Primary frequency modulation optimization control method and system based on adaptive gain adjustment

Abstract

The invention discloses a primary frequency modulation optimization control method and system based on adaptive gain adjustment, and the method comprises the steps: obtaining and preprocessing the operation state parameters of a power grid and the historical data of the frequency of the power grid, and constructing a power grid frequency stability classification model; power grid operation parameters are obtained in real time and normalized, and a power grid operation state classification result is obtained according to normalized data; preliminarily judging the stability based on the power grid operation parameters and the classification model, and if not, calculating a power grid frequency stability index and judging whether the power grid frequency stability index is lower than a threshold value; and a self-adaptive gain adjustment model is constructed, if the power grid operation state classification result does not meet the requirement or is lower than a threshold value, the optimal gain parameter of primary frequency modulation control is obtained based on the power grid frequency stability index and the self-adaptive gain adjustment model, and optimization control is realized according to the optimal gain parameter. Accurate adaptive control of the power grid frequency is realized, and a powerful guarantee is provided for safe and stable operation of a power system.

Description

Technical Field

[0001] The present invention belongs to the field of information technology, and in particular relates to a primary frequency modulation optimization control method and system based on adaptive gain adjustment. Background Art

[0002] Primary frequency control in power systems is a key technology for maintaining grid frequency stability. However, in actual operation, the operating state of the grid changes dynamically, and the factors affecting grid stability are complex. Traditional primary frequency control uses fixed gain parameters, which are difficult to adapt to real-time changes in grid operating conditions, resulting in poor control effects and may even cause drastic fluctuations in grid frequency. How to dynamically adjust the gain parameters of primary frequency control according to the real-time operating state of the grid to achieve precise control of grid frequency is a technical problem that needs to be solved urgently.

[0003] To address this problem, it is necessary to deeply analyze the inherent mechanism of power grid operation and explore the key variable indicators that affect power grid stability. Through a large amount of data collection and analysis, a power grid stability assessment model is established to calculate the stability index of the power grid in real time. On this basis, an adaptive gain adjustment model is constructed, with the power grid stability index as the optimization target and the key variable indicators as input, and the gain parameters are optimized through intelligent algorithms. The optimized gain parameters are fed back to the primary frequency regulation control system in a timely manner to accurately adjust the generator set. During the operation of the power grid, it is also necessary to continuously collect signals and dynamically update the adaptive gain adjustment model to adapt to changes in the operating status of the power grid. The solution to this technical problem requires a deep theoretical foundation of power systems, rich data analysis experience, and advanced intelligent optimization algorithms. Summary of the invention

[0004] In order to solve the above technical problems, the present invention proposes a primary frequency modulation optimization control method and system based on adaptive gain adjustment to solve the above problems existing in the prior art.

[0005] To achieve the above object, the present invention provides a primary frequency modulation optimization control method based on adaptive gain adjustment, comprising:

[0006] Obtain and preprocess the historical data of power grid operation status parameters and power grid frequency, and build a power grid frequency stability classification model based on the preprocessed historical data;

[0007] Obtain and normalize power grid operation parameters in real time to obtain a standard data set of power grid operation parameters; obtain a classification result of power grid operation status based on the standard data set of power grid operation parameters;

[0008] Preliminarily judging the stability based on the grid operation parameters and the grid frequency stability classification model; if it is unstable, calculating the grid frequency stability index through the pre-trained stability index calculation model and judging whether it is lower than the threshold;

[0009] An adaptive gain adjustment model is constructed. If the grid operation status classification result does not meet the requirements or the grid frequency stability index is lower than the threshold, the grid frequency stability index is used as the input of the adaptive gain adjustment model to obtain the optimal gain parameters of the primary frequency modulation control, and the optimization control is realized based on the optimal gain parameters of the primary frequency modulation control.

[0010] Optionally, the process of constructing a grid frequency stability classification model includes:

[0011] The historical data of the power grid operation status parameters are cleaned and normalized to obtain standardized power grid operation status parameter data; a mapping relationship between the power grid operation status parameters and the power grid frequency is constructed, and a classification model is constructed based on the support vector machine algorithm. The classification model is trained based on the standardized power grid operation status parameter data, the historical data of the power grid frequency and the mapping relationship to obtain a power grid frequency stability classification model.

[0012] Optionally, the grid operation parameters acquired in real time are denoised and standardized to obtain a grid operation parameter standard data set; feature extraction is performed on the grid operation parameter standard data set to obtain a grid operation state feature vector, and a grid operation state classification result is obtained based on the grid operation state feature vector.

[0013] Optionally, the process of calculating the grid frequency stability index includes:

[0014] Sensors and measuring devices are arranged at key nodes of the power grid to collect power grid operating parameters in real time; the collected power grid operating parameter data are preprocessed; a mathematical model describing the stability of the power grid is constructed, and the preprocessed power grid operating parameters are input into the mathematical model. By solving the model equation, the power grid frequency stability index is calculated.

[0015] Optionally, a stability index threshold is set based on expert experience to determine whether the grid frequency stability index exceeds the stability index threshold. If it is lower than the stability index threshold, an early warning is issued to obtain the optimal gain parameters of the primary frequency regulation control based on the grid frequency stability index and the adaptive gain adjustment model.

[0016] Optionally, the process of obtaining an optimal gain parameter of a frequency modulation control further includes:

[0017] The grid operation status data is obtained and multiple grid frequency stability indexes are calculated as input variables of the fuzzy controller; according to the numerical range of the grid frequency stability index, a fuzzy control rule base is established to form multiple fuzzy control subsystems; for each fuzzy control subsystem, fuzzy reasoning calculation is performed to obtain the fuzzy gain parameters of the primary frequency modulation control; the fuzzy gain parameters of multiple fuzzy control subsystems are weightedly fused, and the optimal gain parameters of the primary frequency modulation control are obtained through defuzzification processing.

[0018] Optionally, before inputting the grid frequency stability index into the adaptive gain adjustment model, the safety margin of the grid is calculated, and a real-time grid optimization control strategy is obtained based on the calculation results of the grid frequency stability index and the safety margin in combination with the optimization control strategy in the expert knowledge base.

[0019] Optionally, the process of achieving optimized control based on the optimal gain parameters of the primary frequency regulation control includes: integrating the optimal gain parameters of the primary frequency regulation control with the grid frequency deviation and active power, and generating a primary frequency regulation control instruction in combination with a real-time grid optimization control strategy; judging whether the primary frequency regulation control instruction meets the preset rationality conditions, and if so, sending the instruction to the target generator set through the communication network; if not, recalculating the optimal gain parameters of the primary frequency regulation control.

[0020] Optionally, grid operation parameters and generator set status information are obtained as input data for the adaptive gain adjustment model, and the model is trained and optimized in real time through signal dynamic update technology; data preprocessing technology is used to remove noise and extract features from the obtained grid operation parameters and generator set status information to obtain high-quality input data; the preprocessed data is input into the adaptive gain adjustment model, and the model is continuously optimized through a machine learning algorithm.

[0021] The present invention provides a primary frequency modulation optimization control system based on adaptive gain adjustment, comprising:

[0022] The grid frequency stability mathematical modeling module is used to obtain and preprocess the historical data of grid operation status parameters and grid frequency, and to build a grid frequency stability classification model based on the preprocessed historical data;

[0023] The grid operation parameter online monitoring module is used to obtain the grid operation parameters in real time and normalize them to obtain a grid operation parameter standard data set; based on the grid operation parameter standard data set, a grid operation status classification result is obtained;

[0024] A grid stability state judgment module, which preliminarily judges the stability based on the grid operation parameters and the grid frequency stability classification model. If the grid is unstable, the grid frequency stability index is calculated by a pre-trained stability index calculation model and whether it is lower than a threshold value is determined;

[0025] An adaptive gain adjustment module is used to construct an adaptive gain adjustment model. If the grid operation status classification result does not meet the requirements or the grid frequency stability index is lower than a threshold, the grid frequency stability index is used as an input of the adaptive gain adjustment model to obtain an optimal gain parameter for primary frequency modulation control.

[0026] The primary frequency modulation control instruction generation module is used to integrate the optimal gain parameters of the primary frequency modulation control with the grid frequency deviation and active power, and generate the primary frequency modulation control instructions in combination with the real-time grid optimization control strategy; judge whether the primary frequency modulation control instructions meet the preset rationality conditions, if so, send the instructions to the target generator set through the communication network; if not, recalculate the optimal gain parameters of the primary frequency modulation control;

[0027] The adaptive gain adjustment model optimization module is used to obtain the grid operation parameters and generator set status information as the input data of the adaptive gain adjustment model, and to train and optimize the model in real time through the signal dynamic update technology; among them, the acquired grid operation parameters and generator set status information are subjected to noise removal and feature extraction by data preprocessing technology to obtain high-quality input data; the preprocessed data is input into the adaptive gain adjustment model, and the model is continuously optimized through the machine learning algorithm.

[0028] Compared with the prior art, the present invention has the following advantages and technical effects:

[0029] The present invention establishes a mathematical model of power grid operation status and frequency stability, obtains power grid operation parameters through online monitoring, and calculates the power grid stability index. When the stability index is lower than the preset threshold, the power grid is judged to be unstable and a frequency modulation control is required. For different stability indexes, a fuzzy control algorithm is used to establish an adaptive gain adjustment model, and the stability index is used as input to obtain the optimal gain parameter through fuzzy reasoning. A frequency modulation control instruction is generated according to the parameter and sent to each generator set to adjust the active power output. During the control process, the operating parameters and generator set status information are continuously collected, and the model is trained and optimized online through the signal dynamic update technology to continuously improve the accuracy and robustness. The present invention realizes the precise adaptive control of the power grid frequency, effectively improves the stability and reliability of the power grid, and provides a strong guarantee for the safe and stable operation of the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0031] Figure 1The figure is a flow chart of a method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0033] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0034] Embodiment 1

[0035] like Figure 1 As shown, this embodiment provides a primary frequency modulation optimization control method based on adaptive gain adjustment, including:

[0036] To achieve the above object, the present invention provides a primary frequency modulation optimization control method based on adaptive gain adjustment, comprising:

[0037] Obtain and preprocess the historical data of power grid operation status parameters and power grid frequency, and build a power grid frequency stability classification model based on the preprocessed historical data;

[0038] In some specific implementations, the process of constructing a grid frequency stability classification model includes:

[0039] The historical data of the power grid operation status parameters are cleaned and normalized to obtain standardized power grid operation status parameter data; a mapping relationship between the power grid operation status parameters and the power grid frequency is constructed, and a classification model is constructed based on the support vector machine algorithm. The classification model is trained based on the standardized power grid operation status parameter data, the historical data of the power grid frequency and the mapping relationship to obtain a power grid frequency stability classification model.

[0040] Specifically, according to the power system theory, the grid operation status parameters and the grid frequency stability parameters are obtained, and the mapping relationship between the grid operation status and the grid frequency stability is established; for the obtained grid operation status parameters, the parameter data is cleaned and normalized by using data preprocessing technology to obtain standardized grid operation status parameters; according to the standardized grid operation status parameters, the support vector machine algorithm in the machine learning algorithm is used to train and establish a classification model of the grid operation status and the grid frequency stability; in the classification model training process, the kernel function, penalty coefficient and other hyperparameters of the support vector machine algorithm are optimized through grid search and cross-validation methods to improve the classification performance of the model.

[0041] For example, the relationship between the operation state of the power grid and the frequency stability is one of the core issues of the operation of the power system. First, it is necessary to obtain the parameters of the operation state of the power grid, such as generator power, load demand, line flow, etc., as well as frequency stability parameters such as system inertia and frequency regulation capability. These parameters reflect the real-time operation of the power grid. It is necessary to preprocess the acquired raw data. For example, the generator power data is normalized to unify the output of units of different capacities to the range of 0-1, which is convenient for subsequent modeling and analysis. At the same time, it is also necessary to process outliers, such as eliminating obvious erroneous data caused by communication failures. Support vector machine is an excellent classification algorithm suitable for the discrimination of power grid frequency stability. During the training process, the grid search method can be used to optimize the kernel function parameters. For example, for the Gaussian kernel, different γ values ​​(such as 0.1, 1, 10, etc.) can be tried to select the parameters with the best classification effect. At the same time, the penalty coefficient C should be optimized to balance the complexity of the model and the classification accuracy.

[0042] Obtain and normalize power grid operation parameters in real time to obtain a standard data set of power grid operation parameters; obtain a classification result of power grid operation status based on the standard data set of power grid operation parameters;

[0043] In some specific implementations, the grid operation parameters acquired in real time are denoised and standardized to obtain a grid operation parameter standard data set; feature extraction is performed on the grid operation parameter standard data set to obtain a grid operation state feature vector, and a grid operation state classification result is obtained based on the grid operation state feature vector.

[0044] Specifically, according to the pre-established threshold range of power grid operation parameters, it is determined whether the real-time collected power grid operation parameters exceed the threshold range. If they exceed the threshold range, an early warning signal is triggered, indicating that the power grid operation status is abnormal. The power grid operation parameters are collected in real time through online monitoring technology, and parameter data such as voltage, current, active power and reactive power are obtained, and the collected parameter data are transmitted to the data processing center. In the data processing center, data cleaning and data preprocessing techniques are used to denoise and standardize the collected power grid operation parameter data to obtain a standardized power grid operation parameter data set. Feature parameters are extracted from the standardized power grid operation parameter data set to construct a power grid operation status feature vector as input data for judging the power grid operation status. The support vector machine algorithm is used to classify the power grid operation status feature vector to obtain the power grid operation status classification result to determine whether the current power grid operation status is normal.

[0045] Exemplarily, the threshold range of power grid operation parameters is an important basis for ensuring the safe and stable operation of the power grid. For example, for a transmission line with a voltage level of 220kV, the normal operating voltage range is usually 198kV to 242kV. If the voltage value is monitored in real time to exceed this range, the system will trigger an early warning signal. This early warning mechanism can promptly detect potential power grid operation anomalies and provide an important basis for operation and maintenance personnel to take corresponding measures. Online monitoring technology is the key to obtaining power grid operation parameters in real time. Taking the substation as an example, through sensors installed on key equipment such as the main transformer and circuit breaker, parameters such as voltage, current, active power and reactive power can be collected in real time. These data are transmitted to the data processing center through the communication network to provide raw data support for subsequent analysis. Data cleaning and preprocessing are important links to ensure data quality. For example, for the collected voltage data, there may be abnormal values ​​caused by equipment failure or communication interference. These noises can be effectively removed by methods such as median filtering to improve data reliability. At the same time, normalizing parameters of different dimensions can make subsequent analysis more accurate. Feature extraction is a key step in constructing the feature vector of power grid operation status. For example, indicators such as voltage deviation rate and power factor can be calculated as feature parameters. These feature parameters can better reflect the operation status of the power grid and provide effective input for subsequent classification judgment. The support vector machine algorithm has a good performance in the classification of power grid operation status. By training historical data, a classification model that can distinguish between normal and abnormal operation states can be established. For example, when the input feature vector shows that the voltage deviation rate exceeds 5% and the power factor is less than 0.85, the model may judge it as an abnormal state.

[0046] Preliminarily judging the stability based on the grid operation parameters and the grid frequency stability classification model; if it is unstable, calculating the grid frequency stability index through the pre-trained stability index calculation model and judging whether it is lower than the threshold;

[0047] In some specific implementations, the process of calculating the power grid frequency stability index includes:

[0048] Sensors and measuring devices are arranged at key nodes of the power grid to collect power grid operating parameters in real time; the collected power grid operating parameter data are preprocessed; a mathematical model describing the stability of the power grid is constructed, and the preprocessed power grid operating parameters are input into the mathematical model. By solving the model equation, the power grid frequency stability index is calculated.

[0049] Specifically, by arranging sensors and measuring devices at key nodes of the power grid, the power grid operation parameters, including voltage, current, active power, reactive power, etc., are collected in real time. The collected power grid operation parameter data is transmitted to the data center, and after data preprocessing, outliers and noise data are eliminated to obtain standardized parameter data. According to the physical model and operation characteristics of the power system, a mathematical model describing the stability of the power grid is established, and the model comprehensively considers static stability and dynamic stability. The real-time collected power grid parameters are input into the established mathematical model, and the stability index reflecting the stability level of the power grid at the current moment is calculated by solving the model equation. An adaptive gain adjustment model is constructed. If the classification result of the power grid operation status does not meet the requirements or the power grid frequency stability index is lower than the threshold, the power grid frequency stability index is used as the input of the adaptive gain adjustment model to obtain the optimal gain parameter of the primary frequency modulation control, and the optimal control is realized based on the optimal gain parameter of the primary frequency modulation control.

[0050] In some specific implementations, a stability index threshold is set based on expert experience to determine whether the grid frequency stability index exceeds the stability index threshold. If it is lower than the stability index threshold, an early warning is issued, and the optimal gain parameters of the primary frequency regulation control are obtained based on the grid frequency stability index and the adaptive gain adjustment model.

[0051] Specifically, it is determined whether the calculated stability index exceeds the preset safety threshold. If so, an early warning is issued to indicate that there is a risk of instability in the power grid operation and measures need to be taken.

[0052] In some specific implementations, the process of obtaining the optimal gain parameter of the primary frequency modulation control further includes:

[0053] The grid operation status data is obtained and multiple grid frequency stability indexes are calculated as input variables of the fuzzy controller; according to the numerical range of the grid frequency stability index, a fuzzy control rule base is established to form multiple fuzzy control subsystems; for each fuzzy control subsystem, fuzzy reasoning calculation is performed to obtain the fuzzy gain parameters of the primary frequency modulation control; the fuzzy gain parameters of multiple fuzzy control subsystems are weightedly fused, and the optimal gain parameters of the primary frequency modulation control are obtained through defuzzification processing.

[0054] Specifically, the grid operation status data is obtained, and multiple grid stability indexes are calculated as input variables of the fuzzy controller; according to the numerical range of the grid stability index, a fuzzy control rule base is established to form multiple fuzzy control subsystems; for each fuzzy control subsystem, fuzzy reasoning calculation is performed to obtain the fuzzy gain parameters of the primary frequency modulation control; the fuzzy gain parameters of multiple fuzzy control subsystems are weightedly fused, and the optimal gain parameters of the primary frequency modulation control are determined through defuzzification processing.

[0055] Exemplarily, first obtain the real-time operation status data of the power grid, such as the output power of the generator set, the line flow, the node voltage, etc., which are the basis for evaluating the stability of the power grid. Then, based on these data, calculate multiple power grid stability indexes, such as the frequency stability index, the voltage stability index and the small disturbance stability index. Assume that at a certain moment, the frequency stability index is calculated to be 0.95, the voltage stability index is 0.98, and the small disturbance stability index is 0.92. These indexes together reflect the current stability state of the power grid. Take these three stability indexes as input variables of the fuzzy controller. According to the numerical range of each power grid stability index and the actual operation experience, establish a fuzzy control rule base. For example, the range of the frequency stability index is divided into three fuzzy sets: low, medium and high. When the frequency stability index is low, it means that the power grid frequency stability is poor and a larger frequency modulation gain is required; when the frequency stability index is high, a smaller frequency modulation gain can be used. For different combinations of stability indexes, formulate corresponding fuzzy control rules. For example, when the frequency stability index is "low" and the voltage stability index is "medium", the output fuzzy gain parameter is "large"; when the frequency stability index is "high" and the voltage stability index is "low", the output fuzzy gain parameter is "medium". These rules constitute the fuzzy control subsystem. For each fuzzy control subsystem, fuzzy reasoning calculation is performed according to the current power grid stability index. For example, the current frequency stability index is 0.95, which belongs to the "high" fuzzy set, and the voltage stability index is 0.98, which belongs to the "medium" fuzzy set. According to the corresponding fuzzy rules, it can be inferred that the fuzzy gain parameter of the primary frequency modulation control is "medium". The fuzzy gain parameters of multiple fuzzy control subsystems are weighted and fused to obtain a comprehensive fuzzy gain parameter. For example, the fuzzy gain parameters output by the three subsystems are "large", "medium" and "small" respectively. According to the weights of each subsystem, a weighted average is performed, and finally a comprehensive fuzzy gain parameter of "medium to large" is obtained. Then, the fuzzy gain parameter is converted into an accurate value through defuzzification. For example, the fuzzy gain parameter of "medium to large" is converted into a specific value, such as 1.2, and this value will be used as the actual gain parameter of a frequency modulation control.

[0056] Furthermore, if the grid stability index changes, the fuzzy controller is triggered to adaptively adjust the gain parameters and update the fuzzy rule base; the optimized gain parameters are applied to the primary frequency modulation control algorithm to adjust the active output of the generator set in real time to maintain grid frequency stability; the grid stability index change trend is continuously monitored to evaluate the primary frequency modulation control effect, and the above steps are repeated when necessary to achieve adaptive optimization of the gain parameters.

[0057] Specifically, if the grid stability index changes, for example, the frequency stability index drops from 0.95 to 0.85, the fuzzy controller is triggered to adaptively adjust the gain parameters. According to the new stability index, fuzzy reasoning and weighted fusion are re-performed to obtain new fuzzy gain parameters. At the same time, the fuzzy rule base is updated according to historical operation data and expert experience to adapt to different grid operation states. The optimized gain parameters are applied to the primary frequency modulation control algorithm to adjust the active output of the generator set in real time. For example, when the grid frequency drops, according to the gain parameter 1.2, the generator set is controlled to increase the active output to increase the grid frequency. In this way, the grid frequency change can be quickly responded to and the frequency can be maintained within the allowable range. With fuzzy control, the control parameters can be adaptively adjusted according to the grid stability state to improve the accuracy and robustness of the control. During the primary frequency modulation control process, the changing trend of the grid stability index is continuously monitored to evaluate the effect of the primary frequency modulation control. For example, observe whether the frequency stability index returns to the normal range and whether the voltage stability index remains stable. If the control effect is not good, for example, the frequency stability index is still low, it is necessary to repeat the above steps, readjust the gain parameters, and optimize the fuzzy rule base until a satisfactory control effect is achieved. Through this feedback mechanism, the adaptive optimization of gain parameters can be achieved, and the performance of primary frequency modulation control can be improved. The use of fuzzy control and adaptive optimization technology can improve the stability and reliability of the power grid and ensure the safe operation of the power system.

[0058] In some specific embodiments, before the grid frequency stability index is input into the adaptive gain adjustment model, the safety margin of the grid is calculated, and a real-time grid optimization control strategy is obtained based on the calculation results of the grid frequency stability index and the safety margin in combination with the optimization control strategy in the expert knowledge base.

[0059] In some specific embodiments, the process of achieving optimized control based on the optimal gain parameters of the primary frequency modulation control includes: integrating the optimal gain parameters of the primary frequency modulation control with the grid frequency deviation and active power, and generating a primary frequency modulation control instruction in combination with a real-time grid optimization control strategy; determining whether the primary frequency modulation control instruction meets the preset rationality conditions, and if so, sending the instruction to the target generator set through the communication network; if not, recalculating the optimal gain parameters of the primary frequency modulation control.

[0060] Specifically, according to the real-time operation parameters of the power grid, an adaptive gain adjustment model is used to calculate the optimal gain parameter as the basis for generating the frequency control instruction. The gain parameter output by the adaptive gain adjustment model is integrated with the parameters such as the power grid frequency deviation and active power, and a primary frequency control instruction is generated through a machine learning algorithm. It is judged whether the generated primary frequency control instruction meets the preset rationality conditions. If it does, the instruction is sent to the target generator set through the communication network; if it does not, it returns to step 1 to recalculate the gain parameter. In the communication network, security mechanisms such as encrypted transmission and identity authentication are used to ensure the integrity and reliability of the primary frequency control instruction and prevent the instruction from being tampered with or counterfeited. After receiving the primary frequency control instruction, the generator set extracts the active power adjustment value in the instruction and determines the final active power output adjustment value according to the operating status and adjustment capability of the unit. The control system of the generator set adjusts the parameters such as the feed water and steam turbine start and stop of the unit according to the determined active power output adjustment value to achieve precise control of the active power output, and feeds back the adjustment result to the dispatching center. Acquire real-time grid frequency data to determine whether the frequency regulation control effect has reached the expected target. If not, the frequency deviation and other parameters will be fed back to the adaptive gain adjustment model for the next round of iterative optimization. If it has been achieved, maintain the current frequency regulation state and continuously monitor the changes in grid frequency.

[0061] Exemplarily, the adaptive gain adjustment model is a core component of the primary frequency modulation control system, which optimizes the control gain by analyzing the grid operation parameters in real time. For example, when the grid frequency deviation is large, the model may output a larger proportional gain and integral gain to speed up the adjustment speed; when the frequency is close to the rated value, the gain may be reduced to avoid over-regulation. Machine learning algorithms play an important role in generating frequency modulation control instructions. Taking the support vector machine (SVM) as an example, the frequency deviation, active power change, etc. in the historical data can be used as features, and the corresponding frequency modulation instructions can be trained as labels. The trained model can quickly generate appropriate frequency modulation instructions according to the current grid state. Instruction rationality judgment is a key link to ensure system security. Multiple judgment conditions can be set, such as the instruction amplitude does not exceed 10% of the rated power of the unit, the adjustment rate does not exceed the unit climbing rate, etc. If the conditions are not met, the system will recalculate the gain and generate a new instruction until the requirements are met. Communication security is crucial to the power system. The AES encryption algorithm can be used to encrypt the frequency modulation instruction, and the digital signature technology can be used to ensure the integrity of the instruction. At the same time, a VPN channel can be established between the dispatching center and the power plant to further improve communication security. After receiving the command, the generator set needs to process it according to its own characteristics. For example, for thermal power units, it may be necessary to adjust the water supply and fuel supply at the same time; while for hydropower units, it is mainly to adjust the guide vane opening. The regulation characteristics of different types of units are different, so it is necessary to design control strategies in a targeted manner. Accurate control of active power is the basis for achieving frequency regulation. Taking thermal power units as an example, the generator output can be controlled by adjusting parameters such as boiler feed water and turbine regulating valve opening. The control system needs to consider factors such as the thermal characteristics and mechanical characteristics of the unit, and use algorithms such as PID control to achieve fast and accurate power regulation. The evaluation of frequency regulation effect is a continuous process. The frequency deviation can be set to no more than ±0.02Hz as the standard. If it does not meet the standard, the system will feedback the deviation between the actual frequency and the target frequency, the overshoot during the regulation process and other indicators to the adaptive model for the next round of optimization. This closed-loop feedback mechanism can continuously improve the frequency regulation performance of the system. Through the coordinated cooperation of the above links, the primary frequency regulation control system can effectively maintain the stability of the power grid frequency. This not only improves the quality of power, but also provides an important guarantee for the safe and stable operation of the power grid. At the same time, the application of advanced technologies such as adaptive regulation and machine learning has enabled the system to have stronger adaptability and intelligence, enabling it to better cope with the complex and changing power grid operating environment.

[0062] In some specific implementations, grid operation parameters and generator set status information are obtained as input data for an adaptive gain adjustment model, and the model is trained and optimized in real time through signal dynamic update technology; wherein, data preprocessing technology is used to remove noise and extract features from the obtained grid operation parameters and generator set status information to obtain high-quality input data; the preprocessed data is input into the adaptive gain adjustment model, and the model is continuously optimized through a machine learning algorithm.

[0063] Specifically, the grid operation parameters and generator set status information are obtained and used as input data of the adaptive gain adjustment model. The model is trained and optimized in real time through the signal dynamic update technology. For the obtained grid operation parameters and generator set status information, data preprocessing technology is used to remove noise and extract features to obtain high-quality input data. The preprocessed data is input into the adaptive gain adjustment model, and the model is trained through machine learning algorithms such as support vector machines or neural networks to continuously optimize the model parameters. In the model training process, cross-validation and early stopping mechanisms are introduced to avoid model overfitting and improve the generalization ability and robustness of the model. According to the optimized adaptive gain adjustment model obtained by training, the newly collected grid operation parameters and generator set status information are predicted and analyzed to obtain the optimal gain adjustment strategy. The optimal gain adjustment strategy is applied to the frequency modulation control process, and the output power of the generator set is adaptively adjusted to achieve stable control of the grid frequency. The grid operation status and generator set performance are continuously monitored. If an abnormality occurs or the prediction accuracy decreases, the retraining and optimization of the adaptive gain adjustment model is triggered to ensure the reliability and effectiveness of the frequency modulation control.

[0064] Exemplarily, the adaptive gain adjustment model is the core technology of power grid frequency control, and its input data includes power grid operation parameters and generator set status information. For example, power grid operation parameters may include system frequency, active power balance, voltage level, etc., while generator set status information may include unit output, speed, excitation current, etc. These data usually have noise and redundancy, so they need to be preprocessed. Data preprocessing technology can use filtering algorithms to remove high-frequency noise, such as using low-pass filters or median filters. Feature extraction can apply dimensionality reduction methods such as principal component analysis (PCA) to extract the most representative features. For example, for a generator set, the original data may contain dozens of parameters, which can be reduced to several main features through PCA, greatly reducing the amount of subsequent calculations. Machine learning algorithms play an important role in the adaptive gain adjustment model. Support vector machine (SVM) is suitable for small sample learning and can effectively handle emergencies in the power grid. Neural networks perform well in processing large amounts of historical data and can capture complex nonlinear relationships. For example, a multi-layer perceptron network can be constructed, with the input layer corresponding to the preprocessed features, the hidden layer using the ReLU activation function, and the output layer giving the optimal gain parameters. To prevent the model from overfitting, cross validation is a common technique. K-fold cross validation can be used to divide the data set into K parts, and each time K-1 parts are used for training and 1 part for validation, and the cycle is repeated K times. The early stopping mechanism stops training when the performance of the validation set no longer improves. The combination of these two methods can effectively improve the generalization ability of the model. The application of the optimal gain adjustment strategy is the key to the whole process. For example, when it is detected that the system frequency deviates from the rated value, the model may give suggestions: increase the adjustment system gain for some fast-responding units, and appropriately reduce the gain for slower-responding units. This differentiated strategy can make full use of the characteristics of various units to achieve more accurate frequency control. Continuous monitoring and model retraining mechanisms ensure the long-term reliability of the system. For example, a sliding time window can be set to regularly evaluate the prediction accuracy of the model in the recent period. If the accuracy is lower than the preset threshold, such as 95%, the model retraining process is triggered. This can adapt to seasonal changes in grid load characteristics or dynamic changes in the system caused by the addition of new power generation equipment. Through this adaptive gain adjustment technology, the grid frequency regulation control system can better cope with challenges such as load fluctuations and new energy access, and improve the stability and reliability of the grid. At the same time, due to more precise control, it can also reduce the wear of the generator set, extend the life of the equipment, and achieve both economic and technical benefits.

[0065] Embodiment 2

[0066] This embodiment provides a primary frequency modulation optimization control system based on adaptive gain adjustment, including:

[0067] The grid frequency stability mathematical modeling module is used to obtain and preprocess the historical data of grid operation status parameters and grid frequency, and to build a grid frequency stability classification model based on the preprocessed historical data;

[0068] The grid operation parameter online monitoring module is used to obtain the grid operation parameters in real time and normalize them to obtain a grid operation parameter standard data set; based on the grid operation parameter standard data set, a grid operation status classification result is obtained;

[0069] A grid stability state judgment module, which preliminarily judges the stability based on the grid operation parameters and the grid frequency stability classification model. If the grid is unstable, the grid frequency stability index is calculated by a pre-trained stability index calculation model and whether it is lower than a threshold value is determined;

[0070] An adaptive gain adjustment module is used to construct an adaptive gain adjustment model. If the grid operation status classification result does not meet the requirements or the grid frequency stability index is lower than a threshold, the grid frequency stability index is used as an input of the adaptive gain adjustment model to obtain an optimal gain parameter for primary frequency modulation control.

[0071] The primary frequency modulation control instruction generation module is used to integrate the optimal gain parameters of the primary frequency modulation control with the grid frequency deviation and active power, and generate the primary frequency modulation control instructions in combination with the real-time grid optimization control strategy; judge whether the primary frequency modulation control instructions meet the preset rationality conditions, if so, send the instructions to the target generator set through the communication network; if not, recalculate the optimal gain parameters of the primary frequency modulation control;

[0072] The adaptive gain adjustment model optimization module is used to obtain the grid operation parameters and generator set status information as the input data of the adaptive gain adjustment model, and to train and optimize the model in real time through the signal dynamic update technology; among them, the acquired grid operation parameters and generator set status information are subjected to noise removal and feature extraction by data preprocessing technology to obtain high-quality input data; the preprocessed data is input into the adaptive gain adjustment model, and the model is continuously optimized through the machine learning algorithm.

[0073] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A primary frequency modulation optimization control method based on adaptive gain adjustment, characterized in that: The following steps are involved: Obtain and preprocess the historical data of power grid operation status parameters and power grid frequency, and build a power grid frequency stability classification model based on the preprocessed historical data; Obtain and normalize power grid operation parameters in real time to obtain a standardized data set of power grid operation parameters; Obtaining grid operation status classification results based on grid operation parameter specification data set; Preliminarily judging the stability based on the grid operation parameters and the grid frequency stability classification model; if it is unstable, calculating the grid frequency stability index through the pre-trained stability index calculation model and judging whether it is lower than the threshold; An adaptive gain adjustment model is constructed. If the grid operation status classification result does not meet the requirements or the grid frequency stability index is lower than the threshold, the grid frequency stability index is used as the input of the adaptive gain adjustment model to obtain the optimal gain parameters of the primary frequency modulation control, and the optimization control is realized based on the optimal gain parameters of the primary frequency modulation control.

2. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 1, characterized in that: The process of building a grid frequency stability classification model includes: The historical data of the power grid operation status parameters are cleaned and normalized to obtain standardized power grid operation status parameter data; a mapping relationship between the power grid operation status parameters and the power grid frequency is constructed, and a classification model is constructed based on the support vector machine algorithm. The classification model is trained based on the standardized power grid operation status parameter data, the historical data of the power grid frequency and the mapping relationship to obtain a power grid frequency stability classification model.

3. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 1, characterized in that: De-noising and standardizing are performed on the grid operation parameters acquired in real time to obtain a grid operation parameter standard data set; feature extraction is performed on the grid operation parameter standard data set to obtain a grid operation state feature vector, and a grid operation state classification result is obtained based on the grid operation state feature vector.

4. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 1, characterized in that: The process of calculating the grid frequency stability index includes: Sensors and measuring devices are arranged at key nodes of the power grid to collect power grid operating parameters in real time; the collected power grid operating parameter data are preprocessed; a mathematical model describing the stability of the power grid is constructed, and the preprocessed power grid operating parameters are input into the mathematical model. By solving the model equation, the power grid frequency stability index is calculated.

5. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 1, characterized in that: The stability index threshold is set based on expert experience to determine whether the power grid frequency stability index exceeds the stability index threshold. If it is lower than the stability index threshold, an early warning is issued. The optimal gain parameters of the primary frequency regulation control are obtained based on the power grid frequency stability index and the adaptive gain adjustment model.

6. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 5, characterized in that: The process of obtaining the optimal gain parameter of the primary frequency modulation control also includes: The grid operation status data is obtained and multiple grid frequency stability indexes are calculated as input variables of the fuzzy controller; according to the numerical range of the grid frequency stability index, a fuzzy control rule base is established to form multiple fuzzy control subsystems; for each fuzzy control subsystem, fuzzy reasoning calculation is performed to obtain the fuzzy gain parameters of the primary frequency modulation control; the fuzzy gain parameters of multiple fuzzy control subsystems are weightedly fused, and the optimal gain parameters of the primary frequency modulation control are obtained through defuzzification processing.

7. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 5, characterized in that: Before inputting the grid frequency stability index into the adaptive gain adjustment model, the safety margin of the grid is calculated. According to the calculation results of the grid frequency stability index and the safety margin, combined with the optimization control strategy in the expert knowledge base, a real-time grid optimization control strategy is obtained.

8. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 1, characterized in that: The process of achieving optimized control based on the optimal gain parameters of primary frequency regulation control includes: integrating the optimal gain parameters of primary frequency regulation control with the grid frequency deviation and active power, and generating primary frequency regulation control instructions in combination with real-time grid optimization control strategies; judging whether the primary frequency regulation control instructions meet the preset rationality conditions, and if so, sending the instructions to the target generator set through the communication network; if not, recalculating the optimal gain parameters of primary frequency regulation control.

9. The primary frequency modulation optimization control method based on adaptive gain adjustment according to claim 1, characterized in that: Also includes: The grid operation parameters and generator set status information are obtained as input data for the adaptive gain adjustment model, and the model is trained and optimized in real time through the signal dynamic update technology; the obtained grid operation parameters and generator set status information are subjected to noise removal and feature extraction by using data preprocessing technology to obtain high-quality input data; The preprocessed data is input into the adaptive gain adjustment model, and the model is continuously optimized through machine learning algorithms.

10. The method and system for primary frequency modulation optimization control based on adaptive gain adjustment according to claim 1, characterized in that: The grid frequency stability mathematical modeling module is used to obtain and preprocess the historical data of grid operation status parameters and grid frequency, and to build a grid frequency stability classification model based on the preprocessed historical data; The grid operation parameter online monitoring module is used to obtain the grid operation parameters in real time and normalize them to obtain a grid operation parameter standard data set; based on the grid operation parameter standard data set, a grid operation status classification result is obtained; A grid stability state judgment module, which preliminarily judges the stability based on the grid operation parameters and the grid frequency stability classification model. If the grid is unstable, the grid frequency stability index is calculated by a pre-trained stability index calculation model and whether it is lower than a threshold value is determined; An adaptive gain adjustment module is used to construct an adaptive gain adjustment model. If the grid operation status classification result does not meet the requirements or the grid frequency stability index is lower than a threshold, the grid frequency stability index is used as an input of the adaptive gain adjustment model to obtain an optimal gain parameter for primary frequency modulation control. The primary frequency modulation control instruction generation module is used to integrate the optimal gain parameters of the primary frequency modulation control with the grid frequency deviation and active power, and generate the primary frequency modulation control instructions in combination with the real-time grid optimization control strategy; judge whether the primary frequency modulation control instructions meet the preset rationality conditions, if so, send the instructions to the target generator set through the communication network; if not, recalculate the optimal gain parameters of the primary frequency modulation control; The adaptive gain adjustment model optimization module is used to obtain the grid operation parameters and generator set status information as the input data of the adaptive gain adjustment model, and to train and optimize the model in real time through the signal dynamic update technology; among them, the acquired grid operation parameters and generator set status information are subjected to noise removal and feature extraction by using data preprocessing technology to obtain high-quality input data; The preprocessed data is input into the adaptive gain adjustment model, and the model is continuously optimized through machine learning algorithms.

Images