A primary frequency modulation test method and system for a hydroelectric power station governor

Abstract

The application relates to the field of power system control, and discloses a primary frequency modulation test method and system for a hydroelectric station governor, the method comprising the following steps: step 1, establishing a digital twin model and synchronizing physical hydroelectric unit operation information; step 2, performing a virtual primary frequency modulation test to obtain a virtual power response and a virtual dynamic stress response; step 3, quantifying a primary frequency modulation technical performance index of the physical hydroelectric unit; step 4, quantifying an incremental health cost generated by the virtual primary frequency modulation test; and step 5, updating a health state vector of the digital twin model. According to the application, a digital twin model of a physical hydroelectric unit is established, and a primary frequency modulation test is performed based on the virtual model, so that the test process is transferred from a physical entity to a digital space, actual operation on the physical unit is avoided, and thus operation risks, economic expenditure and physical damage to equipment that may exist in the test process are eliminated.

Description

Technical Field

[0001] This invention relates to the field of power system control, specifically to a primary frequency regulation test method and system for a hydropower station governor. Background Technology

[0002] Hydropower, as a clean and renewable energy source, plays a vital role in modern power systems. Hydropower turbine generators are not only the primary producers of electricity, but also, due to their excellent rapid response capabilities, are key resources for providing grid ancillary services and maintaining system frequency stability. Primary frequency regulation is one of the most basic and important ancillary services of hydropower units. It requires the units to automatically and quickly respond to grid frequency deviations, balancing the system's supply and demand by increasing or decreasing output to ensure the safe and stable operation of the grid. With the increasing proportion of intermittent renewable energy generation such as wind and solar power, grid frequency volatility has significantly increased, placing unprecedented demands on and reliance on the performance of primary frequency regulation by hydropower units.

[0003] To verify and evaluate whether a hydropower unit has the primary frequency regulation capability to meet the grid requirements, the traditional approach is to conduct an on-site primary frequency regulation test. This test involves artificially introducing a simulated frequency disturbance signal into the unit's control system and directly observing and measuring its power response process on the physical unit.

[0004] However, this traditional on-site testing method has inherent and insurmountable drawbacks. Because the testing process directly affects the expensive and structurally complex physical hydroelectric generating units, any unexpected dynamic process can threaten the unit's safety, bringing significant operational risks. Furthermore, conducting such tests often requires unit shutdown or deviating from normal power generation plans, resulting in direct losses in power generation efficiency and requiring substantial investment of manpower and resources, constituting high economic costs. More insidiously, frequency regulation testing itself is a disturbance to the unit; frequent testing operations can accelerate the accumulation of fatigue damage to critical components, generating unquantifiable health costs and thus shortening the unit's effective lifespan.

[0005] To mitigate the risks and costs of physical testing, existing technologies employ computer simulation as an alternative. While simulation theoretically avoids physical losses, traditional simulations typically build mathematical models based on a fixed set of idealized parameters. Such static models cannot reflect the degradation of the mechanical properties of hydropower units due to fatigue, wear, and other factors after long-term frequency regulation. Therefore, when assessing the long-term impact of a frequency regulation strategy, the accuracy of traditional simulation models decreases significantly over time because they cannot simulate equipment aging, failing to provide a reliable basis for long-term operational decisions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a primary frequency regulation test method and system for hydropower station governors. This solves the problems of high risk and high cost associated with traditional physical tests, and the long-term inaccurate evaluation caused by the static nature of traditional simulations and their inability to simulate unit aging, making it difficult to quantify the performance and health losses of frequency regulation strategies.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] The first aspect of this invention provides a primary frequency regulation test method for a hydropower station governor, the method comprising the following steps:

[0009] Step S1: Based on the physical hydropower unit's operating information, establish a digital twin model and synchronize the physical hydropower unit's operating information. In a specific implementation, the physical hydropower unit's operating information includes: real-time operating data of the physical hydropower unit acquired through a sensor system, pre-set structural and material data, and operating safety constraint data. Synchronizing the physical hydropower unit's operating information specifically involves: acquiring the physical hydropower unit's operating information at the input signal... Actual measurement output under action And run an online parameter identification algorithm to use the digital twin model with the same input signal. Predicted output under the action With the actual measurement output With the goal of minimizing the error between the parameters, the model parameters of the digital twin model are continuously calibrated. The optimization objective formula is expressed as:

[0010] ;

[0011] in, For a preset loss function, such as the mean squared error function,

[0012] Step S2: Perform a virtual primary frequency regulation test based on the digital twin model to obtain the virtual power response and virtual dynamic stress response of the physical hydropower unit.

[0013] Step S3: Based on the virtual power response, quantify the primary frequency regulation performance indicators of the physical hydropower unit.

[0014] Step S4: Based on the virtual dynamic stress response, quantify the incremental health cost generated by the virtual primary frequency tuning test. In one specific implementation, this step is completed collaboratively by a dynamic stress analysis model coupled to the digital twin model and a fatigue damage accumulation model. The dynamic stress analysis model extracts the amplitude and number of stress cycles based on the virtual dynamic stress response using methods such as rainflow counting. Subsequently, the fatigue damage accumulation model calculates incremental fatigue damage based on the material's SN curve and the Palmgren-Miner linear cumulative damage criterion. The formula for this calculation process is expressed as follows:

[0015] ;

[0016] In the formula, It is in the The actual number of stress cycles occurring at each stress level It is in the The maximum number of cycles allowed for the material at a given stress level. The incremental health cost. Defined as the incremental fatigue damage The function.

[0017] Step S5: Based on the incremental health cost, update the health state vector of the digital twin model and perform self-evolution on the digital twin model. The health state vector... It is a vector containing multiple variables characterizing the health status of the unit, including at least the cumulative fatigue damage values ​​of the key components of the physical hydropower unit. The incremental health cost leads to an update of the cumulative fatigue damage value. The model parameters of the digital twin model... Defined as the health state vector The function, i.e. In one specific implementation, the model parameters This includes model parameters related to the mechanical properties of the physical hydroelectric generator. When the health state vector... After the update, the model parameters The model is then updated, thus enabling it to evolve on its own.

[0018] Furthermore, the method of the present invention may further include the following steps:

[0019] Define at least one primary frequency regulation strategy to be evaluated, wherein each primary frequency regulation strategy to be evaluated is characterized by a set of governor control parameters applied in the digital twin model.

[0020] In an optional implementation, steps S2 to S5 are repeated for multiple different primary frequency modulation (FM) strategies to be evaluated. Based on the obtained incremental health costs and FM technical performance indicators corresponding to each FM strategy to be evaluated, multi-objective optimization is performed to generate a Pareto optimal frontier. This multi-objective optimization problem can be formulated as finding a set of frequency modulation strategies. , so that the objective function vector The optimal formula is:

[0021] ;

[0022] In the formula, To adopt a strategy The performance indicators of primary frequency modulation technology at that time. To adopt a strategy The incremental health costs over time.

[0023] In another optional embodiment, the method of the present invention may further include the following steps: generating an ordered disturbance sequence representing a preset operating cycle based on historical power grid data, the sequence containing multiple virtual frequency disturbance signals; sequentially executing the virtual frequency disturbance signals in the ordered disturbance sequence according to the primary frequency regulation strategy to be evaluated, and repeating steps S2 to S5 in each execution to simulate the evolution process of the digital twin model within the operating cycle; and evaluating the long-term risk and benefit of the primary frequency regulation strategy to be evaluated based on the final result of the evolution process. The long-term risk and benefit specifically include: the cumulative health cost at the end of the preset long-term operating cycle, and the probability that the performance indicators of the primary frequency regulation technology consistently meet the preset performance requirements within the cycle.

[0024] A second aspect of the present invention provides a system for primary frequency regulation testing of a hydropower station governor, the system comprising:

[0025] The digital twin model module is used to establish a digital twin model based on the physical hydropower unit's operating information and synchronize the physical hydropower unit's operating information.

[0026] A virtual test module, connected to the digital twin model module, is used to perform a virtual primary frequency regulation test based on the digital twin model module to obtain the virtual power response and virtual dynamic stress response of the physical hydropower unit.

[0027] A performance quantification module, connected to the virtual test module, is used to quantify the primary frequency regulation performance indicators of the physical hydropower unit based on the virtual power response.

[0028] A health cost quantification module, connected to the virtual test module, is used to quantify the incremental health cost caused by the virtual primary frequency modulation test based on the virtual dynamic stress response.

[0029] The model evolution module, connected to the health cost quantification module and the digital twin model module, is used to update the health state vector of the digital twin model module according to the incremental health cost, and update the model parameters of the digital twin model module based on the updated health state vector, so as to enable the digital twin model module to self-evolve.

[0030] This invention provides a method and system for primary frequency regulation testing of a hydropower station governor. It offers the following advantages:

[0031] 1. This invention establishes a digital twin model of a physical hydroelectric generator unit and performs a frequency regulation test based on the virtual model. This method transfers the test process from the physical entity to the digital space, fundamentally avoiding actual operation on the physical unit, thus eliminating potential operational risks, economic expenditures, and physical damage to the equipment during the test.

[0032] 2. This invention quantifies the incremental health cost generated by virtual experiments and uses this cost to update the health state vector of the digital twin model, thereby adjusting the model parameters related to this vector function. This achieves a self-optimization mechanism, enabling the model to simulate the health degradation of the unit caused by continuous frequency regulation tasks. This overcomes the shortcomings of traditional fixed-parameter simulations, which cannot reflect the equipment aging process, and significantly improves the accuracy of predicting the long-term operating performance of the unit.

[0033] 3. This invention calculates the technical performance indicators and incremental health costs of multiple different primary frequency regulation strategies to be evaluated, and performs multi-objective optimization to generate the Pareto optimal frontier. It can intuitively show the quantitative constraint relationship between frequency regulation performance and unit health loss, and provide objective and quantitative decision support for operators to weigh and choose between different strategies. Attached Figure Description

[0034] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0035] Figure 2 This is a schematic diagram of the system architecture of the present invention.

[0036] Among them, 10 is the digital twin model module; 20 is the virtual experiment module; 30 is the performance quantification module; 40 is the health cost quantification module; and 50 is the model evolution module. Detailed Implementation

[0037] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] To better understand the present invention, the above content will be described in detail below with reference to specific embodiments.

[0039] Please see the appendix Figure 1 and attached Figure 2 , attached Figure 1 This is a schematic flowchart of a primary frequency regulation test method for a hydropower station governor according to an embodiment of the present invention. Figure 2 This is a schematic diagram of the system architecture of a primary frequency regulation test system for a hydropower station governor, according to another embodiment of the present invention.

[0040] Specifically, in step S1, the digital twin model module 10 performs the operation. The digital twin model module 10 first establishes a digital twin model based on preset physical hydropower unit operating information. This physical hydropower unit operating information includes: real-time operating data of the physical hydropower unit acquired through a sensor system, pre-set structural and material data, and operational safety constraint data.

[0041] The digital twin model module 10 also performs a synchronization operation, which is achieved through an online parameter identification algorithm. Specifically, it acquires the physical hydroelectric unit's input signal... Actual measurement output under action ; and using the digital twin model with the same input signal Predicted output under the action With the actual measurement output With the goal of minimizing the error between the parameters, the model parameters of the digital twin model are continuously calibrated. The optimization objective can be expressed as:

[0042] ;

[0043] In the formula, This is a preset loss function, such as the mean squared error function; For the digital twin model in the input signal Predicted output under the influence of; For the physical hydroelectric generator unit under the input signal Actual measurement output under action; These are a set of adjustable model parameters for the digital twin model.

[0044] During step S2, the virtual test module 20 performs the operation. The virtual test module 20 is connected to the digital twin model module 10 and performs a virtual primary frequency regulation test based on the synchronized digital twin model in the digital twin model module 10. This test is achieved by applying a virtual frequency perturbation signal to the model, thereby obtaining the virtual power response and virtual dynamic stress response of the physical hydropower unit, and outputting these two response data to the performance quantification module 30 and the health cost quantification module 40, respectively.

[0045] During step S3, the performance quantification module 30 performs the operation. The performance quantification module 30 is connected to the virtual test module 20, receives virtual power response data, and quantifies the primary frequency regulation technical performance indicators of the physical hydropower unit according to preset calculation rules.

[0046] In one specific implementation, the primary frequency modulation performance indicators include at least the primary frequency modulation adjustment rate. This rate is calculated according to the following formula:

[0047] ;

[0048] In the formula: This refers to the frequency modulation adjustment rate. This represents the maximum change in unit response power within the virtual power response time series. The time taken from the start of the disturbance until the power reaches its maximum change.

[0049] During step S4, the health cost quantification module 40 performs the operation. The health cost quantification module 40 is connected to the virtual test module 20, receives virtual dynamic stress response data, and quantifies the incremental health cost generated by the virtual primary frequency modulation test using a dynamic stress analysis model and a fatigue damage accumulation model. Specifically, the fatigue damage accumulation model, based on the Palmgren-Miner linear cumulative damage criterion, is used to calculate the incremental fatigue damage. The calculation process can be expressed as follows:

[0050] ;

[0051] In the formula, It is in the The actual number of stress cycles occurring at each stress level It is in the The maximum number of cycles allowed for the material at a given stress level. The incremental health cost. Defined as the incremental fatigue damage The function.

[0052] During step S5, the model evolution module 50 performs the operation. The model evolution module 50 is connected to the health cost quantification module 40 and the digital twin model module 10. The model evolution module 50 receives the incremental health cost calculated by the health cost quantification module 40 and updates the health state vector of the digital twin model based on this cost. The health status vector The data should include at least the cumulative fatigue damage values ​​of the key components of the physical hydropower unit. ;

[0053] Subsequently, the model evolution module 50 is based on the updated health state vector. Model parameters of the digital twin model Adjustments were made to the model parameters. Defined as the health state vector The function, whose relationship can be expressed as:

[0054] ;

[0055] In the formula, It is a preset functional relationship used to convert the health state vector Mapping to model parameters The model parameters The model parameters include those related to the mechanical properties of the physical hydroelectric generator unit. Adjusted model parameters. It is transmitted back to the digital twin model module 10 to update the digital twin model therein, thereby completing the self-evolution of the model.

[0056] Specifically, in step S1, the physical hydropower unit operating information upon which the digital twin model is based is a comprehensive description of the multi-dimensional characteristics of the physical hydropower unit. In one specific implementation, this information is divided into the following three categories:

[0057] The first category is real-time operational data. This type of data is collected and transmitted in real time by a sensor network and monitoring system (SCADA) deployed on the physical hydroelectric generator units. Specifically, it includes, but is not limited to: unit speed, active power, reactive power, generator stator and rotor current and voltage, turbine guide vane opening, head formed by the water levels upstream and downstream of the unit, and vibration and temperature measurements of key components (such as the main shaft and bearings).

[0058] The second category is structural and material data. This type of data consists of static or quasi-static information describing the inherent properties of the physical hydroelectric generator unit. Specifically, it includes: detailed three-dimensional geometric model data and dimensional parameters of key pressure-bearing or load-bearing components of the unit (such as turbine blades, main shaft, and volute); material grades and corresponding mechanical property parameters of these components, such as elastic modulus and Poisson's ratio; and material SN curves used for fatigue life calculation, which define the number of cycles required for material failure under different stress amplitudes.

[0059] The third category is operational safety constraint data. This type of data defines the inviolable safety boundary conditions for physical hydroelectric units during operation. Specifically, it includes: preset vibration alarm thresholds and shutdown thresholds for each measuring point; the maximum allowable operating temperature of components such as bearings and generator coils; the allowable pressure fluctuation range in the turbine and water intake system; and the maximum and minimum active power output limits specified by the power grid or the equipment itself. This data is used to determine whether the system behavior is within a safe range during virtual testing and evaluation.

[0060] The digital twin model is mathematically a formalized expression of the dynamic characteristics of a physical hydroelectric generator. The specific structure of this model can be implemented using different technical solutions depending on the requirements for modeling accuracy and computational efficiency.

[0061] In one implementation, the digital twin model is constructed using a mechanism-driven modeling approach. The model consists of multiple sub-models, each describing a physical component or process of the hydroelectric generator unit. For example, the model may include: a governor model describing the governor control logic, a servo system model describing the guide vane relay response characteristics, a turbine model describing the process of converting water flow energy into mechanical energy, and a generator model describing the process of converting mechanical energy into electrical energy. These sub-models are coupled together through predefined physical equations (such as differential or algebraic equations) to form the model.

[0062] In one specific implementation, some or all of the sub-models can be expressed using transfer functions from linear system theory. The general transfer function model formula can be expressed as:

[0063] ;

[0064] In the formula, The transfer function for the sub-model; The Laplace transform of the input signal to the sub-model; The Laplace transform of the output signal of the sub-model; It is a complex frequency variable; , ,…, and , ,…, These are the coefficients of the transfer function.

[0065] In another implementation, the digital twin model is constructed using a data-driven modeling approach. Instead of relying on precise physical equations, the model fits the system's input-output relationships by learning from historical operational data. For example, a recurrent neural network (RNN) or its variants, such as a long short-term memory (LSTM) network, can be used to construct this model. The state updates and outputs of a general recurrent neural network can be represented as:

[0066] ;

[0067] ;

[0068] In the formula, For discrete time steps; In time step The input vector; In time step The hidden state vector, For the previous time step The hidden state vector; In time step The output vector; , , This is the weight matrix of the network; , This is the bias vector of the network; , The activation function is a preset value, such as the sigmoid function or the tanh function. In this implementation, all weight matrices and bias vectors in the network together constitute the model parameters to be identified and evolved. .

[0069] In one implementation, a hybrid modeling approach can be employed. This approach combines mechanistic-driven models with data-driven models. For example, a mechanistic model is used for components of the system with well-defined physical laws (such as the generator swing equation), while a data-driven model is used for compensation and correction of nonlinear elements that are difficult to model accurately (such as hydraulic disturbances or frictional losses). In this way, the advantages of both methods can be combined.

[0070] To ensure that the digital twin model accurately reflects the current state of the physical hydropower unit, especially when the unit characteristics drift due to changes in operating conditions or slight degradation, the present invention further includes an online synchronization and parameter identification process in step S1. This process runs continuously to minimize the error between the predicted output of the digital twin model and the actual measured output of the physical hydropower unit.

[0071] In one implementation, the digital twin model module 10 performs this online synchronization operation. This step is implemented using an online parameter identification algorithm. Specifically, the digital twin model module 10 acquires the physical hydroelectric unit's input signal... The actual measurement output is denoted as Meanwhile, digital twin models operate with the same input signal. Next, based on the current model parameters This produces a predicted output, denoted as... The online synchronization process aims to minimize the error between the predicted output and the actual measured output, and continuously corrects the model parameters of the digital twin model. The optimization objective can be expressed as:

[0072] ;

[0073] In the formula, This is a preset loss function used to quantify the difference between the predicted output and the actual measured output; for example, it could be the square of the difference between the two. This is the vector of model parameters to be corrected; For input signals; The model predicts the output; This is the actual measured output of the physical unit.

[0074] In one specific implementation, the online parameter identification process is implemented using a recursive least squares algorithm. This algorithm is applicable to linear system models or systems that can be linearized near the operating point, and it updates the model parameters online iteratively. .

[0075] First, the system model to be identified is represented in the following linear regression form:

[0076] ;

[0077] In the formula, For discrete time steps; In time step The actual measurement output of the system; In time step Information vector, It is the transpose operator; The parameter vector to be identified contains unknown or uncertain parameters of the model; This could be system noise or model error.

[0078] The recursive least squares algorithm uses the following set of equations at each time step. Estimates of the parameter vector Update:

[0079] ;

[0080] ;

[0081] ;

[0082] In the formula, In time step parameter vector The estimated value; This is the parameter vector estimate from the previous time step; In time step The gain matrix; In time step The covariance matrix reflects the degree of uncertainty of the parameter estimates; In time step The covariance matrix; Forgetting factor is a constant with a value between (0,1) used to adjust the weight of the influence of historical data on the current parameter estimate; It is the identity matrix; In time step The actual measurement output of the system;

[0083] At each time step The digital twin model module 10 acquires real-time input and output data of the physical unit and constructs an information vector. and measurement output Then perform the above recursive calculation to obtain the updated parameter vector. Subsequently, the digital twin model module 10 uses the updated... Adjusting the parameters within the digital twin model completes a synchronization operation, making the model output approximate the actual output of the physical unit.

[0084] In another implementation, for nonlinear systems or scenarios with significant process noise and measurement noise, extended Kalman filtering or unscented Kalman filtering algorithms can be used to simultaneously estimate the system's state and parameters.

[0085] In step S2, the virtual test module 20 performs a virtual primary frequency regulation test. The purpose of this step is to obtain the performance and stress data of the unit in response to standard frequency disturbances without affecting the operation of the physical unit.

[0086] Specifically, the virtual experiment module 20 first receives the latest model parameters, which have been synchronized online, from the digital twin model module 10. The digital twin model. Subsequently, this module defines a standard virtual frequency perturbation signal. In one implementation, the virtual frequency perturbation signal For a step signal, the mathematical expression is:

[0087] ;

[0088] In the formula, For time; The initial time at which the disturbance is applied; A preset frequency deviation amplitude, such as -0.1Hz, is used to simulate a sudden drop in the power grid frequency.

[0089] The virtual test module 20 will use this virtual frequency disturbance signal As input, the speed controller sub-model is applied to the digital twin model. Then, the module performs dynamic simulation of the entire digital twin model using methods such as numerical integration, with the simulation duration covering a preset time interval, for example, from... arrive ,in The total simulation time is sufficient to observe the system reaching a new steady state.

[0090] During the simulation, the virtual experiment module 20 records time-series data of specific variables within the model, thereby obtaining two sets of key outputs. The first set of outputs is the virtual power response. This is a data sequence showing the change in active power output over time from the generator sub-model in the digital twin model. This sequence fully characterizes the power regulation behavior of the unit during the virtual primary frequency regulation process.

[0091] The second set of outputs is the virtual dynamic stress response. This is the first preset step in the digital twin model. The data sequence of equivalent stress over time for key components or locations. This stress response can be obtained by calculating a stress analysis sub-model coupled within the digital twin model. This sub-model takes dynamic variables during the simulation process (such as guide vane opening rate of change, water pressure pulsation, and rotational speed fluctuation) as input, calculates and outputs the dynamic stress value at the corresponding location.

[0092] After the simulation is complete, the virtual test module 20 will display the virtual power response. The complete data is sent to the performance quantization module 30, and the virtual dynamic stress response is transmitted. The data is sent to the health cost quantification module 40 for use in subsequent steps S3 and S4.

[0093] In step S3, the performance quantization module 30 receives the virtual power response data sequence generated by the virtual test module 20. The performance indicators of primary frequency regulation technology are calculated based on this data. In one specific implementation, these indicators include, but are not limited to, dead zone, regulation rate, and regulation accuracy. For example, the dead zone is calculated based on the minimum frequency deviation that triggers the unit's power response; the regulation rate is calculated based on the slope of a specific segment on the power response curve, i.e. The adjustment accuracy is calculated based on the deviation between the stable value of the virtual power and the theoretical target value after the frequency modulation process is completed.

[0094] In parallel, in step S4, the health cost quantification module 40 receives the virtual dynamic stress response data sequence for preset key components generated by the virtual test module 20. This allows for the quantification of the incremental health costs resulting from the virtual experiment. In one specific implementation, this quantification process is accomplished collaboratively by two coupled models: a dynamic stress analysis model and a fatigue damage accumulation model.

[0095] Specifically, firstly, the dynamic stress analysis model responds to the input virtual dynamic stress response time series. The model employs rainflow counting to decompose the irregular stress-time history into a series of independent stress reversal cycles with constant amplitudes. The output of this step is a set of data pairs. ,in The total number of stress amplitude levels obtained from the decomposition. , For the first The stress amplitude at each level, This represents the number of cycles counted at this stress amplitude level.

[0096] Subsequently, the fatigue damage accumulation model receives the output of the rainflow counting method and calculates the incremental fatigue damage value according to the Palmgren-Miner linear cumulative damage criterion. The formula for calculating this criterion is as follows:

[0097] ;

[0098] In the formula, It is in the The actual number of stress cycles occurring at each stress level It is in the The maximum number of cycles allowed for the material at a given stress level. The incremental health cost. Defined as the incremental fatigue damage The function.

[0099] The maximum number of loops The value is determined by the SN curve (stress-life curve) of the material of this critical component. This SN curve is part of pre-defined structural and material data, used to define the stress-life of a specific material under different stress amplitudes. fatigue life For a given stress amplitude , The corresponding SN curve can be obtained by querying it or by using its fitting equation (such as the Basquin equation). Value. Calculated incremental fatigue damage value. The incremental health costs resulting from this virtual trial The quantitative measure of health cost is then output to the model evolution module 50 for subsequent step S5.

[0100] In step S5, the model evolution module 50 executes a model self-evolution process. The model evolution module 50 is connected to the health cost quantification module 40 and the digital twin model module 10. It first receives from the health cost quantification module 40 the incremental health cost calculated in step S4, generated from a single virtual trial; this cost is quantified as an incremental fatigue damage value. .

[0101] The model evolution module 50 maintains a healthy state vector. This vector is a multi-dimensional vector used to store and characterize the cumulative health status of key components of a physical hydroelectric generator. In one specific implementation, the health status vector... It includes at least the cumulative fatigue damage values ​​of one or more key components. .

[0102] Upon receiving incremental fatigue damage values Subsequently, the model evolution module 50 pairs of health state vectors Update the data. Specifically, add the incremental fatigue damage value to the corresponding cumulative fatigue damage value in the vector. If the current time is the [number]th [time unit]... The update process of cumulative fatigue damage after each virtual test can be represented as follows:

[0103] ;

[0104] In the formula, For the first The cumulative fatigue damage value is updated after each virtual test; For the first Cumulative fatigue damage value after one virtual test; For the first The incremental fatigue damage values ​​generated by this virtual experiment. Through this step, the health status vector... Updated, denoted as .

[0105] Subsequently, the model parameters of the digital twin model Defined as a health state vector This function allows the model parameters to adaptively adjust as the unit's health deteriorates. The functional relationship can be expressed as:

[0106] ;

[0107] In the formula, For the first The updated model parameter vector after each virtual trial; For the first The health status vector updated after each virtual trial; A preset mapping function is used to map the health state vector. Mapping to model parameters .

[0108] In one specific implementation, the model parameters This includes parameters related to the mechanical characteristics of the physical hydroelectric generator unit. For example, parameter vectors. The parameter representing the effective stiffness of a spindle component. Its relationship with cumulative fatigue damage The functional relationship can be defined as:

[0109] ;

[0110] in, This is the initial stiffness value of the component. This is a preset material damage degradation coefficient. Similarly, other mechanical property parameters of the system, such as the damping coefficient and friction coefficient, can also be defined as the health state vector. Functions of the corresponding variables in the equation.

[0111] Based on the above functional relationship Calculate the updated model parameter vector Then, the model evolution module 50 will use this new parameter vector The data is transmitted to the digital twin model module 10. The digital twin model module 10 receives the new model parameter vector. Then, the updated digital twin model replaces the original parameters within the model. This completes one cycle of model self-evolution. The updated digital twin model reflects the cumulative damage caused by this virtual experiment and serves as the basis for subsequent virtual experiments or evaluations.

[0112] In another application embodiment, for long-term risk-reward assessment of a specific primary frequency regulation strategy, the present invention provides a further application embodiment. The first step of this embodiment is to generate an ordered perturbation sequence, which is used in subsequent simulations to represent all frequency perturbation events experienced by the physical hydropower unit within a preset operating cycle (e.g., one year or one quarter).

[0113] Specifically, the generation process begins with acquiring a dataset containing high-resolution time-series records of historical power grid frequencies, denoted as . Based on this data, the historical frequency deviation sequence was first calculated. Its formula is:

[0114] ;

[0115] In the formula, The nominal frequency of the power grid (e.g., 50Hz).

[0116] Subsequently, the historical frequency deviation sequence Processing is performed to identify and extract a series of independent, significant frequency disturbance events. In one specific implementation, this is achieved by setting a frequency deviation threshold. All time series that satisfy Continuous time periods are identified as independent frequency disturbance events.

[0117] For each identified disturbance event, its corresponding frequency deviation time segment is completely extracted, forming an independent virtual frequency disturbance signal. By traversing the entire historical frequency dataset, a system containing... A library of independent virtual frequency perturbation signals.

[0118] Finally, all the independent virtual frequency perturbation signals in the library are arranged according to their natural chronological order of occurrence in the original historical data, thus forming the final ordered perturbation sequence. This sequence can be represented as:

[0119] ;

[0120] In the formula, It is the first A virtual frequency disturbance signal extracted from historical data. This ordered disturbance sequence is used as the standard input for subsequent long-term evolution simulations, and its statistical properties are consistent with the disturbance environment faced by physical units in a real power grid.

[0121] Specifically, after generating the ordered perturbation sequence, the present invention further provides a long-term evolution simulation process to evaluate the performance and impact of the "primary frequency modulation strategy to be evaluated" throughout the entire operating cycle.

[0122] First, the "primary frequency regulation strategy to be evaluated" needs to be defined. This strategy is characterized by a specific set of initial control parameters, such as proportional, integral, and derivative gains, as well as transient slip coefficients. These initial parameters are configured into the governor sub-model within the digital twin model. Simultaneously, the health state vector of the digital twin model... Initialized, in one specific implementation, its initial cumulative fatigue damage value It is set to zero.

[0123] Subsequently, the system initiates a long-term evolution simulation. This process is a sequentially executed loop, following an ordered perturbation sequence. In sequence, each virtual frequency perturbation signal is applied to the digital twin model.

[0124] For the first in the sequence (among them, From 1 to Virtual frequency disturbance signal The system executes steps S2 to S5 of the present invention completely once.

[0125] Specifically, in the first In the next loop: Virtual test module 20 will display the disturbance signal Apply to model parameters that have been updated after the previous loop. The digital twin model is used to execute a virtual experiment (step S2). The performance quantification module 30 and the health cost quantification module 40 calculate the technical performance indicators and incremental health costs corresponding to the disturbance, respectively. (Steps S3 and S4).

[0126] Next, the model evolution module 50 performs an evolution operation (step S5). It calculates the incremental health cost. Accumulated into the health state vector In this process, the updated health state vector is obtained. Subsequently, based on the functional relationship... Calculate the new model parameter vector And use it to update the digital twin model for the next cycle (the 1st cycle). (Prepare for the next time).

[0127] Throughout the long-term evolution simulation, the system records the technical performance indicators generated in each cycle (i.e., each perturbation event) and the updated health state vector. .

[0128] When the last perturbation signal in the ordered perturbation sequence After processing, the long-term evolutionary simulation process ends. The final result of this evolutionary process includes: a complete sequence of health state evolutionary trajectories. The output data, along with the corresponding sequence of technical performance indicators, is used for subsequent long-term risk-return assessments. In one specific implementation, the long-term risk-return assessment includes: assessing the cumulative health costs of the physical hydropower units at the end of the preset long-term operating period; and assessing the probability that the primary frequency regulation technical performance indicators will consistently meet performance requirements throughout the preset long-term operating period.

[0129] In another application embodiment, the objective is to identify a set of primary frequency modulation strategies to be evaluated that strike an optimal balance between performance and cost. This problem is formalized as a multi-objective optimization problem.

[0130] The decision variables for this optimization problem are vectors. This vector contains all the adjustable control parameters in the primary frequency modulation strategy to be evaluated.

[0131] In one implementation, the multi-objective optimization problem is based on the results of a single virtual trial, and its objective function is expressed as:

[0132] ;

[0133] In the formula, To adopt a strategy The primary frequency modulation performance indicators calculated in step S3 are as follows: To adopt a strategy The incremental health cost is calculated in step S4, where T is the transpose operator;

[0134] In another, more comprehensive implementation, this multi-objective optimization problem is based on the results of long-term evolutionary simulations, and its objective function is expressed as:

[0135] ;

[0136] In the formula, and respectively adopt strategies The long-term returns and long-term risks of time, where T is the transpose operator;

[0137] To solve this multi-objective optimization problem, a multi-strategy optimization module 60 can be set up. In one specific implementation, this module uses a non-dominated sorting genetic algorithm. The algorithm uses NSGA-II to solve the problem. This algorithm performs iterative selection, crossover, and mutation operations on a population composed of different policies, and finally obtains a set of policies called the Pareto optimal solution set.

[0138] The objective function value corresponding to the Pareto optimal solution is then plotted in a two-dimensional coordinate system to obtain the Pareto front, which serves as an auxiliary decision-making tool. Further, a decision support module 70 is set up. This module receives the maximum acceptable risk threshold and the minimum performance requirement threshold from external inputs, selects a subset of strategies that meet the constraints from the Pareto front, and chooses a final strategy for actual deployment based on preset secondary criteria (such as risk minimization).

[0139] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A primary frequency regulation test method for a hydropower station governor, characterized in that, Includes the following steps: Step 1: Based on the physical hydropower unit operation information, establish a digital twin model and synchronize the physical hydropower unit operation information; Step 2: Perform a virtual primary frequency regulation test based on the digital twin model to obtain the virtual power response and virtual dynamic stress response of the physical hydropower unit; Step 3: Based on the virtual power response, quantify the primary frequency regulation performance indicators of the physical hydropower unit; Step 4: Based on the virtual dynamic stress response, quantify the incremental health cost generated by the virtual primary frequency regulation test. The incremental health cost is a quantitative measure of the incremental fatigue damage value generated on the key components of the physical hydropower unit, calculated by the fatigue damage accumulation model based on the virtual dynamic stress response. Step 5: Based on the incremental health cost, update the health state vector of the digital twin model, wherein the model parameters of the digital twin model are functions of the health state vector, used to enable the digital twin model to self-evolve. The health status vector includes the cumulative fatigue damage value of the key components of the physical hydroelectric generator.

2. The primary frequency regulation test method for a hydropower station speed governor according to claim 1, characterized in that, In step 1, the synchronized physical hydropower unit operation information specifically includes: The system acquires real-time operating data of the physical hydropower unit; based on an online parameter identification algorithm, it continuously calibrates the model parameters of the digital twin model to minimize the error between the predicted output of the digital twin model and the actual measured output of the physical hydropower unit.

3. The primary frequency regulation test method for a hydropower station governor according to claim 2, characterized in that, The physical hydroelectric unit operation information includes: real-time operation data, structural and material data, and operation safety constraint data.

4. The primary frequency regulation test method for a hydropower station governor according to claim 1, characterized in that, The method further includes: generating an ordered disturbance sequence based on historical power grid data, wherein the ordered disturbance sequence includes a virtual frequency disturbance signal; According to the frequency modulation strategy to be evaluated, the virtual frequency perturbation signal in the ordered perturbation sequence is executed sequentially to simulate the evolution process of the digital twin model during the running cycle. The simulation process includes: repeating steps 2 to 5. Based on the final result of the evolution process, the long-term risk and return of the primary frequency modulation strategy to be evaluated are assessed.

5. The primary frequency regulation test method for a hydropower station governor according to claim 4, characterized in that, The long-term risk and benefit include: the cumulative health cost at the end of the preset long-term operation period, and the probability that the performance indicators of the primary frequency modulation technology will always meet the performance requirements during the preset long-term operation period.

6. The primary frequency regulation test method for a hydropower station governor according to claim 1, characterized in that, The method further includes: defining at least one primary frequency regulation strategy to be evaluated, each of the primary frequency regulation strategies to be evaluated being characterized by a set of governor control parameters applied in the digital twin model.

7. The primary frequency regulation test method for a hydropower station governor according to claim 6, characterized in that, For multiple different primary frequency modulation strategies to be evaluated, steps 2 to 5 are repeated respectively; and based on the incremental health cost and primary frequency modulation technical performance index corresponding to each primary frequency modulation strategy to be evaluated, multi-objective optimization is performed to generate the Pareto optimal frontier.

8. The primary frequency regulation test method for a hydropower station governor according to claim 1, characterized in that, The quantification of incremental health cost in step 4 is accomplished collaboratively by the dynamic stress analysis model and the fatigue damage accumulation model coupled in the digital twin model.

9. The primary frequency regulation test method for a hydropower station governor according to claim 1, characterized in that, The model parameters in step 5 include the model parameters of the mechanical properties of the physical hydroelectric generator.

10. A system for primary frequency regulation testing of a hydropower station governor, based on the primary frequency regulation testing method for a hydropower station governor according to any one of claims 1-9, characterized in that, include: The digital twin model module is used to establish a digital twin model based on the physical hydropower unit's operating information and synchronize the physical hydropower unit's operating information. The virtual test module performs a virtual primary frequency regulation test based on the digital twin model module to obtain the virtual power response and virtual dynamic stress response of the physical hydropower unit. The performance quantification module quantifies the primary frequency regulation performance indicators of the physical hydropower unit based on the virtual power response. The health cost quantification module quantifies the incremental health cost caused by the virtual primary frequency modulation test based on the virtual dynamic stress response. The model evolution module is used to update the health state vector of the digital twin model module according to the incremental health cost. The model parameters of the digital twin model module are functions of the health state vector, and the digital twin model module is used to self-evolve.

Images