Multi-island detection method and system based on multi-frequency-point characteristic frequency band impedance increment

Abstract

The invention discloses a multi-island detection method and system based on multi-frequency-point characteristic frequency band impedance increment. The method comprises the following steps: step 1, dividing frequency bands; step 2, carrying out passive island detection; when any one detection value is greater than the corresponding fluctuation early warning threshold but less than or equal to the corresponding fluctuation warning threshold, requesting to execute the step 3; step 3, triggering active island detection of a short time scale; the method comprises the following steps: injecting a broadband signal containing three different frequency point disturbance signals into a power distribution network, calculating difference values between power distribution network system impedance measurement values and corresponding historical values under the three different frequency point disturbance signals, and when the three difference values are all greater than a corresponding first island early warning threshold value and at least one difference value is smaller than a corresponding first island alarm threshold value, sending a warning signal to the power distribution network; if yes, requesting to execute step 4; 4, triggering long-time-scale active island detection; when the average values of the N system impedance difference values in the three frequency bands are all larger than a corresponding second alarm threshold value and the variable coefficients are all smaller than a variable coefficient threshold value, the island state is judged; and a detection blind area is eliminated.

Description

Technical Field

[0001] This invention belongs to the field of detection technology for grid-connected power generation systems of new energy, specifically involving a multi-islanding detection method and system based on impedance increment of multi-frequency characteristic bands in "high-voltage" distribution networks. Background Technology

[0002] With the rapid development of distributed generation technology, wind power, photovoltaic power, and other new energy power generation have gradually become an important part of modern power systems, and distribution networks are exhibiting "high voltage and high frequency" characteristics. However, when a distribution network experiences a fault or maintenance operation, the distributed generation system may lose connection with the main grid, forming an islanded operation. This state can lead to impaired power system stability, and abnormal voltage and frequency fluctuations may damage grid equipment and user equipment, while also posing a threat to the personal safety of maintenance personnel. Accurate and rapid detection of islanding can prevent power outages and power quality degradation caused by islanding operation, thereby improving the reliability of power supply.

[0003] In existing technologies, traditional islanding detection methods rely on abnormal changes in distribution network parameters, such as detection based on voltage and frequency changes and harmonic content changes. These methods suffer from large detection blind zones and difficulties in threshold tuning. Active islanding detection methods detect islanding states by injecting disturbance signals. Common methods, such as active frequency shifting, force the islanded system to become unstable by disturbing the frequency, thereby achieving islanding detection. However, this method affects the frequency stability of the power system, especially when multiple distributed power sources are operating simultaneously. Impedance measurement methods determine whether islanding has occurred by injecting a specific frequency signal into the power system and measuring the system's impedance changes. However, this method requires additional measurement equipment, and the injected disturbance signal and response signal are susceptible to background harmonics or "high-voltage and high-efficiency" distribution network conditions. Furthermore, under complex load conditions, this frequency signal may not excite sufficient impedance shift characteristics, reducing the reliability of islanding detection. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a multi-stage islanding detection method and system based on impedance increments in multiple frequency characteristic bands. The method determines characteristic frequency bands based on the impedance characteristics of common power electronic converters and typical equipment in high-voltage power distribution networks. Disturbance signals are injected using existing devices, and control strategies are provided for controlling these signals to achieve precise control. A triple detection method combining passive and active islanding detection is employed, with the active islanding detection using impedance information from multiple frequency characteristic bands to eliminate detection blind spots.

[0005] The present invention adopts the following technical solutions.

[0006] This invention proposes a multi-island detection method based on impedance increments in multiple frequency characteristic bands, comprising:

[0007] Step 1: Divide the impedance detection frequency band into three characteristic frequency bands: high, medium, and low, based on the characteristic impedance.

[0008] Step 2: Passive islanding detection is performed using the measured values ​​of voltage harmonic fluctuation, voltage amplitude fluctuation, and frequency fluctuation at the grid connection point.

[0009] If any detected value is greater than the corresponding fluctuation warning threshold but less than or equal to the corresponding fluctuation alarm threshold, request to execute step 4; if any detected value is greater than the corresponding fluctuation alarm threshold, determine the islanding status and issue an alarm; otherwise, continue to perform passive islanding detection.

[0010] Step 3: Trigger active islanding detection on a short time scale according to the set first time interval; inject a broadband disturbance signal containing three different frequency points located in three characteristic frequency bands; calculate the difference between the impedance measurement value and the historical value at the three frequency points; if all three differences are greater than the corresponding first islanding warning threshold but at least one is less than the corresponding first islanding alarm threshold, request to execute step 4; if all three differences are not less than the corresponding first islanding alarm threshold, determine the islanding status and issue an alarm; otherwise, wait for another trigger.

[0011] Step 4: Trigger active islanding detection on a long time scale according to the set second time interval or response request; inject broadband disturbance signals at N frequency points in each of the three frequency bands, and calculate the average value and coefficient of variation of the difference between the impedance measurement value and the historical value in each frequency band; when the average value in the three frequency bands is greater than the corresponding second alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, determine the islanding state and issue an alarm; otherwise, wait for another trigger.

[0012] In step 1, the low-frequency band corresponds to the synchronous control characteristics of the strain gauge, with values ​​of (10Hz, 50Hz); the mid-frequency band corresponds to the current loop and voltage loop control characteristics of the strain gauge, with values ​​of (300Hz, 500Hz); and the high-frequency band corresponds to the resonance characteristics of the power electronic equipment filter and line parameters, with values ​​of (1500Hz, 3000Hz).

[0013] In step 2, the fluctuation warning thresholds set for voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations are respectively: voltage harmonic fluctuation warning threshold, voltage amplitude fluctuation warning threshold, and frequency fluctuation warning threshold; the alarm thresholds set are respectively: voltage harmonic fluctuation alarm threshold, voltage amplitude fluctuation alarm threshold, and frequency fluctuation alarm threshold.

[0014] Step 3 includes:

[0015] Step 3.1: Trigger active island detection on a short time scale with a set first time period as the time interval;

[0016] Step 3.2: Select one non-integer frequency point from each of the low-frequency band, mid-frequency band, and high-frequency band, and inject the disturbance signal of the selected frequency point into the distribution network together with the broadband signal;

[0017] Step 3.3: Using the system impedance measurement values ​​of the distribution network in each frequency band obtained when the previous short-timescale active islanding detection was not detected as the historical system impedance values, calculate the system impedance difference between the system impedance measurement values ​​and the historical system impedance values ​​of the distribution network in each frequency band.

[0018] Step 3.4: Construct islanding warning thresholds and islanding alarm thresholds for different frequency bands using the system impedance amplitudes under rated operating conditions at three different frequency points.

[0019] Step 3.5: When the system impedance difference in each frequency band is greater than the corresponding islanding warning threshold, but the system impedance difference in at least one frequency band is less than the corresponding islanding alarm threshold, then request to execute step 4; when the system impedance difference in each frequency band is not less than the corresponding islanding alarm threshold, then determine that the system is in an islanded state and issue an alarm; otherwise, continue to perform active islanding detection on a short time scale.

[0020] The first time interval shall be no less than 15 seconds and no more than 5 minutes.

[0021] Step 3.4 includes:

[0022] 1) Within the low-frequency band, the system impedance amplitude Z under rated operating conditions at frequencies belonging to the low-frequency band. Af1 K f1 The islanding alarm threshold is doubled, and the islanding warning threshold is half of the islanding alarm threshold.

[0023] 2) Within the mid-frequency band, the system impedance amplitude Z under rated operating conditions at frequencies belonging to the mid-frequency band. Af2 K f2 The islanding alarm threshold is doubled, and the islanding warning threshold is half of the islanding alarm threshold.

[0024] 3) Within the high-frequency band, the system impedance amplitude Z under rated operating conditions at a frequency point belonging to the high-frequency band. Af3 K f3 The islanding alarm threshold is doubled, and the islanding warning threshold is half of the islanding alarm threshold.

[0025] Low-frequency band inner multiple K f1 The value is 0.2, and the multiple K is in the mid-frequency band. f2 The value is 0.4, and the multiple K is within the high-frequency band. f3 The value is 0.5.

[0026] Step 4 includes:

[0027] Step 4.1: Using the set second time period as the time interval, trigger long-term proactive island detection according to the set time interval or when a request is made to execute Step 4;

[0028] Step 4.2: Select N non-integer frequency points from each of the low-frequency, mid-frequency, and high-frequency bands, and inject the disturbance signals of the selected 3N frequency points together with the broadband signal into the distribution network, where N≥2;

[0029] Step 4.3: Using the system impedance measurement values ​​in each frequency band obtained when no islanding state was detected during the previous long-term active islanding detection, calculate the system impedance difference between the system impedance measurement values ​​and the system impedance history values ​​in each frequency band.

[0030] Step 4.4: Calculate the average value and coefficient of variation of the N system impedance differences in each frequency band;

[0031] Step 4.5: Construct alarm thresholds for different frequency bands using the average values ​​of system impedance amplitudes under rated operating conditions at N different frequency points, and set the coefficient of variation threshold at the same time.

[0032] Step 4.6: When the average value of N system impedance differences in any frequency band is greater than the corresponding alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, the islanding status judgment logic of that frequency band is determined to be valid; otherwise, the islanding status judgment logic of that frequency band is determined to be invalid.

[0033] If the islanding status judgment logic is valid for all frequency bands, the system is determined to be in an islanding state and an alarm is triggered; otherwise, it cannot be determined whether the system is in an islanding state and waits for the next time interval to end before triggering a long-term active islanding detection again.

[0034] The second time period shall be no less than 15 minutes and no more than 5 hours.

[0035] Step 4.5 includes:

[0036] 1) In the low-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the low-frequency band. K f1 The alarm threshold is multiplied by 1.

[0037] 2) Within the mid-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the mid-frequency band. K f2 The alarm threshold is multiplied by 1.

[0038] 3) Within the high-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the high-frequency band. K f3The alarm threshold is multiplied by 1.

[0039] 4) Within each frequency band, the set threshold value for the coefficient of variation is [0.3, 0.5].

[0040] A multi-amplitude controller is used to inject disturbance signals at 3 or 3N frequency points into the distribution network. The transfer function H(z) of the multi-amplitude controller is:

[0041]

[0042] In the formula, H(z) is the transfer function of the multi-amplitude controller, and N... i The parameters determine the minimum controllable frequency point, and F determines the subsequent controllable frequency points. For N i Order delay, k r H is the proportional coefficient of the controller. f (z) is the transfer function of the compensation stage.

[0043] This invention also proposes a multi-island detection system based on impedance increments in multiple frequency characteristic bands, comprising:

[0044] The disturbance signal generation module is used to divide the impedance detection frequency band of the distribution network system into three different characteristic frequency bands: high frequency band, medium frequency band and low frequency band, based on the characteristic impedance, and to generate disturbance signals at three different frequency points belonging to the three characteristic frequency bands.

[0045] The first detection module is used to perform passive islanding detection using the detection values ​​of the grid connection point. The detection values ​​include voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations. When any detection value is greater than the corresponding fluctuation warning threshold but less than or equal to the corresponding fluctuation alarm threshold, the second detection module is requested to work. When any detection value is greater than the corresponding fluctuation alarm threshold, the islanding state is determined and an alarm is triggered. Otherwise, passive islanding detection continues.

[0046] The second detection module is used to trigger short-timescale active islanding detection according to a set first time interval; it injects a broadband signal containing three different frequency disturbance signals into the distribution network, calculates the difference between the measured impedance value of the distribution network system under the three different frequency disturbance signals and the corresponding historical value, and requests the third detection module to work when all three differences are greater than the corresponding first islanding warning threshold but at least one difference is less than the corresponding first islanding alarm threshold; when all three differences are not less than the corresponding first islanding alarm threshold, it determines the islanding state and issues an alarm; otherwise, it continues to perform short-timescale active islanding detection.

[0047] The third detection module is used to actively trigger long-term islanding detection according to the set second time interval or response request; it injects broadband signals of N disturbance signals at three different frequency points into the distribution network, and calculates the average value and coefficient of variation of the system impedance difference between the measured system impedance value and the historical system impedance value in each frequency band; when the average value of the N system impedance differences in the three frequency bands is greater than the corresponding second alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, the islanding state is determined and an alarm is triggered; otherwise, it is impossible to determine whether the islanding state is in place, and the long-term active islanding detection is triggered again after the set time interval ends.

[0048] Among them, the multi-amplitude controller is used to inject disturbance signals at 3 or 3N frequency points into the power distribution network.

[0049] The present invention is also a terminal, including a processor and a storage medium; the storage medium is used to store instructions; the processor is used to perform operations according to the instructions to execute the steps of the method.

[0050] The present invention is also a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method.

[0051] The beneficial effects of this invention are as follows: Compared with the prior art, this invention employs a combination of passive and active detection to form a triple detection scheme for islanding conditions. The first level is based on passive monitoring of voltage harmonics, amplitude, and frequency; exceeding a threshold triggers the third level of detection. The second level, short-timescale active detection, performs islanding detection by injecting non-integer frequency disturbance signals from multiple frequency characteristic bands, triggering the third level of detection based on impedance increment thresholds or directly determining islanding. The third level, long-timescale active detection, confirms the status by scanning impedance at more frequencies. This triple detection hierarchy and collaborative verification reduces blind spots and false positives, improving the sensitivity and reliability of islanding detection. During the second and third levels of active detection, this invention simultaneously injects non-integer frequency disturbance signals from multiple frequency characteristic bands into the system under test. These disturbance signals do not interfere with the power frequency signal, have minimal impact on power grid power quality, and the response signal is unaffected by power grid harmonic interference. This avoids interference from background harmonics in the power system and comprehensively utilizes impedance results from multiple frequency bands for islanding detection. Furthermore, this invention only requires improvements to the control strategy of existing equipment, without the need for additional complex hardware, reducing system costs and implementation difficulty. Attached Figure Description

[0052] Figure 1 The flowchart shows a multi-islanding detection method based on impedance increment of multiple frequency bands of "dual-high" distribution networks proposed in this invention.

[0053] Figure 2This is a schematic diagram of a multi-island detection system based on impedance increment of multiple frequency bands of "dual-high" distribution networks proposed in this invention.

[0054] Figure 3 This is a control block diagram of the converter when the harmonic signal injection function is reused in an embodiment of the present invention;

[0055] Figure 4 This is the system equivalent impedance diagram when harmonic disturbances are injected in an embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0057] When facing the power grid under test, the active islanding detection system proposed in this invention is configured as follows: Figure 1 As shown, the distribution network under test includes: distribution network, load, n sets of new energy devices, n new energy converters AC / DC, reactive power compensation device, and 1 reactive power compensation converter AC / DC, etc.

[0058] This invention proposes a multi-level islanding detection method based on impedance increments in multiple frequency bands characteristic of "high-voltage" distribution networks. It combines a passive islanding detection method based on voltage harmonics, amplitude fluctuations, and frequency fluctuations with an active islanding detection method based on disturbance signal injection, forming a triple-level islanding detection method.

[0059] like Figure 1 As shown, the method includes the following steps:

[0060] Step 1: Based on the characteristic impedance, the impedance detection frequency band of the distribution network system is divided into three different characteristic frequency bands: high frequency band, medium frequency band, and low frequency band, which are used to generate three different frequency disturbance signals belonging to the three characteristic frequency bands.

[0061] Specifically, the low-frequency band corresponds to the synchronous control characteristics of the strain gauge, with values ​​of (10Hz, 50Hz); the mid-frequency band corresponds to the current loop and voltage loop control characteristics of the strain gauge, with values ​​of (300Hz, 500Hz); and the high-frequency band corresponds to the resonance characteristics of the power electronic equipment filter and line parameters, with values ​​of (1500Hz, 3000Hz). The low-frequency, mid-frequency, and high-frequency bands constitute a wideband.

[0062] This invention first divides the characteristic frequency bands, following the principles of characteristic impedance matching and dynamic characteristic coverage. The frequency bands are divided based on the impedance characteristics of the "high-voltage, high-frequency" distribution network, ensuring coverage of the network's wideband dynamic characteristics. By analyzing the synchronous control characteristics, converter current and voltage loop control characteristics, and the resonance characteristics of lines and filters in the distribution network system, the impedance detection frequency band is divided into multiple characteristic frequency bands corresponding to the equipment characteristics, achieving adaptive frequency band division based on physical characteristics. Specifically, the low-frequency band corresponds to the converter synchronous control characteristics, with values ​​below 50Hz; the mid-frequency band corresponds to the converter current loop and voltage loop control characteristics, with a frequency range set around 1 / 20 of the switching frequency; and the high-frequency band corresponds to the resonance characteristics of power electronic equipment filters and line parameters, with values ​​ranging from 1 / 6 to 1 / 3 of the switching frequency. In the embodiments, the preferred range of characteristic frequency bands includes: (10Hz, 50Hz) for the low-frequency band, (300Hz, 500Hz) for the mid-frequency band, and (1500Hz, 3000Hz) for the high-frequency band. Based on the typical characteristics of low, medium and high frequency bands, we can determine the selection of frequency points and characteristic frequency bands, and also determine the detection threshold.

[0063] Step 2: Perform passive islanding detection using the detection values ​​from the grid connection point. The detection values ​​include voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations.

[0064] When any detected value is greater than the corresponding fluctuation warning threshold but less than or equal to the corresponding fluctuation alarm threshold, step 3 is requested to be executed; when any detected value is greater than the corresponding fluctuation alarm threshold, the islanding status is determined and an alarm is issued; otherwise, passive islanding detection continues.

[0065] Specifically, passive islanding detection is performed using voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations at the grid connection point. When the detected value exceeds the set fluctuation warning threshold, step 4 is requested to be executed. When the detected value exceeds the set fluctuation alarm threshold, the islanding status is determined and an alarm is issued. Otherwise, passive islanding detection continues.

[0066] like Figure 2 As shown, the passive islanding detection implemented in step 2 is the first-level islanding detection scheme. It is achieved through a passive detection module based on voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations. It monitors in real time whether the total harmonic distortion, amplitude fluctuation, and frequency fluctuation deviation of the grid connection point voltage exceed their respective set thresholds. The fluctuation warning thresholds set for voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations are respectively: voltage harmonic fluctuation warning threshold, voltage amplitude fluctuation warning threshold, and frequency fluctuation warning threshold; the set alarm thresholds are respectively: voltage harmonic fluctuation alarm threshold, voltage amplitude fluctuation alarm threshold, and frequency fluctuation alarm threshold.

[0067] Request to execute step 4 if any of the following conditions are met:

[0068] 1) Voltage harmonic fluctuations exceed the voltage harmonic fluctuation warning threshold;

[0069] 2) Voltage amplitude fluctuation exceeds the voltage amplitude fluctuation warning threshold;

[0070] 3) Frequency fluctuations exceed the frequency fluctuation warning threshold;

[0071] When none of the above conditions are met, passive island detection continues.

[0072] An island status is determined and an alarm is triggered when any of the following conditions are met:

[0073] 1) Voltage harmonic fluctuations exceed the voltage harmonic fluctuation alarm threshold;

[0074] 2) Voltage amplitude fluctuation exceeds the voltage amplitude fluctuation alarm threshold;

[0075] 3) Frequency fluctuation exceeds the frequency fluctuation alarm threshold;

[0076] This invention acquires the upper limits of voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations within permissible ranges caused by changes in renewable energy output. The alarm thresholds for voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations are all set to be greater than the upper limits of these limits. The invention proposes setting these alarm thresholds to be greater than the fluctuation values ​​caused by changes in renewable energy output to prevent false islanding detection due to these fluctuations. Furthermore, each warning threshold is set to half of its corresponding alarm threshold to reduce the blind zone in islanding detection. In the embodiment, the alarm threshold for total harmonic distortion of the grid connection point voltage ranges from 5% to 10%, the alarm threshold for amplitude fluctuation ranges from ±10% to ±15% of the rated voltage, and the alarm threshold for frequency fluctuation deviation ranges from 0.5Hz to 1Hz.

[0077] Step 3: Trigger short-timescale active islanding detection according to the set first time interval; inject a broadband signal containing three different frequency disturbance signals into the distribution network, calculate the difference between the measured impedance value of the distribution network system under the three different frequency disturbance signals and the corresponding historical value; if all three differences are greater than the corresponding first islanding warning threshold but at least one difference is less than the corresponding first islanding alarm threshold, then request to execute step 4; if all three differences are not less than the corresponding first islanding alarm threshold, then determine the islanding status and issue an alarm; otherwise, continue short-timescale active islanding detection.

[0078] Specifically, according to the set time interval, short-timescale active islanding detection is triggered; a broadband signal containing three different frequency disturbance signals is injected into the distribution network, with each disturbance signal belonging to a different frequency band; the system impedance difference between the measured system impedance value and the historical system impedance value in each frequency band is calculated; when the system impedance difference in each frequency band is greater than the corresponding islanding warning threshold but the system impedance difference in at least one frequency band is less than the corresponding islanding alarm threshold, step 4 is requested to be executed; when the system impedance difference in each frequency band is not less than the corresponding islanding alarm threshold, the system is determined to be in an islanding state and an alarm is triggered.

[0079] Specifically, the second island detection method implemented in step 3 is a short-timescale active detection scheme, including:

[0080] Step 3.1: Trigger short-timescale active island detection at a set first time period T2; wherein the first time period T2 is not less than 15 seconds and not more than 5 minutes; because the short-timescale active island detection injects fewer frequency points and has little impact on the system, it is injected frequently.

[0081] Step 3.2: Select one non-integer frequency point from each of the low-frequency band, mid-frequency band, and high-frequency band, and inject the disturbance signal of the selected frequency point into the distribution network together with the broadband signal;

[0082] In this embodiment, when determining the frequency of the disturbance signal, non-integer frequency points are selected in each characteristic frequency band to avoid interference from background harmonics. In this embodiment, in the low frequency band, frequencies such as 22.8Hz and 35.6Hz can be selected, and in the mid frequency band, frequencies such as 371.3Hz and 442.6Hz can be selected.

[0083] Step 3.3: Using the system impedance measurement values ​​of the distribution network in each frequency band obtained when the previous short-timescale active islanding detection was not detected as the historical system impedance values, calculate the system impedance difference between the system impedance measurement values ​​and the historical system impedance values ​​of the distribution network in each frequency band.

[0084] ΔZ f1 =Z f1 -Z′ f1 ΔZ f2 =Z f2 -Z′ f2 ΔZ f3 =Z f3 -Z′ f3

[0085] In the formula, Z f1 , Z f2 , Z f3These are the system impedance measurements (Z′, Z′, Z′) obtained during this short-timescale active islanding detection exercise, representing the low-frequency, mid-frequency, and high-frequency bands. f1 Z′ f2 Z′ f3 These are the system impedance measurements (ΔZ) obtained during the previous short-timescale active islanding detection, representing the low-frequency, mid-frequency, and high-frequency bands, respectively. f1 ΔZ f2 ΔZ f3 These are the system impedance differences in the low-frequency, mid-frequency, and high-frequency bands calculated during this short-timescale active island detection.

[0086] Step 3.4: Construct islanding warning thresholds and islanding alarm thresholds for different frequency bands using the system impedance amplitudes under rated operating conditions at three different frequency points, including:

[0087] 1) Within the low-frequency band, the system impedance amplitude Z under rated operating conditions at frequencies belonging to the low-frequency band. Af1 K f1 The multiple is the islanding alarm threshold, and the islanding early warning threshold is half of the islanding alarm threshold; in the embodiment, the multiple K in the low frequency band f1 The value is 0.2;

[0088] 2) Within the mid-frequency band, the system impedance amplitude Z under rated operating conditions at frequencies belonging to the mid-frequency band. Af2 K f2 The multiple is the islanding alarm threshold, and the islanding early warning threshold is half of the islanding alarm threshold; in the embodiment, the multiple K in the mid-frequency band f2 The value is 0.4;

[0089] 3) Within the high-frequency band, the system impedance amplitude Z under rated operating conditions at a frequency point belonging to the high-frequency band. Af3 K f3 The multiple is the islanding alarm threshold, and the islanding early warning threshold is half of the islanding alarm threshold; in the embodiment, the multiple K in the high-frequency band f3 The value is 0.5;

[0090] The detection method proposed in this invention uses the impedance amplitude at each frequency point under rated operating conditions as a benchmark, follows the principle of differentiated setting for different frequency bands, and sets thresholds for different characteristic frequency bands. Based on the operating characteristics and interference risk level of the "high-voltage and high-frequency" distribution network in each characteristic frequency band, the thresholds are set to achieve a balance between detection sensitivity and reliability.

[0091] The low-frequency band corresponds to the synchronous control characteristics of the distribution network. The multiples in the threshold of the low-frequency band are set relatively small because low-frequency disturbances are prone to affect the synchronous control link of the system. A smaller multiple can capture the frequency offset and phase instability characteristics in the early stage of islanding in advance, and avoid detection lag caused by synchronous loss of control.

[0092] The control bandwidth of the current loop and voltage loop of the converter in the mid-frequency band is associated with the distribution network. The multiple in the mid-frequency band threshold is set to a medium value. The mid-frequency band covers the sensitive range of load dynamic response and converter control parameters. The medium multiple can filter out the interference of conventional load fluctuations and accurately identify the sudden changes in current and voltage control characteristics under islanded conditions.

[0093] The high-frequency band corresponds to the filter parameters and line parameters of power electronic equipment. The multiplier in the high-frequency band threshold is set relatively large. Since high-frequency disturbances can easily induce system resonance, a larger multiplier can avoid misjudgment caused by high-frequency oscillations or transient processes in non-islanded scenarios. Islanding is only determined when the impedance increment significantly exceeds the resonance characteristic threshold.

[0094] Step 3.5: When the system impedance difference in each frequency band is greater than the corresponding islanding warning threshold, but the system impedance difference in at least one frequency band is less than the corresponding islanding alarm threshold, then request to execute step 4; when the system impedance difference in each frequency band is not less than the corresponding islanding alarm threshold, then determine that the system is in an islanded state and issue an alarm; otherwise, continue to perform active islanding detection on a short time scale.

[0095] Specifically, in the second-level islanding detection scheme, the islanding alarm threshold is higher than the islanding warning threshold within the same characteristic frequency band, and the islanding warning threshold and islanding alarm threshold are also different for different characteristic frequency bands. The islanding warning threshold is a low threshold, serving as a preliminary warning signal. When the system impedance difference at each frequency point exceeds the islanding warning threshold, it indicates that the system may be in an islanding edge state. However, the detection results at a single short time scale and three frequency points are uncertain; for example, transient interference or load fluctuations may cause false alarms. Therefore, the third-level islanding detection scheme is triggered for further verification. The islanding alarm threshold is a high threshold, serving as a deterministic criterion. When the system impedance at all three frequency points exceeds the islanding alarm threshold, it indicates that the system impedance has undergone a significant change, and the islanding characteristics are sufficiently clear. The islanding state can be directly determined without additional detection, thus shortening the detection time and preventing the accident from escalating.

[0096] Step 4: Actively trigger long-term islanding detection according to the set second time interval or response request; inject broadband signals of N disturbance signals at three different frequency points into the distribution network, and calculate the average value and coefficient of variation of the system impedance difference between the measured system impedance value and the historical system impedance value in each frequency band; when the average value of the N system impedance differences in the three frequency bands is greater than the corresponding second alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, the islanding state is determined and an alarm is triggered; otherwise, it is impossible to determine whether the islanding state is in place, and the long-term active islanding detection is triggered again after the set time interval ends.

[0097] Specifically, at a set time interval or when the execution of step 4 is requested, trigger the long-time-scale active islanding detection; inject a wideband signal containing 3N disturbance signals with different frequency points into the distribution network, where the frequency points of N disturbance signals belong to the same frequency band, and calculate the average value and coefficient of variation of the system impedance difference between the system impedance measurement value and the system impedance historical value in each frequency band respectively; when the average value of the N system impedance differences in any frequency band is greater than the corresponding alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, it is determined that the islanding state judgment logic of this frequency band is valid, otherwise it is determined that the islanding state judgment logic of this frequency band is invalid;

[0098] When the islanding state judgment logic of all frequency bands is valid, it is determined that the islanding state exists and an alarm is issued; otherwise, it cannot be determined whether the islanding state exists, and the long-time-scale active islanding detection is triggered again at the end of the set time interval.

[0099] Specifically, the long-time-scale active islanding detection method implemented in step 4 is the third-layer islanding detection scheme, including:

[0100] Step 4.1, with a set second time period T3 as the time interval, at a set time interval or when the execution of step 4 is requested, trigger the long-time-scale active islanding detection; where the second time period is not less than 15 minutes and not more than 5 hours;

[0101] Specifically, the third-layer islanding detection scheme is requested to be executed by the first-layer islanding detection result or the second-layer islanding detection result or triggered to execute according to the interval time period T3, and T2 << T3. Compared with the second-layer islanding detection, the third-layer islanding detection will measure the system impedance increment of more frequency points to more comprehensively and reliably analyze the state of the distribution network.

[0102] Step 4.2, select N non-integer frequency points from each of the low-frequency band, medium-frequency band and high-frequency band, and inject the disturbance signals of the selected 3N frequency points into the distribution network together with the wideband signal, where N ≥ 2;

[0103] In the embodiment, when determining the frequency of the disturbance signal, non-integer frequency points are selected in each characteristic frequency band to avoid the interference of background harmonics. In the embodiment, in the low-frequency band, frequency points such as 22.8 Hz and 35.6 Hz can be selected, and in the medium-frequency band, frequency points such as 371.3 Hz and 442.6 Hz can be selected.

[0104] Step 4.3, taking the system impedance measurement values in each frequency band obtained when the long-time-scale active islanding detection was last executed but no islanding state was detected as the system impedance historical values, and calculate the system impedance differences between the system impedance measurement values and the system impedance historical values in each frequency band respectively;

[0105] {ΔZ f1,i}={Z f1,i}-{Z′ f1,i},{ΔZ f2,i}={Z f2,i}-{Z′ f2,i},{ΔZ f3,i}={Z f3,i}-{Z′ f3,i}

[0106] In the formula, i = 1, 2, ..., N, {Z f1,i}、{Z f2,i}、{Z f3,i} represents the set of system impedance measurements in the low-frequency, mid-frequency, and high-frequency bands obtained during this long-term active islanding detection operation, respectively. f1,i}、{Z′ f2,i}、{Z′ f3,i} represents the set of system impedance measurements in the low-frequency, mid-frequency, and high-frequency bands obtained during the last long-term active islanding detection, {ΔZ} f1,i}、{ΔZ f2,i}、{ΔZ f3,i} represents the set of system impedance differences in the low-frequency, mid-frequency, and high-frequency bands calculated during this long-term active island detection.

[0107] Step 4.4: Calculate the average value and coefficient of variation of the N system impedance differences in each frequency band;

[0108] In the embodiments, These represent the average system impedance differences of the distribution network under frequency disturbance signals in the low-frequency, mid-frequency, and high-frequency bands, calculated during this long-term active islanding detection exercise. The coefficient of variation (COP) is the ratio of the standard deviation to the mean, used to measure the relative dispersion of the data.

[0109] Step 4.5: Construct alarm thresholds for different frequency bands using the average system impedance amplitude at N different frequency points under rated operating conditions, and simultaneously set the coefficient of variation threshold.

[0110] 1) In the low-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the low-frequency band. K f1 The multiple is the alarm threshold. In the example, the multiple K in the low-frequency band is... f1 The value is 0.2;

[0111] 2) Within the mid-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the mid-frequency band. K f2The multiple is the alarm threshold. In the example, the multiple K in the mid-frequency band is... f2 The value is 0.4;

[0112] 3) Within the high-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the high-frequency band. K f3 The multiple is the alarm threshold. In the example, the multiple K in the mid-frequency band is... f3 The value is 0.5;

[0113] 4) Within each frequency band, the set value range of the coefficient of variation threshold is [0.3, 0.5]. In the embodiment, the preferred value of the coefficient of variation threshold is 0.3.

[0114] Step 4.6: When the average value of N system impedance differences in any frequency band is greater than the corresponding alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, the islanding status judgment logic of that frequency band is determined to be valid; otherwise, the islanding status judgment logic of that frequency band is determined to be invalid.

[0115] If the islanding status judgment logic is valid for all frequency bands, the system is determined to be in an islanding state and an alarm is triggered; otherwise, it cannot be determined whether the system is in an islanding state and waits for the next time interval to end before triggering a long-term active islanding detection again.

[0116] The third-level islanding detection logic involves multiple characteristic frequency bands, and each characteristic frequency band contains multiple frequency points. Compared to the first and second levels of islanding detection, the third-level islanding detection is comprehensive and reliable. On the one hand, due to the large fluctuations in the "high-voltage and high-efficiency" distribution network, the first-level detection has a higher probability of misjudgment. Therefore, the first and third levels of detection are combined for verification. On the other hand, the second-level detection only injects a disturbance signal at one frequency point in each frequency band, which may result in misjudgment. Therefore, when the second-level detection determines that the system is in an islanding warning state, the third-level detection is triggered. At this time, the disturbance control strategy described later adaptively adjusts when switching between the second and third levels of detection. By adjusting the control parameters, the disturbance signal injected into the third-level detection is implemented to achieve the third-level detection and avoid misjudgment.

[0117] In the triple detection method proposed in this invention, the second and third detections are both active islanding detections. These methods simultaneously consider the impedance variation characteristics of multiple frequency bands. These multiple frequency bands can effectively reflect the dynamic characteristics of typical loads, power electronic converters, and typical equipment in high-voltage power distribution networks, avoiding detection blind spots and improving the reliability of islanding detection. For example, if the operating point of a converter in the system changes significantly or a converter disconnects, the system impedance will change significantly in a single frequency band. Determining the system to be islanded under these circumstances would be unreliable.

[0118] To implement the islanding detection method proposed in this invention and inject disturbance signals at multiple frequency points into the distribution network, this invention also proposes a control strategy for impedance measurement in characteristic frequency bands. This strategy implements a generalized amplitude control, which can accurately inject disturbance signals at multiple frequency points according to the requirements of different detection schemes, thereby achieving accurate measurement of the system impedance increment at each frequency point. The injection of disturbance signals is achieved through the redundancy capacity of existing equipment in the system. The control strategy includes:

[0119] The port voltage v of the disturbance injection device is collected. a v b v c and port current i a 、i b 、i c The fundamental signal and disturbance signal are separated from the port voltage and current signals through a signal separation module; wherein, the signal separation module includes 3 or 3N parallel bandpass filters (BPFs);

[0120] The fundamental frequency signal is input to the fundamental frequency control module; the fundamental frequency control module controls the fundamental frequency signal to achieve the normal function of the disturbance injection device;

[0121] In the embodiments, when the disturbance injection device is a new energy device, the fundamental wave control module implements the control function of the new energy converter according to the fundamental wave signal; or when the disturbance injection device is a reactive power compensation device, the fundamental wave control module implements the control function of the reactive power compensation converter according to the fundamental wave signal.

[0122] Specifically, Figure 3 In the process, the three-phase fundamental current signal i′ is obtained by separation. a 、i′ b 、i′ c The three-phase fundamental voltage signal v′ obtained by separation a v′ b v′ c and DC bus voltage V dc The input is sent to the fundamental frequency control module; the fundamental frequency control module outputs the fundamental frequency signal modulation amount m′. a m′ b m′ c The disturbance signal is input to the disturbance signal control module, which controls the disturbance signal to achieve tracking of the disturbance signal reference value. The disturbance signal control module includes three multi-amplitude controllers, and the three-phase disturbance current reference value i... parefk 、i pbrefk 、i pcrefk The three-phase disturbance current signal i obtained by separation pa 、i pb 、i pcThe differences are respectively input to the corresponding multi-amplitude controllers, and the multi-amplitude controllers output the disturbance signal modulation amount m of the disturbance signal. pa m pb m pc ;

[0123] Among them, a multi-amplitude controller is used to control the disturbance signals of 3 or 3N frequency points in multiple characteristic frequency bands injected into the distribution network. The transfer function H(z) of the multi-amplitude controller is:

[0124]

[0125] In the formula, H(z) is the transfer function of the multi-amplitude controller, and N... i The parameters determine the minimum controllable frequency point, and F determines the subsequent controllable frequency points. For N i Order delay, k r H is the proportional coefficient of the controller. f (z) is the transfer function of the compensation stage.

[0126] The amplitude characteristics of H(z) exhibit high amplitude at multiple frequencies, allowing the reference value to be accurately followed.

[0127] After the fundamental signal modulation amount and the disturbance signal modulation amount are superimposed, the trigger pulses of power devices S1-S6 are obtained through PWM modulation, so that the disturbance injection device can simultaneously inject disturbance current signals of multiple frequencies into the system.

[0128] When a disturbance signal is injected using a reactive power compensation device or a new energy converter, the system equivalent impedance is as follows: Figure 4 As shown. Where Z g Z is the equivalent impedance of the power grid. Load Let Z1, ..., Z be the equivalent impedance of the load. m-1 Z m ,…,Z n Z represents the equivalent impedance of each new energy converter. v This is the equivalent impedance of the reactive power compensation device. When using a reactive power compensation device to inject disturbances, the device is equivalent to a disturbance current source connected in parallel with its internal resistance. When using renewable energy source m to inject disturbances, the converter is equivalent to a disturbance current source connected in parallel with its internal resistance. According to the equivalent impedance diagram, the measured impedance is the parallel impedance of all devices except the disturbance injection device. When the grid connection fails, the impedance Z... g When disconnected from the system, the impedance measurement will increase, which can be used as a basis for determining whether the system is in an islanded state.

[0129] If the operating conditions of the converter in a high-voltage distribution network change, its corresponding impedance value will also change; even when the converter is disconnected from the distribution network, the impedance measurement value will change significantly. The converter impedance and the grid impedance differ significantly in each frequency band. Therefore, detecting impedance in multiple characteristic frequency bands will help avoid misjudgment of islanding conditions.

[0130] This invention also proposes a multi-island detection system based on impedance increments in multiple frequency characteristic bands, comprising:

[0131] The disturbance signal generation module is used to divide the impedance detection frequency band of the distribution network system into three different characteristic frequency bands: high frequency band, medium frequency band and low frequency band, based on the characteristic impedance, and to generate disturbance signals at three different frequency points belonging to the three characteristic frequency bands.

[0132] The first detection module is used to perform passive islanding detection using the detection values ​​of the grid connection point. The detection values ​​include voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations. When any detection value is greater than the corresponding fluctuation warning threshold but less than or equal to the corresponding fluctuation alarm threshold, the second detection module is requested to work. When any detection value is greater than the corresponding fluctuation alarm threshold, the islanding state is determined and an alarm is triggered. Otherwise, passive islanding detection continues.

[0133] The second detection module is used to trigger short-timescale active islanding detection according to a set first time interval; it injects a broadband signal containing three different frequency disturbance signals into the distribution network, calculates the difference between the measured impedance value of the distribution network system under the three different frequency disturbance signals and the corresponding historical value, and requests the third detection module to work when all three differences are greater than the corresponding first islanding warning threshold but at least one difference is less than the corresponding first islanding alarm threshold; when all three differences are not less than the corresponding first islanding alarm threshold, it determines the islanding state and issues an alarm; otherwise, it continues to perform short-timescale active islanding detection.

[0134] The third detection module is used to actively trigger long-term islanding detection according to the set second time interval or response request; it injects broadband signals of N disturbance signals at three different frequency points into the distribution network, and calculates the average value and coefficient of variation of the system impedance difference between the measured system impedance value and the historical system impedance value in each frequency band; when the average value of the N system impedance differences in the three frequency bands is greater than the corresponding second alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, the islanding state is determined and an alarm is triggered; otherwise, it is impossible to determine whether the islanding state is in place, and the long-term active islanding detection is triggered again after the set time interval ends.

[0135] The system also includes a multi-amplitude controller, used to inject disturbance signals at 3 or 3N frequency points into the power distribution network.

[0136] In the triple detection proposed in this invention, the first detection operates continuously. While its continuous operation does not negatively impact the system, it has a higher probability of false alarms due to the significant fluctuations in high-voltage and high-efficiency power distribution networks. Both the second and third detections inject disturbance signals into the system, negatively impacting it. The second detection injects less disturbance signal, thus operating at shorter intervals, but it may still misdiagnose islanding issues. The third detection injects disturbance signals at more frequencies, resulting in a greater negative impact, hence requiring longer intervals. The third detection has high reliability; therefore, when the first and second detections indicate the system may be in an islanded state, the third detection is used to determine the system status.

[0137] This disclosure can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of this disclosure.

[0138] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0139] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0140] Computer program instructions used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, status setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing the status information of the computer-readable program instructions to implement various aspects of this disclosure.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A method for detecting multiple islands based on impedance increments in multiple frequency characteristic bands, characterized in that, include: Step 1: Divide the impedance detection frequency band into three characteristic frequency bands: high, medium, and low, based on the characteristic impedance. Step 2: Passive islanding detection is performed using the detected values ​​of voltage harmonic fluctuation, voltage amplitude fluctuation, and frequency fluctuation at the grid connection point. If any detected value is greater than the corresponding fluctuation warning threshold but less than or equal to the corresponding fluctuation alarm threshold, request to execute step 4; if any detected value is greater than the corresponding fluctuation alarm threshold, determine the islanding state and issue an alarm. Otherwise, continue with passive island detection; Step 3: Trigger active islanding detection on a short time scale according to the set first time interval; inject a broadband disturbance signal containing three different frequency points located in three characteristic frequency bands; calculate the difference between the impedance measurement value and the historical value at the three frequency points; if all three differences are greater than the corresponding first islanding warning threshold but at least one is less than the corresponding first islanding alarm threshold, request to execute step 4; if all three differences are not less than the corresponding first islanding alarm threshold, determine the islanding status and issue an alarm; otherwise, wait for another trigger. Step 4: Trigger active islanding detection on a long time scale according to the set second time interval or response request; inject broadband disturbance signals at N frequency points in each of the three frequency bands, and calculate the average value and coefficient of variation of the difference between the impedance measurement value and the historical value in each frequency band; when the average value in the three frequency bands is greater than the corresponding second alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, determine the islanding state and issue an alarm; otherwise, wait for another trigger.

2. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 1, characterized in that, In step 1, the low-frequency band corresponds to the synchronous control characteristics of the strain gauge, with values ​​of (10Hz, 50Hz); the mid-frequency band corresponds to the current loop and voltage loop control characteristics of the strain gauge, with values ​​of (300Hz, 500Hz); and the high-frequency band corresponds to the resonance characteristics of the power electronic equipment filter and line parameters, with values ​​of (1500Hz, 3000Hz).

3. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 1, characterized in that, In step 2, the fluctuation warning thresholds set for voltage harmonic fluctuation, voltage amplitude fluctuation, and frequency fluctuation are respectively: voltage harmonic fluctuation warning threshold, voltage amplitude fluctuation warning threshold, and frequency fluctuation warning threshold. The alarm thresholds set are: voltage harmonic fluctuation alarm threshold, voltage amplitude fluctuation alarm threshold, and frequency fluctuation alarm threshold.

4. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 1, characterized in that, Step 3 includes: Step 3.1: Trigger active island detection on a short time scale with a set first time period as the time interval; Step 3.2: Select one non-integer frequency point from each of the low-frequency band, mid-frequency band, and high-frequency band, and inject the disturbance signal of the selected frequency point into the distribution network together with the broadband signal; Step 3.3: Using the system impedance measurement values ​​of the distribution network in each frequency band obtained when the previous short-timescale active islanding detection was not detected as the historical system impedance values, calculate the system impedance difference between the system impedance measurement values ​​and the historical system impedance values ​​of the distribution network in each frequency band. Step 3.4: Construct islanding warning thresholds and islanding alarm thresholds for different frequency bands using the system impedance amplitudes under rated operating conditions at three different frequency points. Step 3.5: When the system impedance difference in each frequency band is greater than the corresponding islanding warning threshold, but the system impedance difference in at least one frequency band is less than the corresponding islanding alarm threshold, then request to execute step 4; when the system impedance difference in each frequency band is not less than the corresponding islanding alarm threshold, then determine that the system is in an islanded state and issue an alarm; otherwise, continue to perform active islanding detection on a short time scale.

5. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 4, characterized in that, The first time interval shall be no less than 15 seconds and no more than 5 minutes.

6. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 4, characterized in that, Step 3.4 includes: 1) Within the low-frequency band, the system impedance amplitude Z under rated operating conditions at frequencies belonging to the low-frequency band. Af1 K f1 The islanding alarm threshold is doubled, and the islanding warning threshold is half of the islanding alarm threshold. 2) Within the mid-frequency band, the system impedance amplitude Z under rated operating conditions at frequencies belonging to the mid-frequency band. Af2 K f2 The islanding alarm threshold is doubled, and the islanding warning threshold is half of the islanding alarm threshold. 3) Within the high-frequency band, the system impedance amplitude Z under rated operating conditions at a frequency point belonging to the high-frequency band. Af3 K f3 The islanding alarm threshold is doubled, and the islanding warning threshold is half of the islanding alarm threshold.

7. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 6, characterized in that, Low-frequency band inner multiple K f1 The value is 0.2, and the multiple K is in the mid-frequency band. f2 The value is 0.4, and the multiple K is within the high-frequency band. f3 The value is 0.

5.

8. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 1, characterized in that, Step 4 includes: Step 4.1: Using the set second time period as the time interval, trigger long-term proactive island detection according to the set time interval or when a request is made to execute Step 4; Step 4.2: Select N non-integer frequency points from each of the low-frequency, mid-frequency, and high-frequency bands, and inject the disturbance signals of the selected 3N frequency points together with the broadband signal into the distribution network, where N≥2; Step 4.3: Using the system impedance measurement values ​​in each frequency band obtained when no islanding state was detected during the previous long-term active islanding detection, calculate the system impedance difference between the system impedance measurement values ​​and the system impedance history values ​​in each frequency band. Step 4.4: Calculate the average value and coefficient of variation of the N system impedance differences in each frequency band; Step 4.5: Construct alarm thresholds for different frequency bands using the average values ​​of system impedance amplitudes under rated operating conditions at N different frequency points, and set the coefficient of variation threshold at the same time. Step 4.6: When the average value of N system impedance differences in any frequency band is greater than the corresponding alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, the islanding status judgment logic of that frequency band is determined to be valid; otherwise, the islanding status judgment logic of that frequency band is determined to be invalid. If the islanding status judgment logic is valid for all frequency bands, the system is determined to be in an islanding state and an alarm is triggered; otherwise, it cannot be determined whether the system is in an islanding state and waits for the next time interval to end before triggering a long-term active islanding detection again.

9. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 8, characterized in that, The second time period shall be no less than 15 minutes and no more than 5 hours.

10. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 8, characterized in that, Step 4.5 includes: 1) In the low-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the low-frequency band. K f1 The alarm threshold is multiplied by 1. 2) Within the mid-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the mid-frequency band. K f2 The alarm threshold is multiplied by 1. 3) Within the high-frequency band, the average value of the system impedance amplitude under rated operating conditions at N frequency points belonging to the high-frequency band. K f3 The alarm threshold is multiplied by 1. 4) Within each frequency band, the set threshold value for the coefficient of variation is [0.3, 0.5].

11. The multi-island detection method based on impedance increment of multi-frequency characteristic bands according to claim 1, characterized in that, A multi-amplitude controller is used to inject disturbance signals at 3 or 3N frequency points into the distribution network. The transfer function H(z) of the multi-amplitude controller is: In the formula, H(z) is the transfer function of the multi-amplitude controller, and N... i The parameters determine the minimum controllable frequency point, and F determines the subsequent controllable frequency points. For N i Order delay, k r H is the proportional coefficient of the controller. f (z) is the transfer function of the compensation stage.

12. A multi-island detection system based on impedance increments of multiple frequency characteristic bands, used to implement the multi-island detection method based on impedance increments of multiple frequency characteristic bands as described in any one of claims 1 to 11, characterized in that, include: The disturbance signal generation module is used to divide the impedance detection frequency band of the distribution network system into three different characteristic frequency bands: high frequency band, medium frequency band and low frequency band, based on the characteristic impedance, and to generate disturbance signals at three different frequency points belonging to the three characteristic frequency bands. The first detection module is used to perform passive islanding detection using the detection values ​​of the grid connection point. The detection values ​​include voltage harmonic fluctuations, voltage amplitude fluctuations, and frequency fluctuations. When any detection value is greater than the corresponding fluctuation warning threshold but less than or equal to the corresponding fluctuation alarm threshold, the second detection module is requested to work. When any detection value is greater than the corresponding fluctuation alarm threshold, the islanding state is determined and an alarm is triggered. Otherwise, continue with passive island detection; The second detection module is used to trigger active islanding detection on a short time scale according to a set first time interval; inject a broadband signal containing three different frequency disturbance signals into the distribution network, calculate the difference between the measured impedance value of the distribution network system under the three different frequency disturbance signals and the corresponding historical value, and request the third detection module to work when all three differences are greater than the corresponding first islanding warning threshold but at least one difference is less than the corresponding first islanding alarm threshold. When all three differences are not less than the corresponding first islanding alarm threshold, the islanding status is determined and an alarm is triggered. Otherwise, continue with active island detection on a short timescale; The third detection module is used to actively trigger long-term islanding detection according to the set second time interval or response request; it injects broadband signals of N disturbance signals at three different frequency points into the distribution network, and calculates the average value and coefficient of variation of the system impedance difference between the measured system impedance value and the historical system impedance value in each frequency band; when the average value of the N system impedance differences in the three frequency bands is greater than the corresponding second alarm threshold and the coefficient of variation is less than the coefficient of variation threshold, the islanding state is determined and an alarm is triggered; otherwise, it is impossible to determine whether the islanding state is in place, and the long-term active islanding detection is triggered again after the set time interval ends.

13. The multi-island detection system based on impedance increment of multi-frequency characteristic bands according to claim 12, characterized in that, Also includes: A multi-amplitude controller is used to inject disturbance signals at 3 or 3N frequency points into the power distribution network.

14. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-11.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-11.

Images