AGC (Automatic Gain Control) coordinated optimization control system with adaptive capability

Abstract

The invention relates to the technical field of AGC (automatic gain control), and discloses an AGC coordinated optimization control system with self-adaptive ability, which comprises the following steps: establishing a data sensing module: laying a solid data foundation for the AGC coordinated optimization control system to realize accurate regulation and control, and constructing a comprehensive and standard feature data set adaptive to a dynamic scene; the limitation of a traditional fixed parameter prediction model is broken through, and an accurate load prediction value fitting the real-time operation state of the power grid is output; an AGC instruction correction module is established: an accurate load prediction result is converted into a specific parameter adjustment action of an AGC system, dynamic adaptation of system parameters is realized, and the regulation and control precision is improved; and a comprehensive evaluation module is established, wherein after the AGC adjustment instruction is executed, an adjustment effect evaluation value is calculated for effect evaluation. According to the system, a closed-loop regulation and control system driven by source network load storage multi-source data is constructed, inherent defects of a traditional AGC system are specifically solved, and safe and stable operation of a power system is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of AGC control technology, specifically to an AGC coordinated optimization control system with adaptive capabilities. Background Technology

[0002] The AGC (Automatic Generation Control) coordination and optimization control system is a core component of the power dispatch automation system. Its core objectives are to maintain grid frequency stability and ensure tie-line power is controlled within planned values. By collecting grid operation data in real time, it intelligently coordinates the output of various generating units within the region. Under the premise of meeting grid security constraints and unit operation limitations, it achieves the optimal allocation of power generation resources, ultimately achieving multiple goals of grid safety, economy, and high-quality operation.

[0003] Traditional AGC systems rely on historical data to build predictive models with fixed parameters. The parameter update cycle is long, making it difficult to adapt to dynamic scenarios such as random fluctuations in new energy output and sudden responses to flexible loads on the user side. They lack real-time perception and adaptive correction capabilities, resulting in a significant increase in prediction errors. This leads to AGC commands lagging behind actual power deviations, ultimately causing problems such as grid frequency deviations from rated values ​​and tie-line power oscillations, threatening the safe and stable operation of the power system. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides an adaptive AGC coordinated optimization control system. It can construct a closed-loop control system driven by multi-source data from the power grid, load, and storage, thereby specifically solving the inherent defects of traditional AGC systems. It fundamentally solves the problem of traditional AGC commands lagging behind actual power deviation, avoiding risks such as grid frequency deviation from rated values ​​and tie-line power oscillation caused by inaccurate prediction and command lag, and ensuring the safe and stable operation of the power system in scenarios with a high proportion of new energy access.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution: an AGC coordination and optimization control system with adaptive capabilities, including a data sensing module: the data sensing module is used to fuse multi-source dynamic feature data, and to construct a dynamic feature data set after normalizing and standardizing the multi-source dynamic feature data;

[0008] An adaptive prediction module is established: The adaptive prediction module is used to calculate the base load prediction value based on a dynamic feature data set, and to correct the base load prediction value based on the adaptive weight coefficient and the real-time prediction deviation.

[0009] Establish an AGC command correction module: The AGC command correction module is used to calculate parameter adjustment values ​​based on the corrected basic load forecast values, and trigger adjustment commands to adjust the parameters of the current AGC system based on the calculation results of the parameter adjustment values;

[0010] Establish a comprehensive evaluation module: The comprehensive evaluation module is used to calculate the evaluation value of the adjustment effect after executing the AGC adjustment command to evaluate the effect.

[0011] Preferably, the data sensing module is used to acquire historical basic data, new energy data, real-time meteorological data, user-side data, and energy storage data to construct a dynamic feature data set, the expression of which is: ; in, Represents a dynamic feature data set; This represents the historical baseline data within the dynamic feature dataset; Represents new energy data in a dynamic feature data set; Real-time meteorological data representing a dynamic feature dataset; This represents user-side and energy storage data within a dynamic feature dataset.

[0012] Preferably, the historical basic data includes historical load data, unit rated output, and historical power of tie lines;

[0013] The new energy data includes real-time photovoltaic power output data, real-time wind power output data, ultra-short-term forecast values, and adjustable margin.

[0014] The real-time meteorological data includes light intensity, wind speed, temperature, and precipitation probability;

[0015] The user-side and energy storage data include resilient load response status, energy storage SOC, and energy storage charging and discharging power.

[0016] Preferably, the formula for calculating the basic load forecast value is as follows: ; in, Represents the basic load forecast value; The first in the historical basic data One parameter; The first in representing new energy data One parameter, The first in the real-time meteorological data One parameter; Representing the user-side and energy storage data, the first One parameter; , , , Represents weight.

[0017] Preferably, in the adaptive prediction module, the formula for correcting the base load forecast value is as follows: ; in, This represents the revised base load forecast. Represents the adaptive weighting coefficient; Represents the absolute value of the real-time prediction deviation; The raw value representing the real-time prediction deviation.

[0018] Preferably, the adaptive weighting coefficient The formula expression is: ; in, Represents the sensitivity adjustment coefficient; represent.

[0019] Preferably, the formula for calculating the absolute value of the real-time prediction deviation is: ; in, represent Actual load data at any given time; represent Actual load data at any given time.

[0020] Preferably, in the AGC instruction correction module, the formula for calculating the parameter adjustment value is: ; in, Represents the current AGC system. The parameter adjustment value of each parameter is the optimal parameter of the current coefficient after correcting the basic load forecast value; Represents the current AGC system. The current values ​​of the parameters; Represents the rated load of the power grid; This represents the allowable deviation threshold;

[0021] Represents the adjustment direction coefficient; when These parameters belong to the categories of predictive sensitivity parameters or adjustment resource allocation parameters. It is a positive number; when This belongs to the category of parameters that adjust dead zone. It is a negative number.

[0022] Preferably, in the AGC instruction correction module, the condition for triggering the adjustment instruction is:

[0023] when When this happens, an adjustment command is triggered to adjust the parameters of the current AGC system;

[0024] when At that time, keep .

[0025] Preferably, in the comprehensive evaluation module, the formula for calculating the adjustment effect evaluation value is: ; in, This represents the evaluation value of the adjustment effect; Frequency deviation before adjustment; This represents the frequency deviation after adjustment; This represents the total power deviation before adjustment; This represents the deviation in total power after adjustment; , Represents weight; Represents the minimum value;

[0026] When the adjustment effect evaluation value is greater than the high threshold of the adjustment effect evaluation value, it means that the adjustment effect is good; when the adjustment effect evaluation value is less than the low threshold of the adjustment effect evaluation value, it means that the adjustment effect is poor. In the case of poor adjustment effect, parameter readjustment is triggered immediately.

[0027] Compared with the prior art, the present invention provides an AGC coordination optimization control system with adaptive capabilities, which has the following beneficial effects:

[0028] This invention acquires multi-source dynamic feature data, including historical basic data, new energy data, real-time meteorological data, user-side data, and energy storage data, through a data sensing module to construct a dynamic feature data set. This breaks the limitation of traditional systems that rely solely on historical basic data and integrates real-time data from new energy, meteorology, user-side data, and energy storage to provide comprehensive input for accurate prediction and control.

[0029] By using the adaptive correction mechanism of the basic load forecast value of the adaptive forecast module, the fixed parameter model is abandoned. The weight coefficient is dynamically adjusted by real-time forecast deviation, which effectively offsets the forecast error caused by random fluctuations in new energy output and sudden response of flexible load, and improves the real-time performance and accuracy of load forecast.

[0030] The AGC command correction module calculates and executes AGC system parameter adjustments based on the corrected load forecast values, achieving dynamic parameter adaptation rather than long-term static updates, and ensuring that the control parameters are accurately matched with the real-time operating status of the power grid.

[0031] Finally, the effect is evaluated through a comprehensive evaluation module, which fundamentally solves the problem of traditional AGC commands lagging behind actual power deviation, avoids risks such as grid frequency deviation from rated value and tie line power oscillation caused by inaccurate prediction and command lag, and ensures the safe and stable operation of the power system in scenarios with a high proportion of new energy access. Attached Figure Description

[0032] Figure 1 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0033] The technical solutions of 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. It is worth noting that this application also relates to prior art. Since prior art is well known to those skilled in the art, it will not be described in detail in this application.

[0034] Please see Figure 1 The adaptive AGC coordinated optimization control system includes a data sensing module. This module acquires multi-source dynamic characteristic data from historical baseline data, new energy data, real-time meteorological data, and user-side and energy storage data. After normalizing and standardizing the multi-source dynamic characteristic data, a dynamic characteristic data set is constructed. The expression for the dynamic characteristic data set is: ; in, Represents a dynamic feature data set; This represents the historical baseline data within the dynamic feature dataset; Represents new energy data in a dynamic feature data set; Real-time meteorological data representing a dynamic feature dataset; This represents user-side and energy storage data within a dynamic feature dataset;

[0035] Historical basic data includes historical load data, unit rated output, and historical power of tie lines;

[0036] New energy data includes real-time photovoltaic power output data, real-time wind power output data, ultra-short-term forecasts, and adjustable margins;

[0037] Real-time meteorological data includes light intensity, wind speed, temperature, and precipitation probability;

[0038] User-side and energy storage data include resilient load response status, energy storage SOC, and energy storage charging and discharging power;

[0039] The data perception module lays a solid data foundation for the AGC coordinated optimization control system to achieve precise regulation. It constructs a comprehensive, standardized, and dynamic scenario-adaptive feature dataset, providing high-quality data input for subsequent load forecasting and parameter adjustment, and ensuring the accuracy and reliability of subsequent algorithm calculations.

[0040] Establish an adaptive forecasting module: The adaptive forecasting module is used to calculate the base load forecast based on a dynamic feature data set, and to correct the base load forecast based on the adaptive weighting coefficient and the real-time forecast deviation.

[0041] The formula for calculating the basic load forecast is: ; in, Represents the basic load forecast value; The first in the historical basic data One parameter; The first in representing new energy data One parameter, The first in the real-time meteorological data One parameter; Representing the user-side and energy storage data, the first One parameter; , , , Represents weight;

[0042] Basic load forecast Based on a dynamic feature dataset, weights are assigned to historical basic data, new energy data, meteorological data, user-side data, and energy storage data respectively to quantify the impact of various types of data on load forecasting.

[0043] The formula for correcting the base load forecast is as follows: ; in, This represents the revised base load forecast. Represents the adaptive weighting coefficient; Represents the absolute value of the real-time prediction deviation; The raw value representing the real-time prediction deviation;

[0044] Through adaptive weight coefficients The absolute value of the real-time forecast deviation relative to the base load forecast value The correction enabled the prediction model to perceive the dynamic changes in the power grid in real time, effectively reducing the prediction error caused by new energy fluctuations and load disturbances, and avoiding the problem of AGC command lag due to inaccurate prediction.

[0045] Adaptive weight coefficients The formula expression is: ; in, Represents the sensitivity adjustment coefficient; represent;

[0046] The formula for calculating the absolute value of real-time prediction deviation is: ; in, represent Actual load data at any given time; represent Actual load data at any given time;

[0047] The adaptive prediction module breaks through the limitations of traditional fixed parameter prediction models and outputs accurate load prediction values ​​that fit the real-time operating status of the power grid, providing a core basis for subsequent parameter adjustment and command issuance.

[0048] Establish an AGC command correction module: This module calculates parameter adjustment values ​​based on the corrected baseline load forecast and triggers adjustment commands to adjust the parameters of the current AGC system based on the calculation results. The formula for calculating the parameter adjustment values ​​is as follows: ; in, Represents the current AGC system. The parameter adjustment value of each parameter is the optimal parameter of the current coefficient after correcting the basic load forecast value; Represents the current AGC system. The current values ​​of the parameters; Represents the rated load of the power grid; This represents the allowable deviation threshold;

[0049] Represents the adjustment direction coefficient; when These parameters belong to the categories of predictive sensitivity parameters or adjustment resource allocation parameters. It is a positive number; when This belongs to the category of parameters that adjust dead zone. It is a negative number;

[0050] when When this happens, an adjustment command is triggered to adjust the parameters of the current AGC system;

[0051] when At that time, keep ;

[0052] Based on the revised base load forecast Based on this as the core basis, standardized formulas are used to calculate the adjustment values ​​of each key parameter. The formula calculates the parameter adjustment value by coordinating the current parameter value, deviation characteristics, grid rated load, allowable deviation threshold, and adjustment direction coefficient, ensuring that the adjustment value fits the current operating scenario. Afterwards, the system makes a judgment to trigger the adjustment command, and dynamically adjusts the key parameters of the AGC system to ensure that the parameter adjustment matches the dynamic characteristics of the power grid, so that the AGC system can switch from fixed parameter operation to adaptive parameter operation, laying the foundation for the efficient execution of subsequent adjustment commands;

[0053] The AGC instruction correction module translates accurate load forecast results into specific parameter adjustment actions for the AGC system, enabling dynamic adaptation of system parameters and improving control accuracy.

[0054] Establish a comprehensive evaluation module: This module is used to calculate the evaluation value of the adjustment effect after executing the AGC adjustment command. The formula for calculating the evaluation value of the adjustment effect is as follows: ; in, This represents the evaluation value of the adjustment effect; Frequency deviation before adjustment; This represents the frequency deviation after adjustment; This represents the total power deviation before adjustment; This represents the deviation in total power after adjustment; , Represents weight; Represents the minimum value;

[0055] When the adjustment effect evaluation value is greater than the high threshold of the adjustment effect evaluation value, it means that the adjustment effect is good; when the adjustment effect evaluation value is less than the low threshold of the adjustment effect evaluation value, it means that the adjustment effect is poor. In the case of poor adjustment effect, parameter readjustment is triggered immediately.

[0056] After the AGC adjustment command is executed, the comprehensive evaluation module calculates the evaluation value of the adjustment effect based on two core indicators: frequency deviation and power deviation, using a standardized evaluation formula. Then, the adjustment effect is judged by comparing it with the preset high and low thresholds, so as to understand the working ability of the system.

[0057] 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. An AGC coordinated optimization control system with adaptive capabilities, characterized in that, This includes establishing a data perception module: the data perception module is used to fuse multi-source dynamic feature data, and construct a dynamic feature data set after normalizing and standardizing the multi-source dynamic feature data; An adaptive prediction module is established: The adaptive prediction module is used to calculate the base load prediction value based on a dynamic feature data set, and to correct the base load prediction value based on the adaptive weight coefficient and the real-time prediction deviation. Establish an AGC command correction module: The AGC command correction module is used to calculate parameter adjustment values ​​based on the corrected basic load forecast values, and trigger adjustment commands to adjust the parameters of the current AGC system based on the calculation results of the parameter adjustment values; Establish a comprehensive evaluation module: The comprehensive evaluation module is used to calculate the evaluation value of the adjustment effect after executing the AGC adjustment command to evaluate the effect.

2. The adaptive AGC coordinated optimization control system according to claim 1, characterized in that, The data sensing module is used to acquire historical basic data, new energy data, real-time meteorological data, user-side data, and energy storage data to construct a dynamic feature data set. The expression for the dynamic feature data set is: ； in, Represents a dynamic feature data set; This represents the historical baseline data within the dynamic feature dataset; Represents new energy data in a dynamic feature data set; Real-time meteorological data representing a dynamic feature dataset; This represents user-side and energy storage data within a dynamic feature dataset.

3. The adaptive AGC coordinated optimization control system according to claim 2, characterized in that, The historical basic data includes historical load data, unit rated output, and historical power of tie lines; The new energy data includes real-time photovoltaic power output data, real-time wind power output data, ultra-short-term forecast values, and adjustable margin. The real-time meteorological data includes light intensity, wind speed, temperature, and precipitation probability; The user-side and energy storage data include resilient load response status, energy storage SOC, and energy storage charging and discharging power.

4. The adaptive AGC coordinated optimization control system according to claim 3, characterized in that, The formula for calculating the basic load forecast value is as follows: ； in, Represents the basic load forecast value; The first in the historical basic data One parameter; The first in representing new energy data One parameter, The first in the real-time meteorological data One parameter; Representing the user-side and energy storage data, the first One parameter; , , , Represents weight.

5. The adaptive AGC coordinated optimization control system according to claim 4, characterized in that, In the adaptive forecasting module, the formula for correcting the base load forecast value is as follows: ； in, This represents the revised base load forecast. Represents the adaptive weighting coefficient; Represents the absolute value of the real-time prediction deviation; The raw value representing the real-time prediction deviation.

6. The adaptive AGC coordinated optimization control system according to claim 5, characterized in that, The adaptive weighting coefficient The formula expression is: ； in, This represents the sensitivity adjustment coefficient; represent.

7. The adaptive AGC coordinated optimization control system according to claim 5, characterized in that, The formula for calculating the absolute value of the real-time prediction deviation is: ； in, represent Actual load data at any given time; represent Actual load data at any given time.

8. The adaptive AGC coordinated optimization control system according to claim 5, characterized in that, In the AGC instruction correction module, the formula for calculating the parameter adjustment value is as follows: ； in, Represents the current AGC system. The parameter adjustment value of each parameter is the optimal parameter of the current coefficient after correcting the basic load forecast value; Represents the current AGC system. The current values ​​of the parameters; Represents the rated load of the power grid; This represents the allowable deviation threshold; Represents the adjustment direction coefficient; when These parameters belong to the categories of predictive sensitivity parameters or adjustment resource allocation parameters. It is a positive number; when This belongs to the category of parameters that adjust dead zone. It is a negative number.

9. The adaptive AGC coordinated optimization control system according to claim 8, characterized in that, In the AGC instruction correction module, the condition for triggering the adjustment instruction is: when When this happens, an adjustment command is triggered to adjust the parameters of the current AGC system; when At that time, keep .

10. The adaptive AGC coordinated optimization control system according to claim 9, characterized in that, In the comprehensive evaluation module, the formula for calculating the evaluation value of the adjustment effect is as follows: ； in, This represents the evaluation value of the adjustment effect; Frequency deviation before adjustment; This represents the frequency deviation after adjustment; This represents the total power deviation before adjustment; This represents the deviation in total power after adjustment; , Represents weight; Represents the minimum value; When the adjustment effect evaluation value is greater than the high threshold of the adjustment effect evaluation value, it means that the adjustment effect is good; when the adjustment effect evaluation value is less than the low threshold of the adjustment effect evaluation value, it means that the adjustment effect is poor. In the case of poor adjustment effect, parameter readjustment is triggered immediately.

Images