Power grid impedance adjusting system based on adaptive event triggering and hybrid deep learning

Abstract

The invention discloses a power grid impedance adjustment system based on adaptive event triggering and hybrid deep learning, and the system specifically comprises an adaptive event triggering impedance estimation module, an HDRL-based impedance optimization module, a real-time feedback adjustment module, and an adaptive event triggering and HDRL integration mechanism. The adaptive event-triggered impedance estimation module, the HDRL-based impedance optimization module and the real-time feedback regulation module realize linkage control through data interaction and signal feedback, and when the fluctuation of PCC voltage or current exceeds a dynamic threshold value, the adaptive event-triggered module activates impedance estimation and takes an estimation result as one of the input states of the HDRL, so that the real-time feedback regulation module can realize real-time feedback regulation. The HDRL determines the optimal impedance control action according to the current PCC state and transmits the optimal impedance control action to the feedback regulation module, through the combined design of a self-adaptive event triggering mechanism, an HDRL optimization strategy and a feedback control mechanism, real-time optimization and intelligent control of power grid impedance are achieved, and the stability and the electric energy quality of a power grid are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system and power quality management, and in particular to a power grid impedance regulation system based on adaptive event triggering and hybrid deep learning. Background Art

[0002] With the rapid development of new energy technologies, the access of distributed energy, such as solar energy and wind energy, to the power grid has become a global trend. The access of distributed energy has not only changed the operation mode of traditional power grids, but also brought a series of power quality problems such as voltage fluctuations, power fluctuations, and harmonic pollution. In order to cope with these problems, the control and optimization technology of the power grid has been continuously developed. In the field of power grid impedance regulation, researchers have proposed a variety of methods to improve the stability and power quality of the power grid. Traditional power grid impedance control methods are generally divided into two categories: passive control methods and active control methods. Passive control methods achieve control of the stability and impedance of the power grid by adjusting physical components. They have good stability, but low accuracy and cannot adapt to the dynamically changing power grid environment. Active control methods estimate and adjust the impedance of the power grid in real time through power electronic equipment, and can accurately respond to changes in the power grid, but they often require complex calculations and frequent signal disturbances, which not only increases the calculation burden, but may also cause power quality problems. Chinese patent CN117630487B proposes a virtual impedance control method based on distributed power sources. The method is used to improve the dynamic characteristics and voltage quality problems of distributed power sources. By adjusting the virtual impedance parameters, the power flow of the power grid is optimized.

[0003] Although the adaptive event triggering mechanism in the current technology has certain advantages and can reduce unnecessary computing burden, its convergence speed is slow and it does not respond promptly to rapidly changing loads. With the introduction of artificial intelligence technology, many studies have begun to try to combine AI technology to solve complex problems in power grid control, especially through deep learning and reinforcement learning to optimize the estimation and adjustment strategy of power grid impedance, but the existing technology still has shortcomings in real-time, adaptability and control accuracy. Summary of the invention

[0004] The present invention provides a power grid impedance regulation system based on adaptive event triggering and hybrid deep learning, which can effectively solve the problems that the traditional methods proposed in the above background technology have poor real-time and adaptability, and are difficult to effectively cope with rapidly changing loads. Most impedance estimation methods rely on frequent calculations or injection of interference signals, which leads to excessive calculation burden and may cause power quality problems. In addition, traditional control methods lack intelligence and are difficult to adapt to complex dynamic load environments, resulting in delayed system response and limited contribution to power grid stability. Therefore, the impedance control efficiency and accuracy of the prior art in dynamic environments are obviously insufficient.

[0005] To achieve the above objectives, the present invention provides the following technical solutions: a grid impedance adjustment system based on adaptive event triggering and hybrid deep learning, impedance estimation based on adaptive event-driven, intelligent control optimization and dynamic feedback adjustment, to achieve real-time adjustment of grid impedance, reduce harmonic distortion, alleviate voltage fluctuations, and improve power quality;

[0006] Specifically, it includes an impedance estimation module triggered by an adaptive event, an impedance optimization module based on HDRL, a real-time feedback adjustment module, and an integration mechanism between adaptive event triggering and HDRL;

[0007] The adaptive event-triggered impedance estimation module, HDRL-based impedance optimization module, and real-time feedback regulation module realize linkage control through data interaction and signal feedback. When the fluctuation of PCC voltage or current exceeds the dynamic threshold, the adaptive event triggering module activates impedance estimation and uses the estimation result as one of the input states of HDRL. HDRL determines the optimal impedance control action according to the current PCC state and transmits it to the feedback regulation module.

[0008] The real-time feedback regulation module further adjusts and optimizes the grid impedance through a weighted combination of PID control signals and HDRL control signals. This full-link linkage control structure can ensure the flexibility and stability of the control strategy in scenarios with drastic load fluctuations, reduce the conduction effect of load disturbances, and significantly improve power quality and grid stability.

[0009] According to the above technical solution, the adaptive event-triggered impedance estimation module dynamically monitors the voltage and current fluctuations of the common coupling point PCC, adopts a dynamic threshold adjustment mechanism and a recursive least squares method RLS combined with a Kalman filter estimation method to achieve real-time estimation of the grid impedance.

[0010] According to the above technical solution, the HDRL-based impedance optimization module divides impedance control into a two-layer control structure of discrete actions and continuous actions based on a hybrid deep reinforcement learning HDRL algorithm to cope with complex load fluctuations and impedance changes in the power grid.

[0011] According to the above technical solution, the real-time feedback regulation module is responsible for quickly correcting and adjusting the voltage and current of the power grid under the coordinated action of the HDRL control signal and the PID control signal. The function of the PID controller is to handle small load fluctuations, while the HDRL compensation module is responsible for quickly correcting the control output of the system when a large load change occurs in the power grid to prevent the voltage or current from exceeding the threshold range.

[0012] According to the above technical solution, the adaptive event-triggered impedance estimation module is responsible for real-time monitoring of the voltage and current fluctuations at the common coupling point PCC of the power grid, and automatically triggers the impedance estimation process according to these fluctuations. The core goal of this module is to reduce the computational burden, respond quickly when the light load fluctuations of the power grid are significant, and improve the stability and power quality of the power grid. It adopts a dynamic threshold adjustment mechanism instead of the traditional fixed threshold mechanism. The event-triggered threshold can be dynamically adjusted according to the real-time voltage and current fluctuations, thereby avoiding excessive calculation and ensuring that the impedance estimation process is accurately triggered when needed.

[0013] According to the above technical solution, the dynamic threshold adjustment mechanism is based on the voltage and current fluctuation amplitude, and uses an adaptive filter to effectively separate the grid signal and noise, automatically identify the effective grid disturbance signal, and when the grid fluctuation is small or the background noise is large, the impedance estimation will not be triggered frequently, which reduces the calculation load of the system and improves the response efficiency of the system. Only when the grid fluctuation reaches a certain level, the impedance estimation process will be triggered to ensure the stability of the system.

[0014] In order to further improve the effect of dynamic threshold adjustment, a sliding window algorithm is proposed. This algorithm dynamically adjusts the threshold by calculating the maximum and minimum values ​​of the voltage within the set time window, and updates the threshold base at each time node. It can effectively cope with common random fluctuations in the power grid and has a high anti-interference ability to light load fluctuations.

[0015] Specifically, suppose that in the time window T win The maximum and minimum values ​​of the voltage signal V(t) are V max (t) and V min (t), the sliding window algorithm can calculate the current fluctuation base value V by the following formula base (t): Based on this fluctuating base value, the system can dynamically adjust the threshold for event triggering to ensure that impedance estimation can be triggered in a timely manner when the power grid changes significantly;

[0016] When the event triggering conditions are met, the system will start the impedance estimation process. In order to achieve accurate real-time estimation, the recursive least squares RLS and Kalman filter algorithms are combined. The RLS algorithm can quickly update the impedance estimation value, and the Kalman filter can effectively suppress the noise in the power grid environment and improve the accuracy of the estimation result. Specifically, the mathematical formula of the impedance estimation process is as follows:

[0017] Z est (t) = Z est (t-1)+K(t)[V(t)-I(t)Z est (t-1)];

[0018] Among them, Z est is the impedance estimation value at the current time t, V(t) is the current voltage value, I(t) is the current current value, K(t) is the adaptive gain, which is dynamically adjusted based on the real-time voltage, current and impedance inverse prediction value, Z est (t-1) is the impedance estimate at the previous moment, and the gain K(t) is calculated based on the current voltage V(t), current I(t) and impedance Z est The prediction error of (t-1) is dynamically calculated, and the adaptive gain can optimize the estimation result according to the real-time fluctuation and background noise;

[0019] The introduction of the Kalman filter enables the impedance estimation process to maintain a high estimation accuracy in a noisy environment. The specific Kalman gain update formula is as follows:

[0020] K(t)=P(t-1)[P(t-1)+R(t)] -1

[0021] P(t)=[P(t-1) -1 +Q(t)] -1 ;

[0022] Among them, P(t) is the estimation error covariance matrix, Q(t) is the process noise covariance, and R(t) is the measurement noise covariance. Through the combination of RLS and Kalman filter, the system can accurately estimate the grid impedance in a noisy environment and has a high real-time response capability.

[0023] According to the above technical solution, the HDRL-based impedance optimization module is the core intelligent control unit, which is designed to adjust the grid impedance in real time, thereby optimizing the power quality and improving the stability of the grid. The module adopts a hybrid deep reinforcement learning algorithm, combining discrete and continuous action spaces, to ensure that the grid can quickly and accurately perform impedance adjustment optimization when the load fluctuates and the grid state changes;

[0024] The HDRL algorithm selects discrete actions through the policy network and refines and optimizes these discrete actions through the value network to achieve the optimal effect of impedance regulation. It can adapt to frequent load changes in the power grid while avoiding frequent computing operations, improving real-time response capabilities and overall stability.

[0025] According to the above technical solution, the HDRL-based impedance optimization module needs to first design the state space of HDRL. The design of the state space involves the system's real-time perception and feedback mechanism of the power grid. The input vector of the system mainly includes the voltage, current and estimated power grid impedance value of the common coupling point PCC in the power grid. Through the voltage and current data collected by real-time sensors, the system can monitor the dynamic changes of the power grid in real time and adjust the power grid impedance according to the changes. The design of the state space can be expressed by the following formula: t =[V PCC ,I PCC ,Z est ], mainly including the voltage and current of the common coupling point PCC in the power grid and the estimated power grid impedance value. Through this state space design, the system can quickly respond to changes in the power grid load and adjust the impedance in time to ensure the stability of the power grid;

[0026] The action space design of HDRL is one of the keys to this module. The action space is divided into two parts: discrete action and continuous action. The combination of the two can flexibly respond to various fluctuations in the power grid. Discrete action is used for a wide range of impedance adjustment, which is suitable for situations where the power grid load fluctuates violently or changes suddenly. Each discrete action represents a large adjustment step for impedance, which is usually used to quickly respond to significant fluctuations. Continuous action is used for fine adjustment to ensure that the power grid can still operate smoothly when the load changes slightly. Continuous action provides fine-grained control of impedance adjustment, avoids frequent and large adjustments, maintains the stability of the power grid, and avoids the negative impact caused by excessive adjustments.

[0027] The action space can be defined by the following formula: α(t) = {α dis (t),α con (t)};

[0028] Among them, α dis (t) is a discrete action, α con (t) is a continuous action;

[0029] In order to enable the HDRL model to learn the optimal impedance adjustment strategy, a reward function is designed. The goal of the reward function is to minimize the voltage and current fluctuations of the power grid and improve the power quality and grid stability. The specific formula is as follows:

[0030] R(s,a)=-α|V target -V PCC |-β|I target -I PCC |;

[0031] Among them, α and β are the stability weight coefficients of voltage and current respectively, which can be adjusted according to actual needs to ensure the power quality and grid stability. The learning process of HDRL adopts a hierarchical neural network structure, including a strategy network and a value network. The strategy network is responsible for selecting appropriate discrete actions from the current state space, which is suitable for coping with a wide range of load changes. The value network further refines the discrete actions selected by the strategy network and optimizes the accuracy of impedance adjustment. This refinement process enables the system to make fine adjustments through continuous actions when there are small load changes to ensure the smooth operation of the grid.

[0032] After each adjustment, the system calculates the reward and updates the network parameters according to the reward signal to optimize the impedance adjustment strategy. Through the reward feedback mechanism, the HDRL model continuously learns and adjusts its strategy, so that the power grid can maintain efficient and stable operation in a dynamic load environment.

[0033] During the training process, the Q-learning algorithm of reinforcement learning is used to update the parameters. The update formula is as follows:

[0034] Q(s t ,α t )=Q(s t ,α t )+α[R(t)+γmaxQ(s t+1 ,α')-Q(s t ,α t )];

[0035] Among them, Q(s t ,α t ) is state s t Next, take action α t Q value, α is the learning rate, γ is the discount factor, maxQ(s t+1 ,α') is the maximum Q value in the next state. The learning algorithm enables the HDRL model to gradually learn the optimal impedance adjustment strategy, so as to adaptively adjust the grid impedance in practical applications.

[0036] According to the above technical solution, the real-time feedback regulation module is responsible for fine-tuning the dynamic correction of impedance regulation during the operation of the power grid to cope with the rapidly changing load environment and complex power grid fluctuations. It adopts a control strategy that combines the classic PID controller with the compensation term based on hybrid deep reinforcement learning HDRL, combining the rapid response capability of the traditional control method with the optimization capability of the intelligent control method, significantly improving the real-time, robustness and adaptability of the system.

[0037] The real-time feedback regulation module consists of two key parts: the PID controller and the HDRL compensation term. The PID controller is used to preliminarily adjust the grid impedance, respond to rapid load fluctuations and perform basic dynamic compensation. The PID controller ensures that the system can quickly stabilize near the target value through simple and efficient proportional, integral and differential control.

[0038] The HDRL compensation term is based on the PID controller. The HDRL compensation term provides the system with more accurate adjustment capabilities by learning complex nonlinear dynamic characteristics. The compensation term is generated in real time based on the HDRL algorithm and can correct the part that the PID controller cannot fully compensate. The combination of the two is achieved through the following control formula:

[0039] u(t)=u PID (t)+u HDRL (t);

[0040] Among them, u(t) is the total output control signal, u PID (t) is the output of PID controller, u HDRL (t) is the output of the HDRL compensation term. This combined control strategy not only utilizes the high response speed of PID control, but also gives full play to the adaptive learning ability of HDRL in complex scenarios, achieving fast and accurate impedance adjustment.

[0041] in, e(t)=Z ref (t)-Z est (t), is the error between the target impedance and the estimated impedance, where K p , K i and K d are proportional, integral and derivative gains respectively;

[0042] The design of HDRL compensation term is based on hybrid deep reinforcement learning algorithm, which generates compensation signal u in real time through policy network and value network. HDRL (t), the design goal of the compensation term is to eliminate the complex nonlinear characteristics that the PID controller cannot handle and the insufficient regulation caused by the dynamic changes of the system.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. Through the dynamic threshold adjustment mechanism, the voltage and current fluctuations at the common coupling point are monitored in real time to reduce unnecessary computing load. The sliding window algorithm combines the statistical characteristics of voltage and current to dynamically set the trigger conditions to improve the accuracy and real-time performance of event triggering. The combination of discrete and continuous action spaces is adopted, and the strategy network is used for large-scale adjustments. The value network is fine-tuned to ensure the accuracy of grid impedance regulation. The reward function combines the voltage fluctuations, current fluctuations and trigger frequency of the grid to optimize system stability and efficiency. The PID controller is introduced for initial response, and the HDRL module provides dynamic optimization compensation signals to achieve intelligent control of complex load fluctuations. The dual feedback mechanism enhances the robustness and real-time response capability of the system.

[0045] 2. Start the HDRL module through the event trigger mechanism to realize the combination of efficient triggering and intelligent impedance optimization. The linkage mechanism significantly reduces the consumption of computing resources, while improving the response speed and adjustment accuracy of the system. The collaborative design of the event trigger mechanism and the HDRL optimization module, the combination of adaptive dynamic threshold adjustment and sliding window algorithm, the feedback adjustment system composed of PID controller and HDRL module, the grid impedance estimation method combining recursive least squares method with Kalman filtering, the application of hybrid deep reinforcement learning algorithm in impedance optimization, especially the combination of discrete and continuous action space and hierarchical network structure, dynamic threshold adjustment algorithm and its implementation in the event trigger mechanism.

[0046] 3. Through the adaptive event trigger mechanism, the trigger conditions can be dynamically adjusted according to the real-time fluctuations of the power grid, and impedance estimation and adjustment are performed only when necessary, thereby reducing unnecessary computational interference and ensuring the stability of power grid operation. The hybrid deep reinforcement learning module can quickly respond to drastic fluctuations in load and avoid power grid instability caused by impedance imbalance;

[0047] And through the combination of sliding window algorithm and dynamic threshold adjustment mechanism, the fluctuation characteristics of voltage and current are monitored in real time to ensure the accuracy and timeliness of event triggering. The hierarchical action space design of HDRL algorithm realizes the combination of fast and fine adjustment, which significantly improves the response efficiency of the system. Through the adaptive event triggering mechanism, the impedance estimation and optimization process is started only when there is a significant fluctuation, avoiding frequent resource calls and greatly reducing the computational burden. The combination of recursive least squares method and Kalman filtering further optimizes the impedance estimation algorithm, so that the system achieves a balance between computational accuracy and efficiency.

[0048] 4. By adopting the HDRL algorithm and combining the division of labor between the strategy network and the value network, the system can autonomously learn the optimal impedance adjustment strategy according to the actual load environment, reducing the need for human intervention. The system continuously optimizes the adjustment strategy through the reward mechanism, and can realize intelligent impedance adjustment in complex dynamic load environments. The impedance optimization module can significantly reduce voltage and current fluctuations, reduce harmonic distortion and power loss, and improve power transmission efficiency. The joint compensation strategy of the PID controller and the HDRL module can quickly respond to load changes while maintaining output stability and ensuring power quality.

[0049] In summary, the real-time impedance adjustment of the power grid under complex load conditions is realized through intelligent control methods to effectively improve the power quality. By introducing the event trigger mechanism, the system can adaptively estimate the impedance when significant load fluctuations are detected, reducing unnecessary calculations. The HDRL algorithm further optimizes the adjustment strategy to cope with frequent load changes, improve response speed and adaptability, ensure the stability of voltage and current, and overcome the shortcomings of existing technologies in real-time response, load adaptability and power quality assurance.

[0050] In addition, in response to the problems of power quality fluctuations and grid impedance changes under dynamic conditions such as distributed energy access and complex load fluctuations, a full-link impedance optimization solution from "adaptive perception" to "intelligent control" was proposed. Through the joint design of adaptive event triggering mechanism, HDRL optimization strategy and feedback control mechanism, real-time optimization and intelligent control of grid impedance are achieved, thereby improving the stability and power quality of the grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0052] In the attached picture:

[0053] Figure 1 It is a schematic diagram of the working process of the power grid impedance adjustment system of the present invention. DETAILED DESCRIPTION

[0054] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0055] Example: Figure 1As shown, the present invention provides a technical solution, a grid impedance regulation system based on adaptive event triggering and hybrid deep learning, impedance estimation based on adaptive event-driven, intelligent control optimization and dynamic feedback regulation, so as to achieve real-time adjustment of grid impedance, thereby reducing harmonic distortion, alleviating voltage fluctuations and improving power quality;

[0056] Specifically, it includes an impedance estimation module triggered by an adaptive event, an impedance optimization module based on HDRL, a real-time feedback adjustment module, and an integration mechanism between adaptive event triggering and HDRL;

[0057] The adaptive event-triggered impedance estimation module, HDRL-based impedance optimization module, and real-time feedback regulation module realize linkage control through data interaction and signal feedback. When the fluctuation of PCC voltage or current exceeds the dynamic threshold, the adaptive event triggering module activates the impedance estimation and uses the estimation result as one of the input states of HDRL. HDRL determines the optimal impedance control action according to the current PCC state and transmits it to the feedback regulation module. The PCC state includes voltage, current, and estimated impedance.

[0058] The real-time feedback regulation module further adjusts and optimizes the grid impedance through a weighted combination of PID control signals and HDRL control signals. This full-link linkage control structure can ensure the flexibility and stability of the control strategy in scenarios with drastic load fluctuations, reduce the conduction effect of load disturbances, and significantly improve power quality and grid stability.

[0059] Based on the above technical solution, the adaptive event-triggered impedance estimation module dynamically monitors the voltage and current fluctuations of the common coupling point PCC, and adopts a dynamic threshold adjustment mechanism and a recursive least squares method RLS combined with a Kalman filter estimation method to achieve real-time estimation of the grid impedance. Unlike the traditional fixed threshold trigger method, the event trigger mechanism can dynamically adjust the trigger conditions according to the load fluctuation of the PCC point, so that the impedance estimation can be triggered immediately when the load changes greatly, and unnecessary triggering can be reduced when the load is stable, effectively reducing the computational burden and energy consumption. In addition, by combining the adaptive filter with the dynamic threshold control, the robustness to noise and high-frequency disturbances is further improved. The combination of RLS and Kalman filtering ensures high-frequency dynamic response. In the case of load mutation, it can quickly converge to a new impedance value, thereby improving the stability and accuracy of the estimation.

[0060] Based on the above technical solution, the HDRL-based impedance optimization module is based on the hybrid deep reinforcement learning HDRL algorithm, which divides the impedance control into a two-layer control structure of discrete action and continuous action to cope with the complex load fluctuations and impedance changes in the power grid. Specifically, the discrete action space is responsible for drastically adjusting the impedance to quickly respond to sudden load fluctuations, and the continuous action space is responsible for fine-tuning the impedance to ensure the stability of the power grid operation. The HDRL algorithm is based on the architecture of the policy network Actor and the value network Critic. It continuously adjusts the optimization strategy through the reward and punishment mechanism to ensure efficient response in scenarios of large load fluctuations and small load adjustments. Compared with traditional PID control and support vector machine SVM methods, HDRL has the ability of self-learning and self-adaptation. Even in complex power grid fluctuation scenarios, it can generate the optimal impedance adjustment strategy in real time through the policy network, thereby greatly improving the system's adaptability and response speed.

[0061] Based on the above technical solution, the real-time feedback regulation module is responsible for quickly correcting and adjusting the voltage and current of the power grid under the synergistic effect of the HDRL control signal and the PID control signal. The function of the PID controller is to handle small load fluctuations, while the HDRL compensation module is responsible for quickly correcting the control output of the system when a large load change occurs in the power grid to prevent the voltage or current from exceeding the threshold range. The PID controller uses the voltage and current deviations of the PCC point to calculate the control increment. The compensation term generated by HDRL will be used as a supplementary signal to the PID controller based on the optimal control action generated by the HDRL strategy network, so that the system can respond quickly to sudden load changes. This fusion design of dual control signals significantly improves the response speed and stability of the system, especially in scenarios with frequent load fluctuations.

[0062] Based on the above technical solution, the adaptive event-triggered impedance estimation module is responsible for real-time monitoring of the voltage and current fluctuations at the common coupling point PCC of the power grid, and automatically triggers the impedance estimation process according to these fluctuations. The core goal of this module is to reduce the computational burden, respond quickly when the light load fluctuations of the power grid are significant, and improve the stability and power quality of the power grid. It adopts a dynamic threshold adjustment mechanism instead of the traditional fixed threshold mechanism. The event-triggered threshold can be dynamically adjusted according to the real-time voltage and current fluctuations, thereby avoiding excessive calculation and ensuring that the impedance estimation process is accurately triggered when needed.

[0063] Based on the above technical solution, the dynamic threshold adjustment mechanism is based on the voltage and current fluctuation amplitude, and uses an adaptive filter to effectively separate the grid signal and noise, and automatically identify the effective grid disturbance signal. When the grid fluctuation is small or the background noise is large, the impedance estimation will not be triggered frequently, which reduces the calculation load of the system and improves the response efficiency of the system. Only when the grid fluctuation reaches a certain level will the impedance estimation process be triggered to ensure the stability of the system.

[0064] In order to further improve the effect of dynamic threshold adjustment, a sliding window algorithm is proposed. This algorithm dynamically adjusts the threshold by calculating the maximum and minimum values ​​of the voltage within the set time window, and updates the threshold base at each time node. It can effectively cope with common random fluctuations in the power grid and has a high anti-interference ability to light load fluctuations.

[0065] Specifically, suppose that in the time window T win The maximum and minimum values ​​of the voltage signal V(t) are V max (t) and V min (t), the sliding window algorithm can calculate the current fluctuation base value V by the following formula base (t): Based on this fluctuating base value, the system can dynamically adjust the threshold for event triggering to ensure that impedance estimation can be triggered in a timely manner when the power grid changes significantly;

[0066] When the event triggering conditions are met, the system will start the impedance estimation process. In order to achieve accurate real-time estimation, the recursive least squares RLS and Kalman filter Kalman Filter algorithms are combined. The RLS algorithm can quickly update the impedance estimation value, and the Kalman filter can effectively suppress the noise in the power grid environment and improve the accuracy of the estimation result. Specifically, the mathematical formula of the impedance estimation process is as follows:

[0067] Z est (t) = Z est (t-1)+K(t)[V(t)-I(t)Z est (t-1)];

[0068] Among them, Z est is the impedance estimation value at the current time t, V(t) is the current voltage value, I(t) is the current current value, K(t) is the adaptive gain, which is dynamically adjusted based on the real-time voltage, current and impedance inverse prediction value, Z est (t-1) is the impedance estimate at the previous moment, and the gain K(t) is calculated based on the current voltage V(t), current I(t) and impedance Z est The prediction error of (t-1) is dynamically calculated, and the adaptive gain can optimize the estimation result according to the real-time fluctuation and background noise;

[0069] The introduction of the Kalman filter enables the impedance estimation process to maintain a high estimation accuracy in a noisy environment. The specific Kalman gain update formula is as follows:

[0070] K(t)=P(t-1)[P(t-1)+R(t)] -1

[0071] P(t)=[P(t-1) -1 +Q(t)] -1 ;

[0072] Among them, P(t) is the estimation error covariance matrix, Q(t) is the process noise covariance, and R(t) is the measurement noise covariance. Through the combination of RLS and Kalman filter, the system can accurately estimate the grid impedance in a noisy environment and has a high real-time response capability.

[0073] Based on the above technical solution, the HDRL-based impedance optimization module is the core intelligent control unit, which is designed to adjust the grid impedance in real time, thereby optimizing the power quality and improving the stability of the grid. The module adopts a hybrid deep reinforcement learning algorithm, combining discrete and continuous action spaces, to ensure that the grid can quickly and accurately optimize impedance adjustment when the load fluctuates and the grid state changes;

[0074] The HDRL algorithm selects discrete actions through the policy network. Discrete actions refer to large-scale impedance adjustments, and refines and optimizes these discrete actions through the value network to achieve the optimal effect of impedance regulation. It can adapt to frequent load changes in the power grid, avoid frequent computing operations, and improve real-time response capabilities and overall stability.

[0075] Based on the above technical solution, the HDRL-based impedance optimization module needs to design the HDRL state space first. The design of the state space involves the system's real-time perception and feedback mechanism of the power grid. The input vector of the system mainly includes the voltage, current and estimated power grid impedance value of the common coupling point PCC in the power grid. Through the voltage and current data collected by real-time sensors, the system can monitor the dynamic changes of the power grid in real time and adjust the power grid impedance according to the changes. The design of the state space can be expressed by the following formula: t =[V PCC ,I PCC ,Z est ], mainly including the voltage and current of the common coupling point PCC in the power grid and the estimated power grid impedance value. Through this state space design, the system can quickly respond to changes in the power grid load and adjust the impedance in time to ensure the stability of the power grid;

[0076] The action space design of HDRL is one of the keys of this module. The action space is divided into two parts: discrete action and continuous action. The combination of the two can flexibly respond to various fluctuations in the power grid. Discrete action is used for a large range of impedance adjustment, which is suitable for situations where the power grid load fluctuates violently or changes suddenly. Each discrete action represents a large adjustment step for impedance, which is usually used to quickly respond to significant fluctuations. If the power grid load increases suddenly, the discrete action will make a large adjustment to the impedance to restore the power grid state to a stable level as soon as possible. Continuous action is used for fine adjustment to ensure that the power grid can still operate smoothly when the load changes slightly. Continuous action provides fine-grained control of impedance adjustment, avoids frequent and large adjustments, maintains the stability of the power grid, and avoids the negative impact caused by excessive adjustments.

[0077] The action space can be defined by the following formula: α(t) = {α dis (t),α con (t)};

[0078] Among them, α dis (t) is a discrete action, which means a large adjustment, α con (t) is a continuous action, which means fine adjustment;

[0079] In order to enable the HDRL model to learn the optimal impedance adjustment strategy, a reward function is designed. The goal of the reward function is to minimize the voltage and current fluctuations of the power grid and improve the power quality and grid stability. The specific formula is as follows:

[0080] R(s,a)=-α|V target -V PCC |-β|I target -I PCC |;

[0081] Among them, α and β are the stability weight coefficients of voltage and current respectively, which can be adjusted according to actual needs to ensure the power quality and grid stability. The learning process of HDRL adopts a hierarchical neural network structure, including a strategy network and a value network. The strategy network is responsible for selecting appropriate discrete actions from the current state space, that is, making a large adjustment to the impedance, which is suitable for coping with a large range of load changes. The value network further refines the discrete actions selected by the strategy network and optimizes the accuracy of impedance adjustment. This refinement enables the system to make fine adjustments through continuous actions when there are small load changes to ensure the smooth operation of the power grid;

[0082] After each adjustment, the system calculates the reward and updates the network parameters according to the reward signal to optimize the impedance adjustment strategy. Through the reward feedback mechanism, the HDRL model continuously learns and adjusts its strategy, so that the power grid can maintain efficient and stable operation in a dynamic load environment.

[0083] During the training process, the Q-learning algorithm of reinforcement learning is used to update the parameters. The update formula is as follows:

[0084] Q(s t ,α t )=Q(s t ,α t )+α[R(t)+γmaxQ(s t+1 ,α')-Q(s t ,α t )];

[0085] Among them, Q(s t ,α t ) is state s t Next, take action α t Q value, α is the learning rate, γ is the discount factor, maxQ(s t+1 ,α') is the maximum Q value in the next state. The learning algorithm enables the HDRL model to gradually learn the optimal impedance adjustment strategy, so as to adaptively adjust the grid impedance in practical applications;

[0086] The HDRL-based impedance optimization module has the following advantages: Fast response and efficient regulation: By combining discrete and continuous action spaces, HDRL can quickly make large-scale adjustments when the load fluctuates violently, and make fine adjustments when the load changes slightly, thus ensuring the smooth operation of the power grid;

[0087] Intelligent learning capability: HDRL continuously optimizes impedance adjustment strategies through a reward mechanism, and can adaptively adjust impedance in complex power grid environments to improve grid stability and power quality;

[0088] Reduce computational burden: Through reasonable action space design and reward function optimization, HDRL avoids excessive computational operations, reduces the computational burden of the system, and improves the real-time response capability of the system.

[0089] Based on the above technical solution, the real-time feedback regulation module is responsible for making fine dynamic corrections to the impedance regulation during the operation of the power grid to cope with the rapidly changing load environment and complex power grid fluctuations. It adopts a control strategy that combines the classic PID controller with the compensation term based on hybrid deep reinforcement learning HDRL, combining the fast response capability of the traditional control method with the optimization capability of the intelligent control method, significantly improving the real-time, robustness and adaptability of the system.

[0090] The real-time feedback regulation module consists of two key parts: the PID controller and the HDRL compensation term. The PID controller is used to preliminarily adjust the grid impedance, respond to rapid load fluctuations and perform basic dynamic compensation. The PID controller ensures that the system can quickly stabilize near the target value through simple and efficient proportional, integral and differential control.

[0091] The HDRL compensation term is based on the PID controller. The HDRL compensation term provides the system with more accurate adjustment capabilities by learning complex nonlinear dynamic characteristics. The compensation term is generated in real time based on the HDRL algorithm and can correct the part that the PID controller cannot fully compensate. The combination of the two is achieved through the following control formula:

[0092] u(t)=u PID (t)+u HDRL (t);

[0093] Among them, u(t) is the total output control signal, u PID (t) is the output of PID controller, u HDRL (t) is the output of the HDRL compensation term. This combined control strategy not only utilizes the high response speed of PID control, but also gives full play to the adaptive learning ability of HDRL in complex scenarios, achieving fast and accurate impedance adjustment.

[0094] in, e(t)=Z ref (t)-Z est (t), is the error between the target impedance and the estimated impedance, where K p , K i and K d are proportional, integral and derivative gains respectively;

[0095] The design of HDRL compensation term is based on hybrid deep reinforcement learning algorithm, which generates compensation signal u in real time through policy network and value network. HDRL (t), the design goal of the compensation term is to eliminate the complex nonlinear characteristics that the PID controller cannot handle and the insufficient regulation caused by the dynamic changes of the system.

[0096] Finally, it should be noted that the above description is only a preferred example of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A grid impedance regulation system based on adaptive event triggering and hybrid deep learning, characterized by: Based on adaptive event-driven impedance estimation, intelligent control optimization and dynamic feedback regulation, it can achieve real-time adjustment of grid impedance, reduce harmonic distortion, alleviate voltage fluctuations and improve power quality; Specifically, it includes an impedance estimation module triggered by an adaptive event, an impedance optimization module based on HDRL, a real-time feedback adjustment module, and an integration mechanism between adaptive event triggering and HDRL; The adaptive event-triggered impedance estimation module, HDRL-based impedance optimization module, and real-time feedback regulation module realize linkage control through data interaction and signal feedback. When the fluctuation of PCC voltage or current exceeds the dynamic threshold, the adaptive event triggering module activates impedance estimation and uses the estimation result as one of the input states of HDRL. HDRL determines the optimal impedance control action according to the current PCC state and transmits it to the feedback regulation module. The real-time feedback regulation module further adjusts and optimizes the grid impedance through a weighted combination of PID control signals and HDRL control signals, forming a full-link linkage control structure.

2. The grid impedance adjustment system based on adaptive event triggering and hybrid deep learning according to claim 1, characterized in that: The adaptive event-triggered impedance estimation module dynamically monitors the voltage and current fluctuations of the common coupling point PCC, and adopts a dynamic threshold adjustment mechanism and a recursive least squares method RLS combined with a Kalman filter estimation method to estimate the grid impedance in real time.

3. The grid impedance adjustment system based on adaptive event triggering and hybrid deep learning according to claim 1, characterized in that: The HDRL-based impedance optimization module divides impedance control into a two-layer control structure of discrete actions and continuous actions based on a hybrid deep reinforcement learning HDRL algorithm.

4. The grid impedance adjustment system based on adaptive event triggering and hybrid deep learning according to claim 1, characterized in that: The real-time feedback regulation module is responsible for quickly correcting and adjusting the voltage and current of the power grid under the coordinated action of the HDRL control signal and the PID control signal. The function of the PID controller is to handle small load fluctuations, while the HDRL compensation module is responsible for quickly correcting the control output of the system when a large load change occurs in the power grid.

5. The grid impedance regulation system based on adaptive event triggering and hybrid deep learning according to claim 2, characterized in that: The adaptive event-triggered impedance estimation module is responsible for real-time monitoring of voltage and current fluctuations at the common coupling point PCC of the power grid, and automatically triggers the impedance estimation process according to these fluctuations. The core goal of this module is to reduce the computational burden and respond quickly when the light load fluctuations of the power grid are significant. It adopts a dynamic threshold adjustment mechanism instead of the traditional fixed threshold mechanism, and can dynamically adjust the event-triggered threshold according to the real-time voltage and current fluctuations.

6. The grid impedance regulation system based on adaptive event triggering and hybrid deep learning according to claim 5, characterized in that: The dynamic threshold adjustment mechanism is based on the voltage and current fluctuation amplitude, and uses an adaptive filter to effectively separate the grid signal and noise, automatically identify the effective grid disturbance signal, and reduce the system's computing load when the grid fluctuation is small or the background noise is large; In order to further improve the effect of dynamic threshold adjustment, a sliding window algorithm is proposed. This algorithm dynamically adjusts the threshold by calculating the maximum and minimum values ​​of the voltage within the set time window, and updates the threshold base at each time node. It can effectively cope with common random fluctuations in the power grid and has a high anti-interference ability to light load fluctuations. Specifically, suppose that in the time window T win The maximum and minimum values ​​of the voltage signal V(t) are V max (t) and V min (t), the sliding window algorithm can calculate the current fluctuation base value V by the following formula base (t): Based on this fluctuating base value, the system can dynamically adjust the threshold for event triggering to ensure that impedance estimation can be triggered in a timely manner when the power grid changes significantly; When the event triggering conditions are met, the system will start the impedance estimation process. In order to achieve accurate real-time estimation, the recursive least squares RLS and Kalman filter algorithms are combined. The RLS algorithm can quickly update the impedance estimation value, and the Kalman filter can effectively suppress the noise in the power grid environment and improve the accuracy of the estimation result. Specifically, the mathematical formula of the impedance estimation process is as follows: Z est (t)=Z est (t-1)+K(t)[V(t)-I(t)Z est (t-1)]; Among them, Z est is the impedance estimation value at the current time t, V(t) is the current voltage value, I(t) is the current current value, K(t) is the adaptive gain, which is dynamically adjusted based on the real-time voltage, current and impedance inverse prediction value, Z est (t-1) is the impedance estimate at the previous moment, and the gain K(t) is calculated based on the current voltage V(t), current I(t) and impedance Z est The prediction error of (t-1) is dynamically calculated, and the adaptive gain can optimize the estimation result according to the real-time fluctuation and background noise; The introduction of the Kalman filter enables the impedance estimation process to maintain a high estimation accuracy in a noisy environment. The specific Kalman gain update formula is as follows: K(t)=P(t-1)[P(t-1)+R(t)] -1 P(t)=[P(t-1) -1 +Q(t)] -1 ; Among them, P(t) is the estimation error covariance matrix, Q(t) is the process noise covariance, and R(t) is the measurement noise covariance. By combining RLS with the Kalman filter, the system can accurately estimate the grid impedance in a noisy environment.

7. The grid impedance regulation system based on adaptive event triggering and hybrid deep learning according to claim 3, characterized in that: The HDRL-based impedance optimization module is the core intelligent control unit, which is designed to adjust the grid impedance in real time, thereby optimizing the power quality and improving the stability of the grid. The module adopts a hybrid deep reinforcement learning algorithm that combines discrete and continuous action spaces; The HDRL algorithm selects discrete actions through a policy network and refines and optimizes these discrete actions through a value network.

8. The grid impedance adjustment system based on adaptive event triggering and hybrid deep learning according to claim 7, characterized in that: The HDRL-based impedance optimization module needs to design the HDRL state space first. The design of the state space involves the system's real-time perception and feedback mechanism of the power grid. The input vector of the system mainly includes the voltage, current and estimated power grid impedance value of the common coupling point PCC in the power grid. Through the voltage and current data collected by real-time sensors, the system can monitor the dynamic changes of the power grid in real time and adjust the power grid impedance according to the changes. The design of the state space can be expressed by the following formula: t =[V PCC ,I PCC ,Z est ], mainly including the voltage and current of the common coupling point PCC in the power grid and the estimated grid impedance value, through this state space design; The action space design of HDRL is one of the keys to this module. The action space is divided into two parts: discrete action and continuous action. The combination of the two can flexibly respond to various fluctuations in the power grid. Discrete action is used for a wide range of impedance adjustment, which is suitable for situations where the power grid load fluctuates violently or changes suddenly. Each discrete action represents a large adjustment step for impedance, which is usually used to quickly respond to significant fluctuations. Continuous action is used for fine adjustment to ensure that the power grid can still operate smoothly when the load changes slightly. Continuous action provides fine-grained control of impedance adjustment, avoids frequent and large adjustments, maintains the stability of the power grid, and avoids the negative impact caused by excessive adjustments. The action space can be defined by the following formula: α(t) = {α dis (t),α con (t)}; Among them, α dis (t) is a discrete action, α con (t) is a continuous action; In order to enable the HDRL model to learn the optimal impedance adjustment strategy, a reward function is designed. The goal of the reward function is to minimize the voltage and current fluctuations of the power grid and improve the power quality and grid stability. The specific formula is as follows: R(s,a)=-αV target -V PCC |-βI target -I PCC |; Among them, α and β are the stability weight coefficients of voltage and current respectively, which can be adjusted according to actual needs to ensure the power quality and grid stability. The learning process of HDRL adopts a hierarchical neural network structure, including a strategy network and a value network. The strategy network is responsible for selecting appropriate discrete actions from the current state space, which is suitable for coping with a wide range of load changes. The value network further refines the discrete actions selected by the strategy network and optimizes the accuracy of impedance adjustment. This refinement process enables the system to make fine adjustments through continuous actions when there are small load changes. After each adjustment, the system calculates the reward and updates the network parameters based on the reward signal to optimize the impedance adjustment strategy. Through the reward feedback mechanism, the HDRL model continuously learns and adjusts its strategy. During the training process, the Q-learning algorithm of reinforcement learning is used to update the parameters. The update formula is as follows: Q(s t ,α t )=Q(s t ,α t )+α[R(t)+γmaxQ(s t+1 ,α')-Q(s t ,α t )]; Among them, Q(s t ,α t ) is state s t Next, take action α t Q value, α is the learning rate, γ is the discount factor, maxQ(s t+1 ,α') is the maximum Q value in the next state. The learning algorithm enables the HDRL model to gradually learn the optimal impedance adjustment strategy, so as to adaptively adjust the grid impedance in practical applications.

9. The grid impedance adjustment system based on adaptive event triggering and hybrid deep learning according to claim 4, characterized in that: The real-time feedback regulation module is responsible for fine-tuning the impedance regulation dynamically during the operation of the power grid to cope with the rapidly changing load environment and complex power grid fluctuations. It adopts a control strategy that combines the classic PID controller with the compensation term based on hybrid deep reinforcement learning HDRL, combining the fast response capability of the traditional control method with the optimization capability of the intelligent control method. The real-time feedback regulation module consists of two key parts: the PID controller and the HDRL compensation term. The PID controller is used to preliminarily adjust the grid impedance, respond to rapid load fluctuations and perform basic dynamic compensation. The PID controller ensures that the system can quickly stabilize near the target value through simple and efficient proportional, integral and differential control. The HDRL compensation term is based on the PID controller. The HDRL compensation term provides the system with more accurate adjustment capabilities by learning complex nonlinear dynamic characteristics. The compensation term is generated in real time based on the HDRL algorithm and can correct the part that the PID controller cannot fully compensate. The combination of the two is achieved through the following control formula: u(t)=u PID (t)+u HDRL (t); Among them, u(t) is the total output control signal, u PID (t) is the output of PID controller, u HDRL (t) is the output of the HDRL compensation term. This combined control strategy not only utilizes the high response speed of PID control, but also gives full play to the adaptive learning ability of HDRL in complex scenarios, achieving fast and accurate impedance adjustment. in, e(t)=Z ref (t)-Z est (t), is the error between the target impedance and the estimated impedance, where K p , K i and K d are proportional, integral and derivative gains respectively; The design of HDRL compensation term is based on hybrid deep reinforcement learning algorithm, which generates compensation signal u in real time through policy network and value network. HDRL (t), the design goal of the compensation term is to eliminate the complex nonlinear characteristics that the PID controller cannot handle and the insufficient regulation caused by the dynamic changes of the system.

Images