A multi-source cooperative harmonic adaptive suppression method for island microgrid

Abstract

The application discloses a kind of island microgrid-oriented multi-source collaborative harmonic adaptive suppression method, it is related to island microgrid harmonic suppression technical field, comprising the following steps: extracting the fundamental component of island microgrid local inverter output voltage signal, obtain the real-time fundamental angular frequency perceived;Based on real-time fundamental angular frequency, separate harmonic component is realized through dynamic reconstruction signal delay chain, and the physical phase alignment of compensation vector and grid distortion source is carried out through feedforward correction logic, to prevent system shock induced by insufficient phase margin;Through the consistency protocol of information layer, and coupling aperiodic event trigger mechanism, after dynamic on-demand allocation of channel resources, closed-loop regulation is carried out to local harmonic compensation current.The application significantly reduces the throughput pressure of distributed channel and the invalid task load of processor, greatly improves the engineering reliability and system expansibility of distributed governance architecture in resource-poor environment.

Description

Technical Field

[0001] This invention relates to the field of harmonic suppression technology for islanded microgrids, and more specifically, to a multi-source cooperative harmonic adaptive suppression method for islanded microgrids. Background Technology

[0002] Isolated microgrids play a vital role in remote areas and emergency power supply. However, their operation, independent of the main power grid, causes significant fluctuations in system frequency and voltage with changes in renewable energy output and load, leading to prominent power quality issues, particularly harmonic pollution caused by nonlinear loads. Harmonics can cause electrical equipment malfunctions, reduced energy efficiency, and system resonance risks. Therefore, effective suppression of harmonics is crucial to ensuring the safe and stable operation of isolated microgrids.

[0003] In the field of harmonic mitigation, the existing technologies most similar to the technical concept of this invention can be mainly divided into the following three categories, each of which has obvious limitations in the application of isolated microgrids: 1. Harmonic Detection Technology Based on Frequency Adaptive Phase-Locked Loops: This type of technology (such as the method based on dual second-order generalized integrator frequency-locked loops DSOGI-FLL) tracks grid frequency changes through adaptive loops, improving the performance of fixed-parameter detection methods under frequency fluctuations. However, in islanded microgrids, its core drawback lies in the tight coupling between the gain of the frequency loop and the input voltage amplitude. When the grid experiences voltage dips or swells due to load switching or faults (amplitude fluctuations can reach 0.2-1.2 pu), the sensitivity and dynamic response of its frequency tracking change significantly, leading to frequency reference inaccuracy under severe voltage fluctuations, and consequently causing fundamental errors in all subsequent harmonic extraction stages that rely on this frequency.

[0004] 2. Harmonic Extraction and Compensation Techniques Based on Discrete Delay Signal Cancellation: This type of technique (such as Delay Signal Cancellation (DSC) and its improved algorithms) separates harmonics by constructing a signal delay chain, resulting in a simple structure. Its fundamental problems lie in two aspects: First, the delay length is usually calculated based on a fixed rated frequency. When the frequency of an isolated microgrid continuously drifts, this causes "defocusing" in harmonic extraction, introducing significant detection errors and phase lag. Second, the inherent group delay of the algorithm and the computational delay of the digital control system are not effectively compensated, leading to a non-negligible phase difference between the final compensation command and the real-time harmonic distortion. In severe cases, this not only weakens the suppression effect but may also trigger system oscillations.

[0005] 3. Distributed Cooperative Suppression Technology Based on Periodic Communication: To achieve coordination among multiple distributed management devices (such as grid-connected inverters), existing solutions generally adopt centralized or distributed architectures, relying on periodic, high-frequency communication between nodes to exchange status information and achieve synchronous control. The limitations of this approach stem directly from the physical reality of isolated microgrids: the system deployment environment often has weak communication infrastructure, limited network bandwidth, high latency, and low reliability. High-frequency periodic communication easily causes network congestion, leading to performance degradation or even instability in cooperative control due to data loss or delay, failing to meet real-time requirements. Simultaneously, most existing cooperative mechanisms do not fully consider branch impedance differences and lack effective online current sharing strategies, resulting in uneven output from each management unit and generating harmful harmonic circulating currents.

[0006] In summary, while the closest existing technologies have made improvements in certain aspects, none have systematically solved the three core challenges faced by harmonic suppression in islanded microgrids: the adaptability of the detection stage under large-scale frequency and voltage fluctuations, the accurate compensation for phase lag in the digital control system, and the reliability and balance of multi-machine collaboration under low-bandwidth, high-latency communication conditions. These shortcomings limit the effectiveness and reliability of existing technologies in real-world islanded scenarios. Summary of the Invention

[0007] To address the complex challenges in harmonic suppression in isolated microgrids, including detection inaccuracies caused by frequency and voltage fluctuations, compensation synchronization issues due to processing delays, collaborative burdens from periodic communication, and circulating current losses induced by parameter inconsistencies, this invention aims to provide a multi-source collaborative adaptive harmonic suppression method for isolated microgrids. This method simultaneously achieves robustness of the harmonic detection process to frequency and voltage fluctuations across all operating conditions, accurate self-correction of the compensation signal to the inherent delays in the detection and control links, on-demand intelligent scheduling of communication resources through distributed collaboration, and autonomous balancing of inter-machine compensation loads without relying on precise physical parameters.

[0008] To achieve the above technical objectives, this application provides a multi-source cooperative harmonic adaptive suppression method for isolated microgrids, comprising the following steps: Extract the fundamental component of the output voltage signal of the local inverter in the isolated microgrid and obtain the sensed real-time fundamental angular frequency. Based on the real-time fundamental angular frequency, the separation of harmonic components is achieved by dynamically reconstructing the signal delay chain, and the compensation vector is aligned with the physical phase of the power grid distortion source through feedforward correction logic to prevent system oscillations induced by insufficient phase margin. By using a consistency protocol at the information layer and coupling a non-periodic event triggering mechanism, channel resources are dynamically allocated on demand, and then the local harmonic compensation current is adjusted in a closed loop.

[0009] Preferably, when extracting the fundamental component, to address the DC offset and specific order harmonic interference that microgrid sensors are prone to generate in high-frequency switching environments, the signal is converted into components in the αβ stationary coordinate system using Clark transform. A dual second-order generalized integrator (DSOGI) is then used as a front-end state observer with adaptive notch characteristics to extract the fundamental component. The damping gain coefficient k of the DSOGI is set to... The fundamental component is extracted by using a combination of bandpass and lowpass filters.

[0010] Preferably, a gain normalization update law is introduced when acquiring the real-time fundamental angular frequency to control the real-time fundamental angular frequency. Adaptive evolution is performed according to the following nonlinear integral law: ; in, The value of the standardized frequency gain determines the dynamic response bandwidth of frequency sensing. This is the frequency deviation discrimination signal obtained by multiplying and subtracting orthogonal voltage vectors; and The orthogonal components of the DSOGI output; denominator terms is the square of the instantaneous modulus of the orthogonal component.

[0011] Preferably, when extracting harmonic components, a nonlinear mapping model is established between the number of discrete sampling points d of the real-time fundamental angular frequency; the symmetry of the fundamental frequency within half a period is used for cancellation; and multiple of the above units are cascaded to form a multi-order comb filter to achieve enhanced extraction of specific harmonics, output the original harmonic vector, and complete the separation and extraction of harmonic components.

[0012] Preferably, when performing dynamic on-demand allocation of channel resources, a sparse communication network is constructed based on the physical topology of the multi-machine parallel inverters, and a dynamic consensus algorithm is introduced to balance the governance burden of each node; by setting an adaptive event triggering mechanism, a triggering decision execution mechanism based on the Lyapunov stability criterion, and a minimum triggering interval guarantee mechanism based on the Lipschitz continuity condition, the dynamic on-demand allocation of channel resources is completed.

[0013] Preferably, when performing multi-machine harmonic current closed-loop equalization control, a PI equalization controller based on consensus deviation is set according to the obtained harmonic current consensus mean value. Using a PI equalizer, a harmonic current correction signal is generated based on the consensus tracking error obtained by comparing the locally measured harmonic current with the consensus mean value of the harmonic current. The harmonic current correction signal is transformed into a stationary coordinate system component through an inverse rotation transformation, and then vector-superimposed with the compensation vector to form the final PWM voltage command correction term.

[0014] Preferably, when acquiring the compensation vector, the total phase lag generated by the harmonic detection link is quantitatively analyzed. The delay source includes two parts: one is the inherent group delay caused by the discrete DSC operator structure, whose lag angle at the target harmonic frequency is expressed as... Secondly, there is the hardware computation delay introduced by the sampling, calculation, and PWM update mechanism in the digital signal processing stage. The phase offset generated by this part is expressed as: ; in, Let be the angular frequency of the target harmonic, and h be the harmonic order. ; Define the total compensation angle The sum of the two items above: ; In the formula: The total lead angle used for vector correction; n is the operator characteristic order; This refers to the fixed sampling step size of the underlying hardware.

[0015] For the original harmonic components and Construct a 2×2 rotation transformation matrix Achieve vector lead deflection; the corrected compensation vector expression is as follows: ; After this transformation, the compensation vector It was moved forward on the timeline. This offsets the delay in the detection and control link, enabling a near-ideal 180° counter-offset on the time axis against the actual harmonic distortion in the power grid.

[0016] Based on the same inventive concept, this invention discloses a multi-source cooperative harmonic adaptive suppression system for isolated microgrids, comprising: The fundamental component extraction module is used to extract the fundamental component of the output voltage signal of the local inverter in the islanded microgrid and obtain the sensed real-time fundamental angular frequency. The harmonic component extraction module is used to separate harmonic components based on the real-time fundamental angular frequency by dynamically reconstructing the signal delay chain, and to align the compensation vector with the physical phase of the power grid distortion source through feedforward correction logic to prevent system oscillations induced by insufficient phase margin. The adaptive suppression module is used to dynamically allocate channel resources on demand through the information layer's consistency protocol and coupled with an aperiodic event triggering mechanism, and then perform closed-loop adjustment of the local harmonic compensation current.

[0017] The present invention discloses the following technical effects: This invention, by introducing standardized gain tracking logic and a deep reconstruction mapping of the discrete domain buffer, effectively alleviates the detection defocusing problem caused by non-stationary frequency evolution during droop adjustment in islanded microgrids, significantly enhances the frequency robustness of the detection link, and reduces the numerical step noise common in discrete systems. Simultaneously, the introduction of a phase rotation compensation matrix mathematically offsets the inherent physical and computational delays of the detection link, improving the phase margin of the closed-loop control system and aiming to reduce the risk of self-oscillation during high-frequency compensation. More importantly, by deeply coupling a distributed event triggering mechanism with Lyapunov stability constraints with dynamic consistency closed-loop equalization logic, this invention not only achieves automatic proportional sharing of harmonic load rates among multi-machine parallel systems, effectively suppressing harmonic circulating currents caused by line impedance mismatch, but also, while ensuring governance accuracy, realizes a paradigm shift from "periodic indiscriminate consumption" of communication resources to "state-driven on-demand allocation," significantly reducing the throughput pressure of distributed channels and the ineffective task load on processors, and greatly improving the engineering reliability and system scalability of the distributed governance architecture in resource-scarce environments. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is the overall flowchart of the distributed multi-source cooperative harmonic suppression method described in this invention; Figure 2 This is the logic diagram of the adaptive discrete detection operator described in this invention; Figure 3 This is the distributed resource scheduling and closed-loop collaborative equilibrium logic diagram described in this invention; Figure 4 This is the topology diagram of the islanded microgrid harmonic collaborative suppression system described in this invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] like Figures 1-4 As shown, this invention provides a multi-source cooperative harmonic adaptive suppression method for isolated microgrids, the overall process of which is as follows: Figure 1 As shown, the main steps include: S1: Frequency adaptive sensing based on a standardized gain frequency-locked loop: To address the frequency dynamic drift characteristics of isolated microgrids caused by droop control, this step aims to construct a frequency sensing chain with high amplitude decoupling capability and DC offset suppression characteristics. The specific implementation process is as follows: First, the output voltage signal of the local inverter in the isolated microgrid is sampled in real time. To address the DC offset and specific-order (e.g., 5th and 7th) harmonic interference that easily occur in microgrid sensors under high-frequency switching environments, this step converts the signal into a Clark transform. The components in the stationary coordinate system are used, and a dual second-order generalized integrator (DSOGI) is employed as a front-end state observer with adaptive notch characteristics. To achieve optimal damping characteristics and dynamic response speed under complex operating conditions, this invention increases the damping gain coefficient of the DSOGI. Set as It utilizes its unique combination of band-pass and low-pass filters to achieve accurate extraction of the fundamental frequency component.

[0022] This step utilizes two sets of orthogonal transfer functions within DSOGI. and The mathematical expression for bandpass and lowpass filtering of the input signal is as follows: ; In the formula, s is the Laplace operator. and These are the in-phase and quadrature components of the observer output, respectively. The real-time fundamental angular frequency sensed by the system.

[0023] To address the interference of frequent nonlinear load switching on frequency locking accuracy in isolated microgrids caused by significant voltage amplitude fluctuations, this step further introduces a gain normalization update law. Real-time fundamental angular frequency. Adaptive evolution is performed according to the following nonlinear integral law: ; in, The value of the standardized frequency gain determines the dynamic response bandwidth of frequency sensing. This is the frequency deviation discrimination signal obtained by multiplying and subtracting orthogonal voltage vectors; and These are the orthogonal components of the DSOGI output. Denominator terms. This is the square of the instantaneous modulus of the orthogonal components. The amplitude range of voltage fluctuations in islanded microgrids (e.g., 0.2 pu-1.2 pu) far exceeds that of ordinary power electronic systems. This normalization process maintains a constant frequency tracking bandwidth under "large disturbance environments," avoiding nonlinear coupling interference between droop control and harmonic detection. Ultimately, the generated dynamic angular frequency... The feedback is sent to the state observer to form an adaptive closed loop, and is output as the core time-stamped parameter to drive the real-time focusing of the notch point of the subsequent harmonic detection operator.

[0024] S2: Discrete-domain harmonic extraction based on dynamic delay reconstruction: To address the inaccuracy of traditional delay operators caused by frequency shifts in isolated microgrids, this step aims to improve the accuracy of the time-varying angular frequency obtained in step S1. This is converted into discrete addressing logic executable by a digital controller, and the precise separation of harmonic components is achieved through dynamic reconstruction of the signal delay chain. Its logic block diagram is as follows: Figure 2 As shown.

[0025] First, the system establishes the angular frequency based on the real-time sensed frequency state. With the number of discrete sampling points The nonlinear mapping model between them is expressed by the following formula: ; In the formula: d is the number of discrete delay points stored in the digital memory, representing the logical offset of the signal on the discrete time axis; n is the order of the delay signal cancellation (DSC) operator, used to set the target harmonic order; The system sampling period is This is a rounding function. It applies when the calculated d exceeds the hardware buffer limit. At that time, the system executes maximum addressing depth truncation protection.

[0026] Secondly, to address the issue of numerical discontinuity caused by data pointer jumps under variable frequency operating conditions, this step involves setting up a circular buffer in the processor memory. This buffer is based on the sampling frequency. Real-time voltage sample input , .

[0027] The system uses equation (4) to calculate the address offset in real time. Dynamically adjust the read pointer position to read samples. , This mechanism ensures a smooth transition in delay depth when the frequency changes, avoiding signal discontinuities.

[0028] After completing discrete depth addressing alignment, the system executes the Discrete Delay Signal Cancellation (Discrete DSC) operator, which... The discrete calculation formula in the stationary coordinate system is described as follows: ; In the formula, and This refers to the orthogonal voltage component samples input within the current sampling step. and This is a historical component sample that is delayed by d sampling points and is accurately retrieved from the circular buffer.

[0029] This operation utilizes the symmetry of the fundamental wave within half a period for cancellation. By cascading multiple of the above units, a multi-order comb filter can be constructed to enhance and extract specific harmonics, outputting the original harmonic vector. .

[0030] S3: Detection link phase lead compensation based on rotation transformation matrix: To address the physical lag generated during step S2 extraction and the inherent computational delay of the digital control system, this step uses feedforward correction logic to align the compensation vector with the physical phase of the grid distortion source, preventing system oscillations induced by insufficient phase margin. Figure 2 As shown.

[0031] This invention first quantifies and analyzes the total phase lag generated by the harmonic detection link. The delay primarily originates from two parts: one is the inherent group delay caused by the discrete DSC operator structure, whose lag angle at the target harmonic frequency is expressed as... Secondly, there is the hardware computation delay introduced by the sampling, calculation, and PWM update mechanism in the digital signal processing stage. The phase offset generated by this part is expressed as: ; in, Let be the angular frequency of the target harmonic, and h be the harmonic order. ; Define the total compensation angle The sum of the two items above: ; In the formula: is the total lead angle used for vector correction; n is the operator characteristic order defined in step S2; This refers to the fixed sampling step size of the underlying hardware.

[0032] For the original harmonic components output in step S2 and Construct a 2×2 rotation transformation matrix Achieve vector lead deflection. The corrected compensation vector expression is as follows: ; After this transformation, the compensation vector It was moved forward on the timeline. This offsets the delay in the detection and control link, and can achieve a near-ideal 180° counter-offset on the time axis with the actual harmonic distortion in the power grid, thereby improving the suppression effect and system stability.

[0033] S4: Event-triggered distributed coordination and consistency control: Addressing the physical limitations of limited communication bandwidth and heavy computing load among distributed nodes (inverters) in resource-constrained isolated microgrids, this step aims to resolve the uneven distribution of harmonic power among multiple devices through a consensus protocol at the information layer, and couples it with an aperiodic event-triggered mechanism (DETM) to achieve dynamic, on-demand allocation of channel resources. The logic diagram is as follows: Figure 3 As shown.

[0034] S4.1. Construction of Sparse Communication Networks: This invention first abstracts the corresponding communication logic diagram based on the physical topology of a multi-machine parallel inverter. .in, Represents the set of inverter nodes. Represents the communication link between nodes. This is an adjacency matrix. In resource-constrained scenarios, each node does not need to communicate with the global center, but only needs to interact with its physically adjacent nodes through low-bandwidth channels. This framework aims to utilize the mechanism of "local interaction to achieve global synchronization," avoiding the excessive reliance of traditional centralized control on the computing power of a single central controller, and enhancing the system's "plug-and-play" characteristics and resistance to single points of failure.

[0035] S4.2. Consistent Iteration of Harmonic Load Rate: To suppress harmonic circulating currents caused by impedance differences among branches in a parallel system, this step introduces a Dynamic Consensus Algorithm (DCA) to balance the governance burden of each node. Define the first... State variables of Taiwan inverter This represents the harmonic current load rate. Upon receiving the trigger signal, each node updates its state according to the following discrete-time consensus protocol: ; In the formula: For the first Taiwan inverter Estimated harmonic load rate at any given time; These are elements of the adjacency matrix. If a node... , Connectivity Otherwise, it is 0; For nodes The set of adjacent nodes; The convergence weighting coefficient, the magnitude of which determines the speed at which multiple machines reach a consensus on harmonic equalization.

[0036] By propagating and canceling local errors, the harmonic output gain of all parallel inverters is forcibly driven. Converging towards the global average aims to eliminate harmonic circulation caused by inconsistent compensation strengths at the information level.

[0037] S4.3. Adaptive event triggering mechanism: To alleviate the occupancy of limited bandwidth channels by high-frequency synchronization signals, this invention designs an adaptive event triggering criterion with "performance awareness" capability. This step no longer employs a fixed-period sampling and transmission mode, but instead defines a triggering threshold function that dynamically evolves with the system error trajectory. : ; In the formula: This is the trigger gain factor, used to adjust the trade-off between steady-state accuracy and communication frequency; The instantaneous modulus of the harmonic components extracted locally at the current moment is designed to automatically reduce the triggering frequency as the harmonic content decreases. For dynamic decay term, where For steady-state error tolerance, This is the decay rate constant.

[0038] S4.4. Triggering and Execution Based on Lyapunov Stability Criterion: This step compares the consistency error with the dynamic threshold in real time to construct a non-periodic triggering decision maker. Its triggering logic expression is as follows: ; In the formula, This represents the measurement error between the current harmonic state and the state at the time of the last successful trigger. The node only broadcasts the latest compensation vector when this deviation exceeds the dynamic threshold defined by the Lyapunov stability boundary.

[0039] S4.5. Safety justification for excluding the Zeno effect: To ensure the engineering feasibility of the aforementioned aperiodic update strategy, this step introduces a minimum trigger interval (MIET) guarantee mechanism based on the Lipschitz continuity condition. This is achieved by dynamically analyzing the error evolution. By imposing limitations, this invention has mathematically proven the time interval between any two consecutive triggering events. Satisfy the lower bound constraint: ; In the formula, This is the Lipschitz constant in the state equations of the controlled system. It is based on the microgrid circuit topology parameters (inductance, capacitance) and the controller distance step length. The setpoints are pre-calculated offline and stored in the controller ROM. Real-time online calculations are not required, thus avoiding additional computational burden. This mechanism achieves a paradigm shift from "fixed-frequency communication" to "state offset-driven updates," significantly reducing channel throughput while ensuring governance accuracy.

[0040] S5: Multi-machine harmonic current closed-loop equalization control: Based on the consensus reached in step 4, this step performs closed-loop regulation of the local harmonic compensation current to achieve precise current sharing.

[0041] S5.1. Real-time acquisition of the consensus mean of harmonic currents: Each inverter node exchanges locally measured data through the distributed event-triggered communication link established in step S4. The per-unit value of the subharmonic current is used for iterative updates via a Dynamic Consensus Algorithm (DCA), where each node calculates the consensus average value of the global harmonic current in real time. Its core physical significance lies in setting a dynamic reference benchmark for "fair sharing" for each inverter without relying on a central controller.

[0042] S5.2. Design of a PI Equalization Controller Based on Consensus Bias: The system will measure the harmonic current locally. Compared with the consensus mean calculated in step 1 Real-time comparison is performed to generate a consensus tracking error. This error is then fed into a specially designed PI equalizer, whose output harmonic current correction signal... The following regulation law applies: ; In the formula: and These are the proportional and integral coefficients of the equalization controller, respectively. In order to be in The harmonic current equalization adjustment amount generated in the coordinate system. This controller automatically compensates for the current distribution deviation caused by branch impedance mismatch, transforming the originally complex "physical parameter mapping" into an easy-to-implement "automatic error elimination" process.

[0043] S5.3. Final synthesis and injection of compensation vectors: The equilibrium adjustment amount calculated in step 2 via inverse rotation transformation ( arrive Transformation) into stationary coordinate system components Subsequently, the system compares it with the phase correction signal output in step S3. Vector superposition is performed to form the final PWM voltage command correction term.

[0044] Example 1: This invention provides a multi-source cooperative harmonic adaptive suppression method for isolated microgrids, the specific implementation process of which is as follows: 1. Experimental scenario and topology parameter settings: like Figure 4 As shown, this embodiment constructs a parallel islanded microgrid system comprising three distributed power supply units (DG1-DG3) with a rated capacity of 10kVA. Each distributed unit is connected to the common sensitive load bus (SLB) via an LC filter and branch impedance.

[0045] Physical layer parameters: Set the rated output voltage (RMS value), fundamental frequency The branch impedance is set asymmetrically: line impedance 1 is... The line impedance 2 is The line impedance is 3. This is used to simulate branch impedance mismatch conditions.

[0046] Control layer parameters: sampling period Harmonic detection order The bandwidth limit of the distributed communication channel is... .

[0047] 2. Dynamic frequency tracking and discrete mapping process (corresponding to S1-S2): When a nonlinear rectifier load is connected to the system, the droop control causes the main frequency to drop from 50Hz to 49.2Hz.

[0048] Perception performance: The DSOGI-FLL circuit within the local controller captures the angular frequency shift in real time through a normalized gain update law. The normalization factor is calculated... The introduction of this technology allows the frequency tracking loop to converge to the new angular frequency rapidly within 0.05 seconds. .

[0049] Addressing behavior: Discrete addressing depth The signal is then dynamically reconstructed. According to formula (4), the address offset is dynamically mapped from 400 sampling points in the rated state to 406 sampling points. The circular buffer maintains the continuity of the signal flow at the moment of pointer switching, eliminating the stepped discrete noise in the output voltage waveform.

[0050] 3. Phase alignment and event triggering coordination (corresponding to S3-S4): Phase correction: for the phase generated by the detection operator. Group delay, rotation transformation matrix The extracted 5th harmonic components were deflected ahead. Experimental results show that the phase deviation between the compensated signal and the physical distortion source was reduced to within ±5°, greatly expanding the phase margin of the system.

[0051] Trigger determination: Set the trigger sensitivity coefficient Noise tolerance Under steady-state operation, the Lyapunov criterion-based decision maker enters a communication silence state. Compared to the traditional 10kHz periodic synchronization mode, this embodiment achieves over 70% bandwidth savings compared to traditional sampling methods while maintaining accuracy. Furthermore, due to satisfying the Lipschitz continuity constraint, the minimum interval between two communication pulses is... Always maintained at The above effectively avoids processor interrupt storms.

[0052] 4. Load consensus and closed-loop balancing performance (corresponding to S5): Due to the asymmetry of line impedance, DG1 carried too much harmonic current before the coordinated balancing of the three units was activated.

[0053] Collaborative Execution: After enabling the Dynamic Consensus Protocol (DCA), each node exchanges harmonic load rate information. The local PI balancer controller detects load consensus deviations. Adaptive correction of local compensation gain .

[0054] Treatment Result: Through the consistency and equalization adjustment described in this invention, the harmonic sharing deviation among multiple units shows a decrease of orders of magnitude. In this embodiment, the deviation can be suppressed to a low level of less than 2%.

[0055] 5. Summary of Technical Effects: After adopting the method described in this invention, the power quality at the Sensitive Load Bus (SLB) is significantly improved. The total harmonic distortion (THD) of the voltage drops dramatically from its unacceptable state before treatment and eventually stabilizes within the range that meets international power quality standards such as IEEE 519. The system exhibits excellent dynamic adaptability over a wide frequency fluctuation range and greatly reduces its dependence on distributed communication networks and underlying computing resources, demonstrating extremely high engineering practical value.

[0056] Compared with existing technologies, the present invention exhibits significant adaptive characteristics, phase accuracy, and high efficiency in resource scheduling, making it well-suited for the governance of isolated microgrids with limited bandwidth and computing power.

[0057] This invention employs a frequency-sensing closed-loop design based on a standardized gain update law: by introducing a nonlinear integral law with the square of the instantaneous magnitude of the orthogonal components as the denominator into the frequency-locked loop of the dual second-order generalized integrator, the real-time dynamic decoupling of frequency tracking sensitivity and grid voltage amplitude is achieved, thereby constructing an adaptive frequency reference that maintains high steady-state accuracy over a wide voltage range, providing a reliable timescale for subsequent harmonic processing.

[0058] This invention establishes a dynamic mapping model from real-time frequency to discrete buffer delay depth, and couples a circular buffer to achieve smooth signal reconstruction for harmonic separation. Furthermore, by calculating the comprehensive compensation angle composed of discrete group delay and system computation delay, and using a rotation transformation matrix to perform advance rotation on the extracted harmonic vector, the compensation signal and the power grid distortion source are precisely phase aligned on the time axis at the physical level, thus eliminating the risk of control instability caused by phase lag from a mechanistic perspective.

[0059] This invention employs a cascaded architecture of "event-triggered communication" and "consensus closed-loop balancing": the front-end triggers aperiodic neighbor communication through an adaptive dynamic threshold function based on Lyapunov stability design, and generates global load consensus using a consensus protocol; the back-end uses this consensus as a dynamic reference to drive the local PI controller to perform closed-loop balancing adjustment of harmonic currents. This mechanism mathematically avoids Zeno's phenomenon and achieves automatic load sharing among multiple harmonic compensation units and on-demand scheduling of communication resources without requiring precise line impedance parameters.

[0060] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0061] In the description of this invention, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0062] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A multi-source cooperative harmonic adaptive suppression method for isolated microgrids, characterized in that, Includes the following steps: Extract the fundamental component of the output voltage signal of the local inverter in the isolated microgrid and obtain the sensed real-time fundamental angular frequency; Based on the real-time fundamental angular frequency, the harmonic components are separated by dynamically reconstructing the signal delay chain, and the compensation vector is aligned with the physical phase of the power grid distortion source by feedforward correction logic to prevent system oscillations induced by insufficient phase margin. By using a consistency protocol at the information layer and coupling a non-periodic event triggering mechanism, channel resources are dynamically allocated on demand, and then the local harmonic compensation current is adjusted in a closed loop.

2. The multi-source cooperative harmonic adaptive suppression method for islanded microgrids according to claim 1, characterized in that: When extracting the fundamental component, considering the DC offset and specific order harmonic interference that microgrid sensors are prone to generate in high-frequency switching environments, the signal is converted into components in the αβ stationary coordinate system using Clark transform. A dual second-order generalized integrator (DSOGI) is then used as a front-end state observer with adaptive notch characteristics to extract the fundamental component. The damping gain coefficient k of the DSOGI is set to... The fundamental component is extracted by using a combination of bandpass and lowpass filters.

3. The multi-source cooperative harmonic adaptive suppression method for islanded microgrids according to claim 2, characterized in that: When acquiring the real-time fundamental angular frequency, a gain normalization update law is introduced to control the real-time fundamental angular frequency. Adaptive evolution is performed according to the following nonlinear integral law: ； in, The value of the standardized frequency gain determines the dynamic response bandwidth of frequency sensing. This is the frequency deviation discrimination signal obtained by multiplying and subtracting orthogonal voltage vectors; and The orthogonal components of the DSOGI output; denominator terms is the square of the instantaneous modulus of the orthogonal component.

4. The multi-source cooperative harmonic adaptive suppression method for islanded microgrids according to claim 3, characterized in that: When extracting harmonic components, a nonlinear mapping model is established between the number of discrete sampling points d of the real-time fundamental angular frequency; the symmetry of the fundamental wave within half a period is used for cancellation; by cascading multiple of the above units, a multi-order comb filter is formed to achieve enhanced extraction of specific harmonics, output the original harmonic vector, and complete the separation and extraction of harmonic components.

5. The multi-source cooperative harmonic adaptive suppression method for islanded microgrids according to claim 4, characterized in that: When performing dynamic on-demand allocation of channel resources, a sparse communication network is constructed based on the physical topology of the multi-machine parallel inverters, and a dynamic consensus algorithm is introduced to balance the governance burden of each node. By setting an adaptive event triggering mechanism, a triggering and execution mechanism based on the Lyapunov stability criterion, and a minimum triggering interval guarantee mechanism based on the Lipschitz continuity condition, the dynamic on-demand allocation of channel resources is completed.

6. The multi-source cooperative harmonic adaptive suppression method for islanded microgrids according to claim 5, characterized in that: When performing multi-machine harmonic current closed-loop equalization control, a PI equalization controller based on the consensus deviation is set according to the obtained harmonic current consensus mean value. Using the PI equalization controller, a harmonic current correction signal is generated based on the consensus tracking error obtained by comparing the locally measured harmonic current with the consensus mean value of the harmonic current. The harmonic current correction signal is transformed into a stationary coordinate system component through an inverse rotation transformation, and then vector-superimposed with the compensation vector to form the final PWM voltage command correction term.

7. The multi-source cooperative harmonic adaptive suppression system for islanded microgrids according to claim 6, characterized in that: When acquiring the compensation vector, the total phase lag generated by the harmonic detection link is quantitatively analyzed. The delay source includes two parts: one is the inherent group delay caused by the discrete DSC operator structure, whose lag angle at the target harmonic frequency is expressed as... Secondly, there is the hardware computation delay introduced by the sampling, calculation, and PWM update mechanism in the digital signal processing stage. The phase offset generated by this part is expressed as: ； in, Let be the angular frequency of the target harmonic, and h be the harmonic order. ； Define the total compensation angle The sum of the two items above: ； In the formula: The total lead angle used for vector correction; n is the operator characteristic order; This refers to a fixed sampling step size for the underlying hardware. For the original harmonic components and Construct a 2×2 rotation transformation matrix Achieve vector lead deflection; the corrected compensation vector expression is as follows: ； After this transformation, the compensation vector It was moved forward on the timeline. This offsets the delay in the detection and control link, enabling a near-ideal 180° counter-offset on the time axis against the actual harmonic distortion in the power grid.

8. A multi-source cooperative harmonic adaptive suppression system for islanded microgrids, used to execute the multi-source cooperative harmonic adaptive suppression method for islanded microgrids as described in claim 1, characterized in that, include: The fundamental component extraction module is used to extract the fundamental component of the output voltage signal of the local inverter in the islanded microgrid and obtain the sensed real-time fundamental angular frequency. The harmonic component extraction module is used to separate harmonic components based on the real-time fundamental angular frequency by dynamically reconstructing the signal delay chain, and to align the compensation vector with the physical phase of the power grid distortion source through feedforward correction logic to prevent system oscillations induced by insufficient phase margin. The adaptive suppression module is used to dynamically allocate channel resources on demand through the information layer's consistency protocol and coupled with an aperiodic event triggering mechanism, and then perform closed-loop adjustment of the local harmonic compensation current.

Images