Power system stochastic disturbance simulation method and AGC control strategy verification method

Abstract

The invention discloses a power system random disturbance simulation method and an AGC control strategy verification method, and the method comprises the steps: building a load disturbance model based on an Oinstein-Urenbeck random process, and enabling the load disturbance model to be used for simulating the load disturbance of different time scales; and injecting load disturbance into the power system, stabilizing frequency fluctuation, analyzing load disturbance characteristics and system frequency probability distribution characteristics at different time scales, and verifying the effectiveness of the AGC control strategy. According to the method, the OU random process is adopted to construct the load disturbance model so as to accurately simulate the random fluctuation of an actual power grid, the control performance of AGC under multiple time scales is considered, and the AGC control strategy suitable for a high-proportion new energy scene is verified, so that the frequency safety of the power grid is effectively guaranteed.

Description

Technical Field

[0001] This invention relates to the field of power system automatic control technology, and in particular to a method for simulating random disturbances in power systems and a method for verifying AGC control strategies. Background Technology

[0002] Grid frequency is a core indicator of stable power system operation. Ideally, it relies on real-time balance between power generation and load to maintain a constant 50Hz. However, with the increasing proportion of installed capacity of new energy sources such as wind power and photovoltaics, the volatility on both the source and load sides has significantly increased, breaking the traditional "deterministic balance" and causing a substantial increase in the frequency deviation caused by net load disturbances.

[0003] First, the large-scale integration of new energy sources directly amplifies power disturbances. The instantaneous output of wind and solar power is significantly affected by meteorological factors such as wind speed, sunlight, and cloud cover, exhibiting significant randomness and intermittency, and fluctuating across multiple time scales. Simultaneously, user-side behaviors such as the start-up and shutdown of industrial motors and the switching of residential loads also bring continuous, slight fluctuations. The combined effect of these two factors results in a net load curve exhibiting significant random disturbance characteristics. Second, in new power systems, to improve the absorption capacity of new energy sources, the operating ratio of hydropower and thermal power units is forced to decrease, making it difficult for traditional frequency regulation mechanisms relying solely on synchronous machines to meet control requirements. Therefore, there is an urgent need for new energy sources to participate in AGC (Automatic Generation Control) frequency regulation.

[0004] However, current simulations and designs for Automatic Generation Control (AGC) are still generally based on deterministic disturbance models, typically employing generator tripping or load increases / decreases to simulate continuous stochastic processes in actual operation. Existing disturbance models are oversimplified and cannot reproduce the random errors caused by inconsistencies between predicted and actual values, making it difficult to accurately reflect the stochastic fluctuation characteristics of the actual power grid; existing AGC simulation models also cannot effectively analyze control performance. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a method for simulating random disturbances in power systems, which solves the problem of oversimplification in load disturbance modeling in traditional simulations and enables the establishment of random disturbance models that conform to real statistical characteristics. Another purpose of this invention is to provide a method for verifying AGC control strategies, which solves the problem of inaccurate evaluation of control performance in traditional simulations, studies the control performance of AGC systems under different mean regression time constants, analyzes the frequency distribution characteristics of the system, and provides a basis for optimizing AGC controller parameters.

[0006] Technical solution: The present invention provides a method for simulating random disturbances in a power system, comprising the following steps: establishing a load disturbance model based on an Ornstein-Uhlenbeck random process to simulate load disturbances at different time scales;

[0007] The mean value in the load disturbance model is a periodic time variable.

[0008] Furthermore, the load perturbation model is established based on the stochastic differential equation of the Ornstein-Uhlenbeck stochastic process, and the load perturbation model is as follows:

[0009] ;

[0010] in Let be the load disturbance value at time t. For the mean regression rate, For volatility, For Wiener process;

[0011] The mean values ​​in the load disturbance model are as follows:

[0012] ;

[0013] in, The magnitude of the periodic error. Angular frequency, For the first phase, For time.

[0014] Furthermore, by adjusting the mean regression time constant... Simulate load disturbances at different time scales.

[0015] The present invention discloses a power system stochastic disturbance simulation system, which includes a load disturbance modeling unit. The load disturbance modeling unit is used to establish a load disturbance model based on the Ornstein-Uhlenbeck stochastic process to simulate net load stochastic fluctuations.

[0016] The mean value in the load disturbance model is a periodic time variable.

[0017] The AGC control strategy verification method of the present invention includes the following steps:

[0018] The aforementioned power system random disturbance simulation method generates load disturbances at different time scales, and injects these load disturbances into the power system to induce system angular frequency deviation.

[0019] When the system angular frequency deviation exceeds the preset dead zone range, a primary frequency regulation control signal is generated, and the power system performs preliminary suppression of frequency fluctuations based on the primary frequency regulation signal.

[0020] The total adjustment required for automatic generation control is calculated based on the system frequency deviation, and an AGC control strategy is generated. The AGC controller generates an active power adjustment command based on the AGC control strategy, and the new energy generation unit adjusts its grid-connected active power according to the active power adjustment command to smooth out frequency fluctuations.

[0021] The system frequency is collected, and the load disturbance characteristics and system frequency probability distribution characteristics at different time scales are analyzed to verify the effectiveness of the AGC control strategy.

[0022] Furthermore, when the system angular frequency deviation exceeds a preset dead zone range, a primary frequency regulation control signal is generated. The power system performs preliminary suppression of frequency fluctuations based on the primary frequency regulation signal, including:

[0023] The system angular frequency deviation is fed into the dead zone detection stage of the governor model. When the system angular frequency deviation exceeds the preset dead zone range, the governor model generates a primary frequency regulation control signal according to its control logic and sends it to the turbine model to drive the turbine to adjust its mechanical power output. The generator converts the adjusted mechanical power into electrical power and outputs it to the power system. Combined with the load's own regulation, the frequency fluctuation is initially suppressed.

[0024] Furthermore, the total adjustment required for automatic generation control is calculated based on the system frequency deviation, and the AGC control strategy is generated, including:

[0025] The system frequency deviation is measured, and the total adjustment required for automatic power generation control is calculated by combining the predetermined frequency deviation coefficient, thereby generating an AGC control strategy. The control signal of the AGC control strategy is processed by a delay circuit that simulates the communication delay of the master station and then transmitted to the AGC controller.

[0026] The AGC control strategy verification system of the present invention includes:

[0027] The AGC control strategy generation and execution unit is used to generate load disturbances at different time scales using the power system random disturbance simulation method, and inject the load disturbances into the power system to induce system angular frequency deviation.

[0028] When the system angular frequency deviation exceeds the preset dead zone range, a primary frequency regulation control signal is generated, and the power system performs preliminary suppression of frequency fluctuations based on the primary frequency regulation signal.

[0029] The total adjustment required for automatic generation control is calculated based on the system frequency deviation, and an AGC control strategy is generated. The AGC controller generates an active power adjustment command based on the AGC control strategy, and the new energy generation unit adjusts its grid-connected active power according to the active power adjustment command to smooth out frequency fluctuations.

[0030] The AGC control strategy verification unit collects system frequency data, analyzes load disturbance characteristics and system frequency probability distribution characteristics at different time scales, and verifies the effectiveness of the AGC control strategy.

[0031] The computer-readable storage medium of the present invention stores a computer program, which, when executed by a processor, implements the power system random disturbance simulation method.

[0032] The computer program product of the present invention includes a computer program that, when executed by a processor, implements the power system random disturbance simulation method.

[0033] Another computer-readable storage medium of the present invention stores a computer program, which, when executed by a processor, implements the AGC control strategy verification method.

[0034] Another computer program product of the present invention includes a computer program that, when executed by a processor, implements the verification method according to the AGC control strategy.

[0035] Beneficial Effects: Compared with existing technologies, the advantages of this invention are as follows: This invention establishes a random disturbance model based on a periodic Ornstein-Uhlenbeck stochastic process that conforms to real statistical characteristics, capable of simulating load disturbances at different time scales and accurately simulating random fluctuations in the actual power grid; This invention considers the control performance of AGC at multiple time scales, analyzes the coupling relationship between the mean regression time constant, disturbance characteristics, and the control performance of new energy AGC, quantitatively evaluates the control effect through frequency probability distribution characteristics, and verifies the AGC control strategy suitable for high-proportion new energy scenarios, thereby effectively ensuring the frequency security of the power grid. This invention provides a scientific basis for parameter optimization and tuning of AGC controllers in high-proportion new energy power grids, effectively improving the frequency stability of the power system under random disturbances. Attached Figure Description

[0036] Figure 1 This is a block diagram of the power system frequency model in Embodiment 3 of the present invention.

[0037] Figure 2 In Embodiment 3 of the present invention The waveform of the disturbance load simulated for the OU process at 10s, and the frequency probability distribution diagrams when AGC is not engaged (blue waveform) and when AGC is engaged (red waveform).

[0038] Figure 3 In Embodiment 3 of the present invention The perturbation load waveform simulated by the OU process at 100s, and the frequency probability distribution diagrams when AGC is not engaged (blue waveform) and when AGC is engaged (red waveform).

[0039] Figure 4 In Embodiment 3 of the present invention The perturbation load waveform simulated for the OU process at 500s, and the frequency probability distribution diagrams for when AGC is not engaged (blue waveform) and when AGC is engaged (red waveform). Detailed Implementation

[0040] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0041] Example 1

[0042] The power system random disturbance simulation method of the present invention includes the following steps.

[0043] Based on the stochastic differential equations of the Ornstein-Uhlenbeck (OU) stochastic process, a load disturbance model is constructed to simulate stochastic fluctuations in net load. The calculation formula is as follows:

[0044] Stochastic differential equations (SDEs) for OU processes:

[0045] (1)

[0046] in: Let be the load disturbance value at time t; This represents the mean regression rate, reflecting how quickly the load reverts to its long-term mean. This is the long-term mean; Volatility reflects the magnitude of load disturbances; For Wiener processes, simulate randomness.

[0047] By adjusting the mean regression time constant Simulate load disturbances at different time scales:

[0048] Short time constant ( ≤10s): Simulates high-frequency fluctuations, such as the start-up and shutdown of industrial motors, and the fluctuations in new energy sources caused by instantaneous changes in wind speed and sunlight;

[0049] Medium time constant ( ≤200s): Simulates mid-frequency fluctuations, such as load changes in commercial areas;

[0050] Time constant ( ): Simulates low-frequency fluctuations, such as changes in new energy output caused by weather changes.

[0051] Considering the periodic correlation error, then in equation (1) It is a time-varying mean, expressed as:

[0052] (2)

[0053] In the formula The magnitude of the periodic error. Angular frequency, This is the initial phase.

[0054] Equation (2) combined with the OU random model forms a complete error signal that combines periodic drift and random jitter. .

[0055] This invention employs an Ornstein-Uhlenbeck (OU) process with periodic characteristics to simulate load disturbances, which include two types of errors: correlated errors and uncorrelated errors. Uncorrelated errors include variable disturbances (fluctuations around the average estimate) and uncertainty disturbances (deviations from the average estimate). Correlated errors include system estimation errors in the forecast, errors caused by the scheduling cycle and the renewable energy generation cycle, and are periodic.

[0056] Example 2

[0057] The present invention discloses a power system stochastic disturbance simulation system, which includes a load disturbance modeling unit. The load disturbance modeling unit is used to establish a load disturbance model based on the Ornstein-Uhlenbeck stochastic process to simulate net load stochastic fluctuations.

[0058] The mean value in the load disturbance model is a periodic time variable.

[0059] Example 3

[0060] like Figure 1 The diagram shown is a block diagram of the power system frequency model in this embodiment, including a load disturbance model, a governor model, a turbine model, a new energy AGC control system, a generator and load model, and a simulation analysis system. The system involves a frequency given... , , All are per-unit values.

[0061] The load disturbance model mentioned above adopts the load disturbance model established in Example 1.

[0062] The governor model generates a control signal based on the error between the reference speed and the actual rotational speed, and adjusts the valve opening to control the flow of steam or gas into the turbine, thereby regulating the turbine's output power.

[0063] The turbine model is a rotating power device that converts steam or gas into mechanical energy. Together with the governor model, it constitutes the primary frequency regulation system of the generator, providing primary frequency regulation power to the unit.

[0064] The new energy AGC control system generates power adjustment commands based on frequency deviation and sends them to the new energy generator unit to achieve automatic power adjustment.

[0065] The generator and load model is a composite effect of all units and loads, which together determine the magnitude and rate of change of the system frequency deviation.

[0066] The simulation data analysis system calculates the system frequency based on the frequency deviation output by the simulation model, analyzes the system frequency probability distribution, and verifies the effect of the AGC control strategy.

[0067] The AGC control strategy verification method of the present invention includes the following steps.

[0068] Step 1: Based on the stochastic differential equations of the Ornstein-Uhlenbeck (OU) stochastic process, construct a load disturbance model to simulate the stochastic fluctuations of net load.

[0069] Step two: The load disturbance model is introduced into the power system, causing a system angular frequency deviation. The difference between the reference angular frequency value and the actual angular frequency value is sent to the dead zone detection stage of the speed governor model. When the difference exceeds the preset dead zone range, the speed governor model generates a primary frequency regulation control signal according to its control logic and sends it to the turbine model to drive the turbine to adjust its mechanical power output. The generator converts the adjusted mechanical power into electrical power and outputs it to the power system. Combined with the load's own regulation, the frequency fluctuation is initially suppressed.

[0070] Step 3: Measure the system frequency deviation and, in conjunction with a predetermined frequency deviation coefficient, calculate the total adjustment required for automatic generation control. The AGC adjustment command signal is processed by a delay circuit simulating the communication delay of the master station and then transmitted to the AGC controller. The AGC controller generates an active power adjustment command according to the control strategy and issues it to the new energy generation unit. The new energy generation unit calls upon its reserved standby capacity or collaborative energy storage system to adjust its grid-connected active power to further smooth out power disturbances and restore and maintain the grid frequency near its rated value.

[0071] The AGC controller uses PI control.

[0072] (3)

[0073] in, Frequency deviation coefficient, For AGC adjustment commands, and These are adjustable control parameters.

[0074] Step four: Using Monte Carlo simulation, a random scenario is generated under the load disturbance model, and data such as system frequency are collected to analyze the load disturbance characteristics and system frequency probability distribution characteristics under different mean regression time constants, thereby verifying the effectiveness of the AGC control strategy.

[0075] The simulation analysis includes time-domain analysis and statistical domain analysis.

[0076] 1) Time domain analysis.

[0077] Observe the frequency deviation curve over time, and analyze the frequency at which the power disturbance causes the frequency deviation to cross the mean based on different mean regression time constants.

[0078] Among them, the frequency at which the frequency deviation crosses the mean refers to the zero-crossing rate. In zero-crossing theory, the zero-crossing rate refers to the rate of change in a stochastic process. The average number of times the line crosses its mean (zero line) per unit of time, expressed as 1 / time.

[0079] For a stationary random process with zero mean Its expected zero crossover rate (ν0) is:

[0080] (4)

[0081] in: It is the autocorrelation function of the process, defined as ; It is variance. ; Is the autocorrelation function of the OU process in The second derivative at point .

[0082] The autocorrelation function of the OU process is:

[0083] (5)

[0084] in It is variance. It is the mean regression time.

[0085] According to the autocorrelation function of the OU process, R(0) = σ²;

[0086] First derivative:

[0087] (6)

[0088] Second derivative:

[0089] (7)

[0090] so, .

[0091] Substituting into formula (4) Rice's formula, calculate the zero crossover rate:

[0092] (8)

[0093] 2) Statistical domain analysis.

[0094] Frequency probability density distribution diagrams were plotted using frequency data of AGC with and without it, and the suppression effect of AGC on random power disturbances was analyzed at different time scales.

[0095] The following specific experiments verify the AGC control strategy verification method described in this invention. Figures 2 to 4 This demonstrates the different mean regression time constants ( The load disturbance waveform simulated by the periodic OU process and the system frequency probability distribution of the method described in this invention when AGC control is engaged and disengaged are compared to verify the effectiveness of the AGC control strategy under different disturbance characteristics.

[0096] Figure 2 Corresponding mean regression time constant =10s working condition, Figure 2 In the example, (a) represents a mean period of 900 seconds. The disturbance load waveform simulated by the OU process is 10s, with the horizontal axis representing time and the vertical axis representing the disturbance load power. Figure 2 (b) in the example has a mean period of 900 seconds. The graph shows the frequency probability distributions for 10 seconds without AGC (blue waveform) and with AGC (red waveform), with the horizontal axis representing frequency and the vertical axis representing probability density. At this time, due to the rapid mean regression of the OU process, the simulated load disturbance exhibits high-frequency fluctuation characteristics and a high zero-crossing rate. Under these conditions, the AGC control system has no significant impact on the frequency distribution. This result demonstrates that for high-frequency random disturbances, the system's primary frequency modulation characteristics play a dominant role.

[0097] Figure 3 Corresponding mean regression time constant =100s working condition, Figure 3 In the example, (a) represents a mean period of 900 seconds. The disturbance load waveform simulated by the OU process is 100s. Figure 3 (b) in the example has a mean period of 900 seconds. The simulation results show the frequency probability distributions with and without AGC (blue waveform) for 100 seconds. In this case, the mean regression speed of the disturbance waveform slows down, the zero-crossing rate decreases, and the high-frequency harmonic content decreases. The simulation results show that the standard deviation of the system frequency decreases after AGC control is implemented. This result proves that when the disturbance characteristics are within a specific frequency band, AGC control begins to show its effect in improving frequency quality.

[0098] Figure 4 Corresponding mean regression time constant =500s working condition, Figure 4 In the example, (a) represents a mean period of 900 seconds. The disturbance load waveform simulated by the OU process is 500s. Figure 4 (b) has a mean period of 900s. The simulation results show the frequency probability distributions with and without AGC (blue waveform) for a duration of 500 seconds. Under these parameters, the load disturbance simulated by the OU process exhibits significant slow variation and drift characteristics, with an extremely low zero crossover rate. The simulation results clearly show that the standard deviation of the system frequency is significantly reduced after AGC control is implemented. This result demonstrates that for such slow-varying, persistent random disturbances, the AGC control strategy can effectively suppress frequency fluctuations and significantly improve system frequency stability.

[0099] The waveforms show that as the mean regression time constant increases, the zero-crossing rate of the random disturbance simulated by the OU process gradually decreases. Under the condition of a fixed AGC response period, the longer the mean regression time, the more significant the AGC frequency regulation effect. By introducing a periodic OU random process to simulate load disturbances using this invention, the random fluctuation characteristics of the actual power grid can be more accurately reflected, thus providing an effective basis for AGC controller parameter tuning.

[0100] Example 4

[0101] The AGC control strategy verification system of the present invention includes:

[0102] The AGC control strategy generation and execution unit is used to generate load disturbances at different time scales using the power system random disturbance simulation method, and inject the load disturbances into the power system to induce system angular frequency deviation.

[0103] When the system angular frequency deviation exceeds the preset dead zone range, a primary frequency regulation control signal is generated, and the power system performs preliminary suppression of frequency fluctuations based on the primary frequency regulation signal.

[0104] The total adjustment required for automatic generation control is calculated based on the system frequency deviation, and an AGC control strategy is generated. The AGC controller generates an active power adjustment command based on the AGC control strategy, and the new energy generation unit adjusts its grid-connected active power according to the active power adjustment command to smooth out frequency fluctuations.

[0105] The AGC control strategy verification unit collects system frequency data, analyzes load disturbance characteristics and system frequency probability distribution characteristics at different time scales, and verifies the effectiveness of the AGC control strategy.

Claims

1. A method for simulating stochastic disturbances in a power system, characterized in that, The steps include: establishing a load perturbation model based on the Ornstein-Uhlenbeck stochastic process to simulate load perturbations at different time scales; The mean value in the load disturbance model is a periodic time variable.

2. The method for simulating random disturbances in a power system according to claim 1, characterized in that, The load perturbation model is established based on the stochastic differential equation of the Ornstein-Uhlenbeck stochastic process, and the load perturbation model is as follows: ； in Let be the load disturbance value at time t. For the mean regression rate, For volatility, For Wiener process; The mean values ​​in the load disturbance model are as follows: ； in, The magnitude of the periodic error. Angular frequency, For the first phase, For time.

3. The method for simulating random disturbances in a power system according to claim 2, characterized in that, By adjusting the mean regression time constant Simulate load disturbances at different time scales.

4. A power system stochastic disturbance simulation system, characterized in that, It includes a load disturbance modeling unit, which is used to establish a load disturbance model based on the Ornstein-Uhlenbeck stochastic process to simulate random fluctuations in net load; The mean value in the load disturbance model is a periodic time variable.

5. A method for verifying AGC control strategies, characterized in that, Includes the following steps: The power system random disturbance simulation method of claim 1 is used to generate load disturbances at different time scales, and the load disturbances are injected into the power system to induce system angular frequency deviation; When the system angular frequency deviation exceeds the preset dead zone range, a primary frequency regulation control signal is generated, and the power system performs preliminary suppression of frequency fluctuations based on the primary frequency regulation signal. The total adjustment required for automatic generation control is calculated based on the system frequency deviation, and an AGC control strategy is generated. The AGC controller generates an active power adjustment command based on the AGC control strategy, and the new energy generation unit adjusts its grid-connected active power according to the active power adjustment command to smooth out frequency fluctuations. The system frequency is collected, and the load disturbance characteristics and system frequency probability distribution characteristics at different time scales are analyzed to verify the effectiveness of the AGC control strategy.

6. The AGC control strategy verification method according to claim 5, characterized in that, When the system angular frequency deviation exceeds the preset dead zone range, a primary frequency regulation control signal is generated. The power system performs preliminary suppression of frequency fluctuations based on the primary frequency regulation signal, including: The system angular frequency deviation is fed into the dead zone detection stage of the governor model. When the system angular frequency deviation exceeds the preset dead zone range, the governor model generates a primary frequency regulation control signal according to its control logic and sends it to the turbine model to drive the turbine to adjust its mechanical power output. The generator converts the adjusted mechanical power into electrical power and outputs it to the power system. Combined with the load's own regulation, the frequency fluctuation is initially suppressed.

7. The AGC control strategy verification method according to claim 5, characterized in that, The total adjustment required for automatic generation control is calculated based on the system frequency deviation, and the AGC control strategy is generated, including: The frequency deviation of the measurement system is combined with a predetermined frequency deviation coefficient to calculate the total adjustment required for automatic power generation control and generate an AGC control strategy. The control signal of the AGC control strategy is processed by a delay circuit that simulates the communication delay of the master station and then transmitted to the AGC controller.

8. An AGC control strategy verification system, characterized in that, include: The AGC control strategy generation and execution unit is used to generate load disturbances at different time scales using the power system random disturbance simulation method described in claim 1, and inject the load disturbances into the power system to induce system angular frequency deviation. When the system angular frequency deviation exceeds the preset dead zone range, a primary frequency regulation control signal is generated, and the power system performs preliminary suppression of frequency fluctuations based on the primary frequency regulation signal. The total adjustment required for automatic generation control is calculated based on the system frequency deviation, and an AGC control strategy is generated. The AGC controller generates an active power adjustment command based on the AGC control strategy, and the new energy generation unit adjusts its grid-connected active power according to the active power adjustment command to smooth out frequency fluctuations. The AGC control strategy verification unit collects system frequency data, analyzes load disturbance characteristics and system frequency probability distribution characteristics at different time scales, and verifies the effectiveness of the AGC control strategy.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the power system random disturbance simulation method according to any one of claims 1-3.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the power system random disturbance simulation method according to any one of claims 1-3.

11. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the AGC control strategy verification method according to any one of claims 5-7.

12. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the AGC control strategy verification method according to any one of claims 5-7.

Images