A method for intelligent primary frequency regulation control of thermal power units

Abstract

This invention relates to the field of automatic generation control technology for power systems, and discloses an intelligent control method for primary frequency regulation of thermal power units. This invention acquires real-time power grid operating data, including grid frequency signals, actual unit power generation, and main steam pressure. Based on an XGBoost regression model, it obtains the predicted grid frequency, calculates the predicted frequency deviation, and calculates a feedforward compensation command based on the deviation. It then dynamically adjusts the proportional and integral coefficients of the PID controller based on the grid frequency signal and the actual unit power generation, resulting in an adjusted PID controller. The actual measured grid frequency deviation is input to the adjusted PID controller to obtain a feedback correction command. The feedforward compensation command and the feedback correction command are weighted and summed to obtain a comprehensive primary frequency regulation command. This dynamic adjustment of the PID controller parameters reduces the possibility of parameter overshoot or even oscillation during frequency regulation, thus improving the stability of thermal power unit operation.

Description

Technical Field

[0001] This invention relates to the field of automatic power generation control technology, specifically to an intelligent control method for primary frequency regulation of thermal power units. Background Technology

[0002] Primary frequency regulation is the first line of defense in maintaining grid frequency stability. When a momentary imbalance occurs between grid load and power generation, causing frequency deviation, each unit in the grid needs to respond quickly and adjust its output according to its own frequency regulation characteristics in order to restore frequency stability. Thermal power units are the main power sources participating in primary frequency regulation.

[0003] Traditional thermal power units often employ fixed-parameter PID controllers for primary frequency control, directly calculating control commands based on real-time frequency deviations. This method is prone to parameter overshoot and oscillations, and fixed-parameter PID controllers struggle to adapt to the dynamic characteristics of the unit under different operating conditions. Setting excessively large proportional and integral coefficients in pursuit of rapid response can easily lead to overshoot or even oscillations during the control process, which is detrimental to frequency stability and unit safety. Furthermore, frequent and large-amplitude control commands exacerbate valve wear, increase unit coal consumption, and reduce operational economy. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides an intelligent control method for primary frequency regulation of thermal power units. This method has the advantages of dynamically adjusting PID control parameters, thereby reducing the possibility of parameter overshoot and oscillation, ensuring that the thermal power unit is in a stable frequency state, and improving the stability of the thermal power unit's operation. This solves the aforementioned technical problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: an intelligent control method for primary frequency regulation of thermal power units, comprising the following steps:

[0006] S1: Real-time acquisition of power grid operating data, including power grid frequency signal, actual unit power generation, and unit main steam pressure;

[0007] S2: Input the power grid operating data collected in S1 into the XGBoost regression model to calculate the power grid predicted frequency deviation. ;

[0008] S3: Calculate the feedforward compensation command based on the predicted frequency deviation obtained from S2. ;

[0009] S4: Based on the grid frequency signal and actual generator power collected by S1, dynamically adjust the proportional coefficient of the PID controller. and integral coefficient ;

[0010] S5: The actual measured grid frequency deviation The input is fed into the PID controller adjusted by S4 to generate a feedback correction command. ;

[0011] S6: Transfer the feedforward compensation command obtained from S3 Feedback correction instructions received from S5 Weighted synthesis is performed to obtain a primary frequency modulation synthesis command. .

[0012] As a preferred embodiment of the present invention, the expression for the S1 power grid operating data is as follows: in, Represent the unit's main steam pressure at the initial moment, ..., time respectively. The unit's main steam pressure, ..., time The main steam pressure of the unit; Represent the unit's actual generating power at the initial moment, ..., and at the time of the event, respectively. Actual generating power of the unit, ..., time The actual generating power of the unit; Represent the initial grid frequency, ..., time, respectively. The power grid frequency, ..., time The power grid frequency.

[0013] As a preferred embodiment of the present invention, S2 calculates the predicted frequency deviation of the power grid. Includes the following steps:

[0014] Step A1: Input the historical frequency data sequence of the power grid into the XGBoost regression model to obtain the predicted frequency of the power grid. ;

[0015] Step A2: Calculate the predicted frequency deviation of the power grid Its expression is as follows: in, Indicates the predicted frequency of the power grid; This indicates the rated frequency of the power grid.

[0016] As a preferred embodiment of the present invention, the expression for the historical frequency data sequence of the power grid is: ,in, Indicates the current moment. Indicates the length of the sliding window; This indicates the current grid frequency.

[0017] As a preferred technical solution of the present invention, the S3 feedforward compensation command The expression is as follows: in, Indicates a feedforward compensation command; Indicates the deviation of the predicted frequency of the power grid; This indicates the droop coefficient of the generator unit; This indicates the rated power of the unit.

[0018] As a preferred embodiment of the present invention, the S4 dynamically adjusts the proportional coefficient of the PID controller. and integral coefficient This includes the following steps:

[0019] Step B1: Calculate the rate of change of the main steam pressure of the unit based on the main steam pressure of the unit. ;

[0020] Step B2: Based on the actual power output and the rate of change of the main steam pressure of the unit, obtain the adjustment amount of the PID parameter baseline value through the rule base, including the proportional coefficient adjustment amount. and integral coefficient adjustment amount ;

[0021] Step B3: Calculate the proportionality coefficient and integral coefficient The relevant expressions are as follows: in, This represents the baseline value of the proportionality coefficient; Indicates the baseline value of the integral coefficient; Indicates the adjustment amount of the proportional coefficient; This indicates the adjustment amount of the integral coefficient.

[0022] As a preferred technical solution of the present invention, the main steam pressure change rate of the unit in step B1 The expression is as follows: in, This indicates the rate of change of the unit's main steam pressure; Indicates time The main steam pressure of the unit; Indicates time The former The unit's main steam pressure over a given time period; Indicates a time period.

[0023] As a preferred embodiment of the present invention, the rule base in step B2 includes multiple corresponding rules, each rule being used to determine the proportional coefficient adjustment amount based on the actual power output of the unit and the level of the main steam pressure change rate of the unit. and integral coefficient adjustment amount .

[0024] As a preferred technical solution of the present invention, the S5 feedback correction command The expression is as follows: in, This indicates a feedback correction instruction; Indicates the proportionality coefficient; Indicates the integral coefficient; Indicates time The actual measured deviation of the power grid frequency; The integral term representing the power grid frequency deviation;

[0025] The time Actual measured grid frequency deviation ,in, Indicates time The power grid frequency; This indicates the rated frequency of the power grid.

[0026] As a preferred technical solution of the present invention, the S6 primary frequency modulation integrated command The expression is as follows: in, This indicates a single frequency modulation integrated command; Indicates a feedforward compensation command; This indicates a feedback correction instruction; This represents the weighting coefficient.

[0027] Compared with the prior art, the present invention provides an intelligent control method for primary frequency regulation of thermal power units, which has the following beneficial effects:

[0028] This invention collects real-time power grid operating data, including power grid frequency signals, actual generating power of the generating units, and main steam pressure of the generating units. Based on the XGBoost regression model, it obtains the predicted power grid frequency, calculates the predicted power grid frequency deviation, and calculates a feedforward compensation command based on the predicted power grid frequency deviation. Based on the power grid frequency signal and the actual generating power of the generating units, it dynamically adjusts the proportional and integral coefficients of the PID controller to obtain the adjusted PID controller. The actual measured power grid frequency deviation is input to the adjusted PID controller to obtain a feedback correction command. The feedforward compensation command and the feedback correction command are weighted and summed to obtain a primary frequency regulation comprehensive command. By dynamically adjusting the PID controller parameters, the stability of thermal power unit operation is improved. Attached Figure Description

[0029] Figure 1This is a schematic diagram of the process of the present invention. Detailed Implementation

[0030] 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.

[0031] Please see Figure 1 A method for intelligent primary frequency regulation control of thermal power units includes the following steps:

[0032] S1: Real-time acquisition of power grid operating data, including power grid frequency signal, actual unit power generation, and main steam pressure of the unit. The expression for the power grid operating data is as follows: in, Represent the unit's main steam pressure at the initial moment, ..., time respectively. The unit's main steam pressure, ..., time The main steam pressure of the unit; Represent the unit's actual generating power at the initial moment, ..., and at the time of the event, respectively. Actual generating power of the unit, ..., time The actual generating power of the unit; Represent the initial grid frequency, ..., time, respectively. The power grid frequency, ..., time The power grid frequency;

[0033] S2: Input the power grid operating data collected in S1 into the XGBoost regression model to calculate the power grid predicted frequency deviation. This includes the following steps:

[0034] Step A1: Input the historical frequency data sequence of the power grid into the XGBoost regression model to obtain the predicted frequency of the power grid. ;

[0035] Step A2: Calculate the predicted frequency deviation of the power grid Its expression is as follows: in, Indicates the predicted frequency of the power grid; Indicates the rated frequency of the power grid;

[0036] The expression for the historical frequency data sequence of the power grid is: ,in, Indicates the current moment. Indicates the length of the sliding window; This indicates the current power grid frequency.

[0037] The XGBoost regression model is a model with a very powerful and efficient regression algorithm, which is particularly suitable for processing large-scale datasets. It is designed to combine a series of weak prediction models into a strong prediction model for numerical prediction and has been applied in many prediction fields.

[0038] S3: Calculate the feedforward compensation command based on the predicted frequency deviation obtained from S2. Its expression is as follows: in, Indicates a feedforward compensation command; Indicates the deviation of the predicted frequency of the power grid; This represents the droop coefficient of the generator unit, ranging from 4% to 6%, and is set to 4% in this embodiment. Indicates the rated power of the unit;

[0039] S4: Based on the grid frequency signal and actual generator power collected by S1, dynamically adjust the proportional coefficient of the PID controller. and integral coefficient This includes the following steps:

[0040] Step B1: Calculate the rate of change of the main steam pressure of the unit based on the main steam pressure of the unit. Its expression is as follows: in, This indicates the rate of change of the unit's main steam pressure; Indicates time The main steam pressure of the unit; Indicates time The former The unit's main steam pressure over a given time period; Indicates a time period;

[0041] Step B2: Based on the actual power output and the rate of change of the main steam pressure of the unit, obtain the adjustment amount of the PID parameter baseline value through the rule base, including the proportional coefficient adjustment amount. and integral coefficient adjustment amount ;

[0042] Step B3: Calculate the proportionality coefficient and integral coefficient The relevant expressions are as follows: in, This represents the baseline value of the proportionality coefficient; Indicates the baseline value of the integral coefficient; Indicates the adjustment amount of the proportional coefficient; This indicates the adjustment amount for the integral coefficient;

[0043] The rule base includes multiple corresponding rules, each of which determines the proportional coefficient adjustment amount based on the unit's actual power output and the level of the unit's main steam pressure change rate. and integral coefficient adjustment amount Set the value range for each level of the unit's actual power output and the rate of change of the unit's main steam pressure. Determine the corresponding level for each of the unit's actual power output and the rate of change of the unit's main steam pressure based on actual conditions. The levels of the unit's actual power output include low, medium, and high, and the levels of the unit's main steam pressure rate of change include negative large, negative small, zero, positive small, and positive large. Adjust the proportional coefficient. and integral coefficient adjustment amount The levels include negative large, negative small, zero, positive small, and positive large. Each level of adjustment corresponds to a set adjustment amount. Some rules in the rule base are shown in Table 1:

[0044] Table 1. Rules corresponding to the rule base section:

[0045] The actual operating condition corresponding to Rule 1 is: Under high grid load, the main steam pressure rises rapidly. To avoid excessive heat storage leading to subsequent overshoot, the proportional action should be slightly reduced. And significantly weaken the integral effect. To ensure smoother control; the actual operating condition corresponding to Rule 2 is: under high grid load, the main steam pressure drops rapidly. To prevent the steam in the boiler from being drawn away too quickly, or even to cause dangerous conditions such as boiler fire extinguishing, the proportional action should be slightly strengthened. With rapid response and significantly enhanced integration effect. To quickly establish a balance command and stabilize the pressure; the actual operating condition corresponding to Rule 3 is: the main steam pressure remains stable under the load of the power grid. At this time, the unit is in an ideal operating condition of dynamic balance, the boiler has sufficient heat storage and no drastic changes, so the proportional action is maintained. The response speed will remain unchanged, while the integral action will be slightly enhanced. The aim is to improve the adjustment accuracy for small, persistent frequency deviations, more thoroughly eliminate static errors, and ensure long-term frequency stability at the rated value. Rule 4 corresponds to the following actual operating condition: under low grid load, the main steam pressure rises rapidly. At this time, the boiler has relatively abundant heat storage, allowing for more active use of heat storage for regulation, thus significantly increasing the proportion... And integral function ;

[0046] S5: The actual measured grid frequency deviation The input is fed into the PID controller adjusted by S4 to generate a feedback correction command. Its expression is as follows: in, This indicates a feedback correction instruction; Indicates the proportionality coefficient; Indicates the integral coefficient; Indicates time The actual measured deviation of the power grid frequency; The integral term representing the power grid frequency deviation;

[0047] time Actual measured grid frequency deviation ,in, Indicates time The power grid frequency; Indicates the rated frequency of the power grid;

[0048] S6: Transfer the feedforward compensation command obtained from S3 Feedback correction instructions received from S5 Weighted synthesis is performed to obtain a primary frequency modulation synthesis command. Its expression is as follows: in, This represents the comprehensive frequency regulation command, i.e., the adjustment amount of the unit's actual generated power; Indicates a feedforward compensation command; This indicates a feedback correction instruction; This represents the weighting coefficient.

[0049] 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 method for intelligent primary frequency regulation control of thermal power units, characterized in that: Includes the following steps: S1: Real-time acquisition of power grid operating data, including power grid frequency signal, actual unit power generation, and unit main steam pressure; S2: Input the power grid operating data collected in S1 into the XGBoost regression model to calculate the power grid predicted frequency deviation. ; S3: Calculate the feedforward compensation command based on the predicted frequency deviation obtained from S2. ; S4: Based on the grid frequency signal and actual generator power collected by S1, dynamically adjust the proportional coefficient of the PID controller. and integral coefficient ; S5: The actual measured grid frequency deviation The input is fed into the PID controller adjusted by S4 to generate a feedback correction command. ; S6: Transfer the feedforward compensation command obtained from S3 Feedback correction instructions received from S5 Weighted synthesis is performed to obtain a primary frequency modulation synthesis command. .

2. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 1, characterized in that: The expression for the S1 power grid operating data is as follows: in, Represent the unit's main steam pressure at the initial moment, ..., time respectively. The unit's main steam pressure, ..., time The main steam pressure of the unit; Represent the unit's actual generating power at the initial moment, ..., and at the time of the event, respectively. Actual generating power of the unit, ..., time The actual generating power of the unit; Represent the initial grid frequency, ..., time, respectively. The power grid frequency, ..., time The power grid frequency.

3. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 1, characterized in that: The S2 calculation yields the predicted frequency deviation of the power grid. Includes the following steps: Step A1: Input the historical frequency data sequence of the power grid into the XGBoost regression model to obtain the predicted frequency of the power grid. ; Step A2: Calculate the predicted frequency deviation of the power grid Its expression is as follows: in, Indicates the predicted frequency of the power grid; This indicates the rated frequency of the power grid.

4. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 3, characterized in that: The expression for the historical frequency data sequence of the power grid is: ,in, Indicates the current moment. Indicates the length of the sliding window; This indicates the current grid frequency.

5. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 1, characterized in that: The S3 feedforward compensation command The expression is as follows: in, Indicates a feedforward compensation command; Indicates the deviation of the predicted frequency of the power grid; This indicates the droop coefficient of the generator unit; This indicates the rated power of the unit.

6. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 1, characterized in that: The S4 dynamically adjusts the proportional coefficient of the PID controller. and integral coefficient This includes the following steps: Step B1: Calculate the rate of change of the main steam pressure of the unit based on the main steam pressure of the unit. ; Step B2: Based on the actual power output and the rate of change of the main steam pressure of the unit, obtain the adjustment amount of the PID parameter baseline value through the rule base, including the proportional coefficient adjustment amount. and integral coefficient adjustment amount ; Step B3: Calculate the proportionality coefficient and integral coefficient The relevant expressions are as follows: in, This represents the baseline value of the proportionality coefficient; Indicates the baseline value of the integral coefficient; Indicates the adjustment amount of the proportional coefficient; This indicates the adjustment amount of the integral coefficient.

7. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 6, characterized in that: The main steam pressure change rate of the unit in step B1 The expression is as follows: in, This indicates the rate of change of the unit's main steam pressure; Indicates time The main steam pressure of the unit; Indicates time The former The unit's main steam pressure over a given time period; Indicates a time period.

8. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 6, characterized in that: The rule base in step B2 includes multiple corresponding rules. Each rule is used to determine the proportional coefficient adjustment amount based on the actual power output of the unit and the level of the main steam pressure change rate of the unit. and integral coefficient adjustment amount .

9. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 1, characterized in that: The S5 feedback correction command The expression is as follows: in, This indicates a feedback correction instruction; Indicates the proportionality coefficient; Indicates the integral coefficient; Indicates time The actual measured deviation of the power grid frequency; The integral term representing the power grid frequency deviation; The time Actual measured grid frequency deviation ,in, Indicates time The power grid frequency; This indicates the rated frequency of the power grid.

10. The intelligent control method for primary frequency regulation of a thermal power unit according to claim 1, characterized in that: The S6 primary frequency modulation integrated command The expression is as follows: in, This indicates a single frequency modulation integrated command; Indicates a feedforward compensation command; This indicates a feedback correction instruction; This represents the weighting coefficient.

Images