Droop control island detection system

Abstract

The invention discloses a droop control island detection system, which comprises a distributed control node and a droop control node, the system is composed of a plurality of control nodes distributed in a power grid, and each node comprises an inverter control unit, an energy management unit, a communication module and a sensor module; each node can perform droop control independently and perform cooperative work through the communication module; and the system is provided with an energy management center which is responsible for monitoring the energy flow condition in the power grid in real time, integrating distributed energy resources, exchanging data with each control node and sending an optimization instruction. The method can dynamically adjust the control parameters of each node according to the real-time state of the power grid, and guarantees the continuous power supply of the power grid and the optimal configuration of energy in the island state.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a droop control island detection system. Background Art

[0002] In modern power systems, the popularity of distributed energy resources (DERs) has brought new challenges to the stability and reliability of power systems. Traditional power systems rely on centralized power generation and control. However, as more and more distributed generation units (such as photovoltaic power generation, wind power generation, etc.) are connected to the power grid, the operating characteristics of the power grid have become more complex, especially in terms of island detection and energy management. Island detection refers to the process of detecting the island state and taking response measures to avoid grid instability when the power grid loses connection with the main grid. Existing island detection methods usually rely on a single detection technology, such as abnormal fluctuation detection of voltage and frequency. However, these methods often have problems such as low detection accuracy and slow response speed.

[0003] At the same time, traditional power grid energy management systems are also unable to meet the large-scale access needs of distributed energy. Due to the wide geographical distribution of distributed energy, traditional centralized energy management models are difficult to achieve effective scheduling of these resources. In addition, with the rapid development of blockchain technology, its application in power systems has gradually attracted attention. Through blockchain technology, the transparency and security of power transactions can be achieved, thereby further improving the reliability and efficiency of the system.

[0004] In summary, the prior art has the following deficiencies:

[0005] 1. The accuracy and reliability of island detection need to be improved. Existing island detection methods perform poorly in complex power grid environments.

[0006] 2. The energy management system is difficult to effectively cope with the large-scale access of distributed energy. The traditional centralized management model is insufficient in efficiency and flexibility.

[0007] 3. The transparency and security of energy transactions are insufficient, and the existing system lacks effective technical means to ensure the fairness and data security of distributed energy transactions.

[0008] To this end, we propose a droop control island detection system to solve the above-mentioned problems. Summary of the invention

[0009] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions will be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0010] In view of the above problems existing in the existing droop control island detection system, the present invention is proposed.

[0011] Therefore, the object of the present invention is to provide a droop control island detection system, which can dynamically adjust the control parameters of each node according to the real-time state of the power grid to ensure continuous power supply and optimal energy configuration of the power grid in the island state.

[0012] In order to solve the above technical problems, the present invention provides the following technical solutions: a droop control island detection system, comprising:

[0013] Distributed control nodes: The system consists of multiple control nodes distributed in the power grid. Each node contains an inverter control unit, energy management unit, communication module and sensor module. Each node can independently perform droop control and work together through the communication module.

[0014] Energy Management Center: The system is equipped with an energy management center, which is responsible for real-time monitoring of energy flow in the power grid, integrating distributed energy resources, and exchanging data and sending optimization instructions with each control node;

[0015] Blockchain trading platform: In order to realize the integration of energy trading and island detection, the system introduces blockchain technology to establish a distributed energy trading platform. Each control node and energy management center can conduct real-time data interaction and energy transaction records on the platform to ensure the transparency and security of the system.

[0016] As a preferred solution of the droop control island detection system of the present invention, the detection method of the droop control island detection system comprises the following steps:

[0017] Step 1: System initialization and node configuration

[0018] When the system starts, the energy management center first initializes all distributed control nodes, including parameter settings and network connections;

[0019] Each node collects local voltage, current, and frequency parameters through the sensor module, and uploads these data to the energy management center through the communication module;

[0020] Step 2: Distributed droop control and real-time optimization

[0021] During normal grid operation, each control node executes a droop control algorithm to balance the grid load by adjusting the output power;

[0022] The energy management center dynamically adjusts the droop control parameters of each node based on real-time data to adapt to grid fluctuations and optimize system performance;

[0023] The blockchain platform records energy transaction information in real time to ensure the transparency and security of system energy management;

[0024] Step 3: Island status warning and collaborative detection

[0025] When the system detects abnormal fluctuations in voltage or frequency, each control node starts to execute the island detection algorithm and makes a preliminary judgment based on the voltage, frequency, and phase information;

[0026] Each node shares the detection results through the communication module. The energy management center executes the data fusion algorithm based on the received distributed information to comprehensively judge the island status of the entire system;

[0027] If an islanding state is detected, each node immediately adjusts the droop control parameters, switches to the islanding operation mode, and records the information of the islanding through the blockchain platform;

[0028] Step 4: Energy management and recovery in isolated island state

[0029] In the isolated island state, the energy management center dynamically adjusts the output power of each node based on real-time distributed energy resource data to ensure load balance and power supply stability within the isolated island;

[0030] The system continues to monitor grid parameters in the island state and waits for grid recovery;

[0031] Once the external power grid resumes power supply, the energy management center will gradually switch each node back to the grid-connected mode, the system will exit the island mode and resume normal operation;

[0032] Step 5: Data analysis and system optimization

[0033] After the islanding state ends, the energy management center analyzes the data of the entire islanding detection process to identify the optimization opportunities that exist;

[0034] The system shares test results and energy management data through the blockchain platform, further improving the system’s robustness and detection accuracy;

[0035] Distributed droop control: Each control node independently executes the droop control algorithm through its inverter control unit, and adjusts the output power according to the voltage, current and frequency parameters monitored in real time to maintain the stability of the power grid. The adjustment of droop parameters is controlled by the intelligent algorithm inside the node and can be dynamically optimized according to the overall needs of the system.

[0036] Collaborative island detection mechanism: Through the communication module, each node can share island detection information and make collaborative decisions. Each node not only detects the island status independently, but also combines the information of other nodes through data fusion technology to improve the accuracy and reliability of detection;

[0037] Dynamic energy management and detection integration: The energy management center globally optimizes and adjusts the droop control parameters of each control node according to the real-time status of distributed energy resources. The system can achieve optimal energy configuration while detecting the island state, ensuring continuous power supply to the power grid in the island state.

[0038] As a preferred solution of the droop control island detection system of the present invention, wherein: the droop control algorithm in step 2 includes active power-frequency droop control and reactive power-voltage droop control;

[0039] The formula for active power-frequency droop control is:

[0040] f=f0-m P (P-P0)

[0041] Where: f is the output frequency of the inverter;

[0042] f0 is the rated frequency, usually 50Hz or 60Hz;

[0043] m P is the active power droop coefficient;

[0044] P is the current output active power of the inverter;

[0045] P0 is the reference value of active power, which is the rated power;

[0046] The formula for reactive power-voltage droop control is:

[0047] V=V0-m Q ·(Q-Q0)

[0048] Where V is the output voltage of the inverter;

[0049] V0 is the rated voltage;

[0050] m Q is the reactive power droop factor;

[0051] Q is the current output reactive power of the inverter;

[0052] Q0 is the reference value of reactive power, usually zero.

[0053] As a preferred solution of the droop control island detection system of the present invention, the adaptive droop coefficient adjustment introduces an adaptive algorithm to adjust the droop coefficient m in order to adapt to the dynamic changes in the power grid. P and m Q Make real-time adjustments to improve the system's response speed and stability;

[0054] formula:

[0055] Adaptive active power droop coefficient adjustment:

[0056]

[0057] in, is the initial active power droop coefficient;

[0058] k P It is the adaptive adjustment factor of active power droop coefficient;

[0059] df / dt is the rate of change of frequency;

[0060] Adaptive reactive power droop coefficient adjustment:

[0061]

[0062] in:

[0063] is the initial reactive droop coefficient;

[0064] k Q is the adaptive adjustment factor of reactive power droop coefficient;

[0065] dV / dt is the rate of change of voltage.

[0066] As a preferred solution of the droop control island detection system of the present invention, when multiple inverters are operating together, the system can share the operating status of each inverter through a communication module and use a distributed algorithm to collaboratively optimize the droop control parameters;

[0067] Optimization goal:

[0068] Minimize frequency deviation:

[0069] Minimize voltage deviation:

[0070]

[0071] Coordinated power distribution:

[0072] and

[0073] Under the premise that the total power meets the system requirements, the power distribution between the inverters should be optimal.

[0074] As a preferred solution of the droop control island detection system of the present invention, the process of the droop control algorithm includes:

[0075] S1. Initial setting: Each control node sets the initial droop coefficient and and perform droop control;

[0076] S2. Real-time monitoring and adjustment: The node continuously monitors the output frequency f, voltage V, active power P, and reactive power Q parameters. When a parameter change is detected, an adaptive algorithm is used to adjust the droop coefficient m. P (t) and m Q (t);

[0077] S3, Collaborative optimization: Through the communication module, each node shares the current power, frequency and voltage data, and the energy management center performs global optimization of the droop coefficient and power distribution to reduce frequency and voltage deviations and ensure system stability;

[0078] S4, feedback and update: Each node continues to execute the control algorithm according to the optimized droop coefficient, and the system executes this process in a loop to achieve real-time dynamic adjustment.

[0079] As a preferred solution of the droop control island detection system of the present invention, the island detection algorithm in step 3 adopts passive detection and active detection methods to improve the accuracy and response speed of island detection, specifically including:

[0080] Passive detection method: By monitoring abnormal changes in grid parameters such as voltage, frequency and phase, it is possible to determine whether an island has occurred.

[0081] Active detection method: detects islanding by injecting specific disturbance signals into the power grid and observing the system response.

[0082] As a preferred solution of the droop control island detection system of the present invention, the passive detection algorithm includes the following:

[0083] 1) Under / overvoltage protection, algorithm steps:

[0084] Voltage monitoring: real-time monitoring of the voltage V(t) at the inverter output end;

[0085] Threshold judgment: Set the undervoltage threshold V minand overvoltage threshold V max ,like

[0086] V(t)<V min or V(t)>V max , then it is determined that an island occurs;

[0087] 2) Under / over frequency protection, algorithm steps:

[0088] Monitoring frequency: Real-time monitoring of inverter output frequency f(t)

[0089] Threshold judgment: Set the underfrequency threshold f min and overfrequency threshold f max ,like

[0090] f(t)<f min or f(t)>f max , then it is determined that an island occurs;

[0091] 3) Frequency change rate, algorithm steps:

[0092] Calculate the rate of change of frequency:

[0093]

[0094] Among them, Δt is the sampling interval time;

[0095] Threshold judgment: Set the frequency change rate threshold like

[0096]

[0097] It is determined that an island occurs;

[0098] 4) Phase jump detection, algorithm steps:

[0099] Monitoring phase: real-time monitoring of voltage phase θ(t);

[0100] Calculate the phase change:

[0101] Δθ=θ(t)-θ(t-Δt)

[0102] Threshold judgment: Set the phase jump threshold Δθ threshold ,like

[0103] |Δθl>Δθ threshold

[0104] It is determined that an island has occurred.

[0105] As a preferred solution of the droop control island detection system of the present invention, the active detection method injects a specific disturbance into the power grid to observe the response of the system, including:

[0106] (1) Active frequency drift, algorithm steps:

[0107] Output current modulation: A sinusoidal modulation term is introduced into the inverter output current so that the current frequency is:

[0108] f output =f0+Δf·sgn(sin(2πf0t))

[0109] Where Δf is the frequency offset, sgn is the sign function;

[0110] Monitor frequency response: If the system is in grid-connected loading, the frequency rigidity of the grid keeps the frequency stable; in an islanded state, the frequency will gradually deviate;

[0111] Threshold judgment: if the frequency deviation exceeds the set threshold, it is judged as an island;

[0112] (2) Sandia frequency shift, algorithm steps:

[0113] Output current modulation: Output current frequency:

[0114] f output =f0+k·Δf·sin(2πf0t)

[0115] Where k is the gain coefficient;

[0116] Monitor frequency response: In an islanding state, the frequency can deviate rapidly, triggering protection.

[0117] As a preferred solution of the droop control island detection system of the present invention, the data fusion algorithm mainly includes the following steps:

[0118] Data preprocessing: Preliminary screening and preprocessing of the data sent by each control node to remove noise and outliers.

[0119] formula:

[0120] For each node i, the voltage data V i (t) Perform moving average processing:

[0121]

[0122] Where n is the smoothing window size;

[0123] Use Kalman filtering to filter the frequency data f i (t) Processing:

[0124]

[0125] Where K(t) is the Kalman gain, is the frequency after filtering;

[0126] Data weighting: The data is weighted according to the credibility or location of each node. The formula is:

[0127] Let the weight factor w of each node i be i , then the weighted data is:

[0128] Voltage:

[0129]

[0130] frequency:

[0131]

[0132] The calculation of the weighting factor is based on the following formula:

[0133] Position weight, assuming the position weight w location，i Based on the distance d to the substation i :

[0134]

[0135] Historical accuracy weight, assuming that accuracy is based on the success rate p of the past N detections i :

[0136] w accuracy，i =p i

[0137] Total weight:

[0138] w i =α·w location，i +β·w accuracy，i

[0139] Among them, α and β are proportional coefficients of weighting factors, which can be adjusted according to actual needs;

[0140] Data fusion: Combine the weighted data of each node, use weighted average, voting mechanism or Bayesian reasoning and other methods to comprehensively judge the island status. The formula is:

[0141] Integrated voltage:

[0142]

[0143] Comprehensive frequency:

[0144]

[0145] Consistency judgment: Based on the fusion results, determine whether the island event is confirmed and feed the results back to each node.

[0146] Beneficial effects of the present invention:

[0147] 1. Improve the accuracy and reliability of island detection: By introducing distributed control nodes and collaborative island detection mechanism, each node can not only perform island detection independently, but also make collaborative decisions through shared information and data fusion technology, which greatly improves the accuracy and reliability of island detection and reduces the possibility of false detection and missed detection.

[0148] 2. Realize the integration of dynamic energy management and detection: The energy management center of the present invention can monitor the energy flow in the power grid in real time and globally optimize the droop control parameters of each control node. The system can dynamically adjust the control parameters of each node according to the real-time status of the power grid to ensure the continuous power supply of the power grid and the optimal configuration of energy in the island state. BRIEF DESCRIPTION OF THE DRAWINGS

[0149] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them:

[0150] Figure 1 It is a schematic diagram of the overall structure of the droop control island detection system of the present invention.

[0151] Figure 2 Schematic diagram of node voltage and frequency changes in embodiments 1 and 2 of the droop control island detection system of the present invention. DETAILED DESCRIPTION

[0152] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0153] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0154] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0155] Secondly, the present invention is described in detail with reference to the schematic diagram. When describing the embodiments of the present invention in detail, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0156] Reference Figure 1 -2, provides a droop control island detection system, including:

[0157] Distributed control nodes: The system consists of multiple control nodes distributed in the power grid. Each node contains an inverter control unit, energy management unit, communication module and sensor module. Each node can independently perform droop control and work together through the communication module.

[0158] Energy Management Center: The system is equipped with an energy management center, which is responsible for real-time monitoring of energy flow in the power grid, integrating distributed energy resources, and exchanging data and sending optimization instructions with each control node;

[0159] Blockchain trading platform: In order to achieve the integration of energy trading and island detection, the system introduces blockchain technology to establish a distributed energy trading platform. Each control node and energy management center can conduct real-time data interaction and energy transaction records on the platform to ensure the transparency and security of the system;

[0160] Distributed droop control: Each control node independently executes the droop control algorithm through its inverter control unit, and adjusts the output power according to the voltage, current and frequency parameters monitored in real time to maintain the stability of the power grid. The adjustment of droop parameters is controlled by the intelligent algorithm inside the node and can be dynamically optimized according to the overall needs of the system;

[0161] Collaborative island detection mechanism: Through the communication module, each node can share island detection information and make collaborative decisions. Each node not only detects the island status independently, but also combines the information of other nodes through data fusion technology to improve the accuracy and reliability of detection;

[0162] Dynamic energy management and detection integration: The energy management center globally optimizes and adjusts the droop control parameters of each control node according to the real-time status of distributed energy resources. The system can achieve optimal energy configuration while detecting the island state, ensuring continuous power supply to the power grid in the island state.

[0163] The detection method of the droop control island detection system includes the following steps:

[0164] Step 1: System initialization and node configuration

[0165] When the system starts, the energy management center first initializes all distributed control nodes, including parameter settings and network connections;

[0166] Each node collects local voltage, current, and frequency parameters through the sensor module, and uploads these data to the energy management center through the communication module;

[0167] Step 2: Distributed droop control and real-time optimization

[0168] During normal grid operation, each control node executes a droop control algorithm to balance the grid load by adjusting the output power;

[0169] The energy management center dynamically adjusts the droop control parameters of each node based on real-time data to adapt to grid fluctuations and optimize system performance;

[0170] The blockchain platform records energy transaction information in real time to ensure the transparency and security of system energy management;

[0171] Step 3: Island status warning and collaborative detection

[0172] When the system detects abnormal fluctuations in voltage or frequency, each control node starts to execute the island detection algorithm and makes a preliminary judgment based on the voltage, frequency, and phase information;

[0173] Each node shares the detection results through the communication module. The energy management center executes the data fusion algorithm based on the received distributed information to comprehensively judge the island status of the entire system;

[0174] If an islanding state is detected, each node immediately adjusts the droop control parameters, switches to the islanding operation mode, and records the information of the islanding through the blockchain platform;

[0175] Step 4: Energy management and recovery in isolated island state

[0176] In the isolated island state, the energy management center dynamically adjusts the output power of each node based on real-time distributed energy resource data to ensure load balance and power supply stability within the isolated island;

[0177] The system continues to monitor grid parameters in the island state and waits for grid recovery;

[0178] Once the external power grid resumes power supply, the energy management center will gradually switch each node back to the grid-connected mode, the system will exit the island mode and resume normal operation;

[0179] Step 5: Data analysis and system optimization

[0180] After the islanding state ends, the energy management center analyzes the data of the entire islanding detection process to identify the optimization opportunities that exist;

[0181] The system shares detection results and energy management data through the blockchain platform to further improve the system's robustness and detection accuracy.

[0182] Wherein, the droop control algorithm in step 2 includes active power-frequency droop control and reactive power-voltage droop control;

[0183] The formula for active power-frequency droop control is:

[0184] f=f0-m P (P-P0)

[0185] Where: f is the output frequency of the inverter;

[0186] f0 is the rated frequency, usually 50Hz or 60Hz;

[0187] m P is the active power droop coefficient;

[0188] P is the current output active power of the inverter;

[0189] P0 is the reference value of active power, which is the rated power;

[0190] The formula for reactive power-voltage droop control is:

[0191] V=V0-m Q ·(Q-Q0)

[0192] Where V is the output voltage of the inverter;

[0193] V0 is the rated voltage;

[0194] m Q is the reactive power droop factor;

[0195] Q is the current output reactive power of the inverter;

[0196] Q0 is the reference value of reactive power, usually zero.

[0197] Furthermore, in order to adapt to the dynamic changes in the power grid, an adaptive algorithm is introduced to adjust the droop coefficient m. P and m Q Make real-time adjustments to improve the system's response speed and stability;

[0198] formula:

[0199] Adaptive active power droop coefficient adjustment:

[0200]

[0201] in, is the initial active power droop coefficient;

[0202] k P It is the adaptive adjustment factor of active power droop coefficient;

[0203] df / dt is the rate of change of frequency;

[0204] Adaptive reactive power droop coefficient adjustment:

[0205]

[0206] in:

[0207] is the initial reactive droop coefficient;

[0208] k Q is the adaptive adjustment factor of reactive power droop coefficient;

[0209] dV / dt is the rate of change of voltage.

[0210] Furthermore, when multiple inverters are operating together, the system can share the operating status of each inverter through the communication module and use the distributed algorithm to collaboratively optimize the droop control parameters;

[0211] Optimization goal:

[0212] Minimize frequency deviation:

[0213] Minimize voltage deviation:

[0214]

[0215] Coordinated power distribution:

[0216] and

[0217] Under the premise that the total power meets the system requirements, the power distribution between the inverters should be optimal.

[0218] The process of the droop control algorithm includes:

[0219] S1. Initial setting: Each control node sets the initial droop coefficient and and perform droop control;

[0220] S2. Real-time monitoring and adjustment: The node continuously monitors the output frequency f, voltage V, active power P, and reactive power Q parameters. When a parameter change is detected, an adaptive algorithm is used to adjust the droop coefficient m. P (t) and m Q (t);

[0221] S3, Collaborative optimization: Through the communication module, each node shares the current power, frequency and voltage data, and the energy management center performs global optimization of the droop coefficient and power distribution to reduce frequency and voltage deviations and ensure system stability;

[0222] S4, feedback and update: Each node continues to execute the control algorithm according to the optimized droop coefficient, and the system executes this process in a loop to achieve real-time dynamic adjustment.

[0223] Specific example: Assume that there are three distributed inverter control nodes A, B, and C in a microgrid system. Each node independently executes the droop control algorithm and works together through the communication module.

[0224] Step 1: System initialization

[0225] Node initialization: When control nodes A, B, and C are started, the initial droop coefficients are configured respectively. and And collect local voltage, current and frequency parameters through sensor modules;

[0226] Initial power output:

[0227] Assume that under initial conditions, node A outputs active power P A =5kw, reactive power Q A =2kVar;

[0228] Node B outputs active power P B =4kw, reactive power Q B =1.5kVar;

[0229] Node C outputs active power P C =6kw, reactive power Q C =2.5kVar;

[0230] Step 2: Real-time monitoring and preliminary droop control

[0231] Grid fluctuation detection: Assume that the grid fluctuates and node A detects that the frequency drops to f A =49.5Hz, Node B detects that the frequency drops to f B =49.4Hz, node C detects a drop to f C =49.6Hz;

[0232] Initial droop adjustment: Node A adjusts the output power according to the droop control formula:

[0233] f A =f0-m p ·(P A -P0)

[0234] The calculation results in: Adjusted P A 5.5kw;

[0235] Node B and node C are adjusted similarly, and the new output powers are P B =4.8kw and P C =6.2kw;

[0236] Voltage adjustment: Nodes A, B, and C adjust their output voltages according to the droop control formula to maintain the system voltage close to the rated voltage V0;

[0237] Step 3: Collaborative Optimization between Nodes

[0238] Each node sends the current power, frequency and voltage data to other nodes and the energy management center through the communication module. According to the received frequency and voltage deviations, the distributed optimization algorithm is executed to calculate the new global optimal droop coefficient.

[0239] Through the above examples and interactive relationship design, we can see that each control node independently executes the droop control algorithm. At the same time, through the collaborative optimization of the energy management center and the real-time data sharing of the communication module, the entire system can more accurately adjust the grid state to ensure that the stability and efficiency of the grid can be maintained in various situations (such as island mode).

[0240] The island detection algorithm uses passive and active detection methods to improve the accuracy and response speed of island detection, including:

[0241] Passive detection method: By monitoring abnormal changes in grid parameters such as voltage, frequency and phase, it is possible to determine whether an island has occurred.

[0242] Active detection method: detects islanding by injecting specific disturbance signals into the power grid and observing the system response.

[0243] Specifically, the passive detection algorithms include the following:

[0244] 1) Under / overvoltage protection, algorithm steps:

[0245] Voltage monitoring: real-time monitoring of the voltage V(t) at the inverter output end;

[0246] Threshold judgment: Set the undervoltage threshold V min and overvoltage threshold V max ,like

[0247] V(t)<V min or V(t)>V max , then it is determined that an island occurs;

[0248] 2) Under / over frequency protection, algorithm steps:

[0249] Monitoring frequency: Real-time monitoring of inverter output frequency f(t)

[0250] Threshold judgment: Set the underfrequency threshold f min and overfrequency threshold f max ,like

[0251] f(t)<f min or f(t)>f max , then it is determined that an island occurs;

[0252] 3) Frequency change rate, algorithm steps:

[0253] Calculate the rate of change of frequency:

[0254]

[0255] Among them, Δt is the sampling interval time;

[0256] Threshold judgment: Set the frequency change rate threshold like

[0257]

[0258] It is determined that an island occurs;

[0259] 4) Phase jump detection, algorithm steps:

[0260] Monitoring phase: real-time monitoring of voltage phase θ(t);

[0261] Calculate the phase change:

[0262] Δθ=θ(t)-θ(t-Δt)

[0263] Threshold judgment: Set the phase jump threshold Δθ threshold ,like

[0264] |Δθ|>Δθ threshold

[0265] It is determined that an island has occurred.

[0266] More specifically, active detection methods inject specific disturbances into the power grid to observe the system's response, including:

[0267] (1) Active frequency drift, algorithm steps:

[0268] Output current modulation: A sinusoidal modulation term is introduced into the inverter output current so that the current frequency is:

[0269] f output=f0+Δf·sgn(sin(2πf0t))

[0270] Where Δf is the frequency offset, sgn is the sign function;

[0271] Monitor frequency response: If the system is in grid-connected loading, the frequency rigidity of the grid keeps the frequency stable; in an islanded state, the frequency will gradually deviate;

[0272] Threshold judgment: if the frequency deviation exceeds the set threshold, it is judged as an island;

[0273] (2) Sandia frequency shift, algorithm steps:

[0274] Output current modulation: Output current frequency:

[0275] f output =f0+k·Δf·sin(2πf0t)

[0276] Where k is the gain coefficient;

[0277] Monitor frequency response: In an islanding state, the frequency can deviate rapidly, triggering protection.

[0278] Furthermore, the data fusion algorithm mainly includes the following steps:

[0279] Data preprocessing: Preliminary screening and preprocessing of the data sent by each control node to remove noise and outliers.

[0280] formula:

[0281] For each node i, the voltage data V i (t) Perform moving average processing:

[0282]

[0283] Where n is the smoothing window size;

[0284] Use Kalman filtering to filter the frequency data f i (t) Processing:

[0285]

[0286] Where K(t) is the Kalman gain, is the frequency after filtering;

[0287] Data weighting: The data is weighted according to the credibility or location of each node. The formula is:

[0288] Let the weight factor w of each node i be i , then the weighted data is:

[0289] Voltage:

[0290]

[0291] frequency:

[0292]

[0293] The calculation of the weighting factor is based on the following formula:

[0294] Position weight, assuming the position weight w location，i Based on the distance d to the substation i :

[0295]

[0296] Historical accuracy weight, assuming that accuracy is based on the success rate p of the past N detections i :

[0297] w accuracy，i =p i

[0298] Total weight:

[0299] w i =α·w location，i +β·w accuracy，i

[0300] Among them, α and β are proportional coefficients of weighting factors, which can be adjusted according to actual needs;

[0301] Data fusion: Combine the weighted data of each node, use weighted average, voting mechanism or Bayesian reasoning and other methods to comprehensively judge the island status. The formula is:

[0302] Integrated voltage:

[0303]

[0304] Comprehensive frequency:

[0305]

[0306] Consistency judgment: Based on the fusion results, determine whether the island event is confirmed and feed the results back to each node.

[0307] Specific example:

[0308] Assume that the voltage and frequency data of the three nodes in the system are as follows:

[0309] Node A: V A (t)=205V,f A (t) = 48.8 Hz;

[0310] Node B: V B (t) = 230 V, f B (t) = 50.2 Hz;

[0311] Node C: V C (t)=254V,f C (t) = 50.8 Hz;

[0312] The weighting factors are:

[0313] W A =0.5

[0314] W B =0.3

[0315] W C =0.2

[0316] The integrated system voltage and frequency after fusion are:

[0317]

[0318]

[0319] After fusion, the system's comprehensive judgment results show that both voltage and frequency are abnormal, and the energy management center can further confirm the occurrence of islanding events based on the set thresholds.

[0320] Reference Figure 2 , specific embodiment:

[0321] Example 1: Island event triggering, primary detection of node A

[0322] The voltage and frequency data of the three nodes in the system are as follows:

[0323] Node A: V A (t) drops from 230V to 200V in [0, 100] seconds, frequency f A (t) from 50 Hz to 48 Hz;

[0324] Node B: V B (t) Maintained at 230V, frequency f B (t) Maintained at 50 Hz;

[0325] Node C: V C (t) Increase from 230V to 255V in [0, 100] seconds, frequency f C (t) from 50 Hz to 52 Hz;

[0326] Example 2: Dynamic adjustment of voltage and frequency in an island state:

[0327] After nodes A and C enter island mode, the voltage and frequency changes are:

[0328] Node A: Voltage V A (t) Recover to 220V within [100,200] seconds, frequency f A (t) Stable at 49 Hz;

[0329] Node C: Voltage V C (t) drops back to 245V within [100,200] seconds, frequency f C (t) Stable at 51 Hz;

[0330] Chart Description:

[0331] 1. Node voltage change diagram: shows the voltage changes of nodes A, B, and C during the islanding event and in the islanding mode. It can be seen that the voltages of nodes A and C fluctuate abnormally during the islanding event, and then return to a relatively stable state in the islanding mode.

[0332] 2. Node frequency change diagram: shows the frequency changes of nodes A, B, and C during the islanding event and in the islanding mode. The frequencies of nodes A and C drifted significantly during the islanding event, and then stabilized at new frequency values ​​in the islanding mode.

[0333] 3. Voltage and frequency diagram under normal state: used to show the stability of the system under normal operation. Here only the voltage and frequency of node B are shown as a representative to show the stability of the system.

[0334] Through these charts, the voltage and frequency changes of the system under different embodiments can be clearly shown, thereby verifying and illustrating the effectiveness of the island detection system.

[0335] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A droop control island detection system, characterized in that: include: Distributed control nodes: The system consists of multiple control nodes distributed in the power grid. Each node contains an inverter control unit, energy management unit, communication module and sensor module. Each node can independently perform droop control and work together through the communication module. Energy Management Center: The system is equipped with an energy management center, which is responsible for real-time monitoring of energy flow in the power grid, integrating distributed energy resources, and exchanging data and sending optimization instructions with each control node; Blockchain trading platform: In order to realize the integration of energy trading and island detection, the system introduces blockchain technology to establish a distributed energy trading platform. Each control node and energy management center can conduct real-time data interaction and energy transaction records on the platform to ensure the transparency and security of the system.

2. The droop control island detection system according to claim 1, characterized in that: The detection method of the droop control island detection system comprises the following steps: Step 1: System initialization and node configuration When the system starts, the energy management center first initializes all distributed control nodes, including parameter settings and network connections; Each node collects local voltage, current, and frequency parameters through the sensor module, and uploads these data to the energy management center through the communication module; Step 2: Distributed droop control and real-time optimization During normal grid operation, each control node executes a droop control algorithm to balance the grid load by adjusting the output power; The energy management center dynamically adjusts the droop control parameters of each node based on real-time data to adapt to grid fluctuations and optimize system performance; The blockchain platform records energy transaction information in real time to ensure the transparency and security of system energy management; Step 3: Island status warning and collaborative detection When the system detects abnormal fluctuations in voltage or frequency, each control node starts to execute the island detection algorithm and makes a preliminary judgment based on the voltage, frequency, and phase information; Each node shares the detection results through the communication module. The energy management center executes the data fusion algorithm based on the received distributed information to comprehensively judge the island status of the entire system; If an islanding state is detected, each node immediately adjusts the droop control parameters, switches to the islanding operation mode, and records the information of the islanding through the blockchain platform; Step 4: Energy management and recovery in isolated island state In the isolated island state, the energy management center dynamically adjusts the output power of each node based on real-time distributed energy resource data to ensure load balance and power supply stability within the isolated island; The system continues to monitor grid parameters in the island state and waits for grid recovery; Once the external power grid resumes power supply, the energy management center will gradually switch each node back to the grid-connected mode, the system will exit the island mode and resume normal operation; Step 5: Data analysis and system optimization After the islanding state ends, the energy management center analyzes the data of the entire islanding detection process to identify the optimization opportunities that exist; The system shares test results and energy management data through the blockchain platform, further improving the system’s robustness and detection accuracy; Distributed droop control: Each control node independently executes the droop control algorithm through its inverter control unit, and adjusts the output power according to the voltage, current and frequency parameters monitored in real time to maintain the stability of the power grid. The adjustment of droop parameters is controlled by the intelligent algorithm inside the node and can be dynamically optimized according to the overall needs of the system. Collaborative island detection mechanism: Through the communication module, each node can share island detection information and make collaborative decisions. Each node not only detects the island status independently, but also combines the information of other nodes through data fusion technology to improve the accuracy and reliability of detection; Dynamic energy management and detection integration: The energy management center globally optimizes and adjusts the droop control parameters of each control node according to the real-time status of distributed energy resources. The system can achieve optimal energy configuration while detecting the island state, ensuring continuous power supply to the power grid in the island state.

3. The droop control island detection system according to claim 2, characterized in that: The droop control algorithm in step 2 includes active power-frequency droop control and reactive power-voltage droop control; The formula for active power-frequency droop control is: f=f0-m P ·(P-P0) Where: f is the output frequency of the inverter; f0 is the rated frequency, usually 50Hz or 60Hz; m P is the active power droop coefficient; P is the current output active power of the inverter; P0 is the reference value of active power, which is the rated power; The formula for reactive power-voltage droop control is: V=V0-m Q ·(Q-Q0) Where V is the output voltage of the inverter; V0 is the rated voltage; m Q is the reactive power droop factor; Q is the current output reactive power of the inverter; Q0 is the reference value of reactive power, usually zero.

4. The droop control island detection system according to claim 3, characterized in that: The droop control parameters are adjusted dynamically. In order to adapt to the dynamic changes in the power grid, an adaptive algorithm is introduced to adjust the droop coefficient m. P and m Q Make real-time adjustments to improve the system's response speed and stability; formula: Adaptive active power droop coefficient adjustment: in, is the initial active power droop coefficient; k P It is the adaptive adjustment factor of active power droop coefficient; df / dt is the rate of change of frequency; Adaptive reactive power droop coefficient adjustment: in: is the initial reactive droop coefficient; k Q is the adaptive adjustment factor of reactive power droop coefficient; dV / dt is the rate of change of voltage.

5. The droop control island detection system according to claim 4, characterized in that: When multiple inverters are running together, the system can share the operating status of each inverter through the communication module and use the distributed algorithm to collaboratively optimize the droop control parameters; Optimization goal: Minimize frequency deviation: Minimize voltage deviation: Coordinated power distribution: and Under the premise that the total power meets the system requirements, the power distribution between the inverters should be optimal.

6. The droop control island detection system according to claim 5, characterized in that: The process of the droop control algorithm includes: S1. Initial setting: Each control node sets the initial droop coefficient and and perform droop control; S2. Real-time monitoring and adjustment: The node continuously monitors the output frequency f, voltage V, active power P, and reactive power Q parameters. When a parameter change is detected, an adaptive algorithm is used to adjust the droop coefficient m. P (t) and m Q (T); S3, Collaborative optimization: Through the communication module, each node shares the current power, frequency and voltage data, and the energy management center performs global optimization of the droop coefficient and power distribution to reduce frequency and voltage deviations and ensure system stability; S4, Feedback and Update: Each node continues to execute the control algorithm according to the optimized droop coefficient, and the system executes this process in a loop to achieve real-time dynamic adjustment.

7. The droop control island detection system according to claim 2, characterized in that: The island detection algorithm in step 3 adopts passive detection and active detection methods to improve the accuracy and response speed of island detection, which specifically includes: Passive detection method: By monitoring abnormal changes in grid parameters such as voltage, frequency and phase, it is possible to determine whether an island has occurred. Active detection method: detects islanding by injecting specific disturbance signals into the power grid and observing the system response.

8. The droop control island detection system according to claim 7, characterized in that: The passive detection algorithms include the following: 1) Under / overvoltage protection, algorithm steps: Voltage monitoring: real-time monitoring of the voltage V(t) at the inverter output end; Threshold judgment: Set the undervoltage threshold V min and overvoltage threshold V max ,like V(t) <V min or V(t)>V max , it is determined that an island occurs; 2) Under / over frequency protection, algorithm steps: Monitoring frequency: Real-time monitoring of inverter output frequency f(t) Threshold judgment: Set the underfrequency threshold f min and overfrequency threshold f max ,like f(t) <f min or f(t)>f max , then it is determined that an island occurs; 3) Frequency change rate, algorithm steps: Calculate the rate of change of frequency: Among them, Δt is the sampling interval time; Threshold judgment: Set the frequency change rate threshold like It is determined that an island occurs; 4) Phase jump detection, algorithm steps: Monitoring phase: real-time monitoring of voltage phase θ(t); Calculate the phase change: Δθ=θ(t)-θ(t-Δt) Threshold judgment: Set the phase jump threshold Δθ threshold ,like |Δθ|>Δθ threshold It is determined that an island has occurred.

9. The droop control island detection system according to claim 8, characterized in that: The active detection method injects specific disturbances into the power grid to observe the response of the system, including: (1) Active frequency drift, algorithm steps: Output current modulation: A sinusoidal modulation term is introduced into the inverter output current so that the current frequency is: f output =f0+Δf·sgn(sin(2πf0t)) Where Δf is the frequency offset, sgn is the sign function; Monitor frequency response: If the system is in grid-connected loading, the frequency rigidity of the grid keeps the frequency stable; in an islanded state, the frequency will gradually deviate; Threshold judgment: if the frequency deviation exceeds the set threshold, it is judged as an island; (2) Sandia frequency shift, algorithm steps: Output current modulation: Output current frequency: f output =f0+k·Δf·sin(2πf0t) Where k is the gain coefficient; Monitor frequency response: In an islanding state, the frequency can deviate rapidly, triggering protection.

10. The droop control island detection system according to claim 2, characterized in that: The data fusion algorithm mainly includes the following steps: Data preprocessing: Preliminary screening and preprocessing of the data sent by each control node to remove noise and outliers. formula: For each node i, the voltage data V i (t) Perform moving average processing: Where n is the smoothing window size; Use Kalman filtering to filter the frequency data f i (t) Processing: Where K(t) is the Kalman gain, is the frequency after filtering; Data weighting: The data is weighted according to the credibility or location of each node. The formula is: Let the weight factor w of each node i be i , then the weighted data is: Voltage: frequency: The calculation of the weighting factor is based on the following formula: Position weight, assuming the position weight w location,i Based on the distance d to the substation i : Historical accuracy weight, assuming that accuracy is based on the success rate p of the past N detections i : w accuracy，i =p i Total weight: w i =α·w location，i +β·w accuracy，i Among them, α and β are proportional coefficients of weighting factors, which can be adjusted according to actual needs; Data fusion: Combine the weighted data of each node, use weighted average, voting mechanism or Bayesian reasoning and other methods to comprehensively judge the island status. Formula: Integrated voltage: Comprehensive frequency: Consistency judgment: Based on the fusion results, determine whether the island event is confirmed and feed the results back to each node.

Images