A continuously adjustable boost control method for detecting gapped surge arresters

Abstract

This invention relates to the field of surge arrester testing technology, and more particularly to a continuously adjustable boost control method for testing surge arresters with gaps. The method includes: S1, installation of a DC tester; S2, internal circuitry of the DC tester; S3, continuous boost mode; using the continuous boost mode, various boost slopes are set by adjusting the duty cycle of the PWM input; S4, fault determination. This invention sets various boost slopes for the DC tester by adjusting the PWM duty cycle, thereby enabling rapid voltage boosting and testing of surge arresters with gaps. In the initial stage of boosting, the slope is adjusted for rapid voltage increase. Near the critical detection point, the boost rate is slowed down to ensure the accuracy of the test results. Simultaneously, a DBSCAN clustering model is established using the boost slope as a feature to determine the possible fault type of the current surge arrester, thereby deciding whether to stop the boosting process.

Description

Technical Field

[0001] This invention relates to the field of surge arrester testing technology, and in particular to a continuously adjustable boost control method for testing surge arresters with gaps. Background Technology

[0002] A surge arrester DC tester is an instrument that detects internal defects in zinc oxide surge arresters used in high-voltage power grids by applying high-voltage DC current to the surge arrester. By measuring key parameters such as the DC reference voltage U1mA of the surge arrester (MOA) and the leakage current at 0.75 U1mA, it provides a scientific basis for assessing the internal defects of the surge arrester.

[0003] Existing surge arrester DC testers typically employ a staged or step-by-step voltage boosting strategy when generating high-voltage DC. This involves controlling the output high voltage to rise in fixed voltage steps, such as 0.5kV or 1kV per step, with a brief pause at each voltage step for sampling and measurement. The step changes in voltage can affect the accuracy of the measurement results. If the steps are set too densely, the test will be smooth but the total test time will be too long. If the steps are set too sparsely, key characteristic points, such as the precise inflection point of U1mA, will be missed. Testers will need to adjust the voltage back, causing the voltage to fluctuate repeatedly around that point, making accurate measurement impossible. Therefore, a control method that can quickly and continuously boost DC voltage is needed. Summary of the Invention

[0004] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description and other accompanying drawings.

[0005] The purpose of this invention is to overcome the above-mentioned shortcomings and provide a continuously adjustable boost control method for detecting gapped surge arresters. By adjusting the duty cycle of the PWM, various boost slopes are set for the DC detector, thereby enabling rapid boost detection of gapped surge arresters. In the initial stage of boost, the boost slope is adjusted for rapid boost. When approaching the critical detection point, the boost rate is slowed down to ensure the accuracy of the detection results. At the same time, a DBSCAN clustering model is established based on the boost slope to determine the possible fault type of the current surge arrester, thereby deciding whether to stop the boost process.

[0006] This invention provides a continuously adjustable boost control method for detecting gapped surge arresters, comprising:

[0007] S1. DC detector installation: For surge arresters with gaps, connect the current detection terminal of the DC detector to the gap terminal of the surge arrester. Ground the negative terminal of the DC detector and take readings from the DC detector.

[0008] S2. Internal circuit of DC detector: The DC detector is equipped with a battery and uses a DC-DC boost chip to convert the low voltage powered by the lithium battery into a high voltage corresponding to the high voltage grid, obtain the DC reference voltage U1mA and leakage current data under the condition of 0.75U1mA, and determine the fault status of the gap arrester.

[0009] S3, Continuous Boost Mode: During the DC-DC boost process, a continuous boost mode is adopted. By adjusting the duty cycle of the PWM input, multiple boost slopes are set. When the DC detector is turned on, the first slope is used for rapid continuous boost, and the leakage current of the surge arrester is monitored. When the leakage current exceeds the first set threshold, the second slope is switched to medium-speed continuous boost, until the leakage current exceeds the second set threshold, and then the third slope is switched to slow-speed continuous boost, thereby accurately boosting the voltage to the high voltage corresponding to the high voltage grid.

[0010] S4. Fault Judgment: For the gap arrester under test, the leakage current data under the DC reference voltage U1mA and 0.75U1mA are sent to the main control unit. The main control unit is equipped with a DBSCAN clustering model, which determines the fault type by judging the distance to each core point and outputs it to the display terminal of the DC detector.

[0011] In some embodiments, in step S2, the GND pin of the DC-DC boost chip is grounded, the VSYS pin is connected to a capacitor and then grounded, the LX pin is connected to an inductor and a capacitor in sequence and then grounded, the VIN pin is connected between the capacitor and the inductor of the LX pin, and a diode is connected between the VSYS pin and the LX pin.

[0012] In some embodiments, the specific steps in step S3 for forming various boost slopes by adjusting the duty cycle of the PWM input are as follows:

[0013] S31. Set the target slope and calculate the desired voltage, which is the voltage value that should be reached in one cycle. The target slope includes the first slope, the second slope and the third slope.

[0014] S32. Sample the high voltage output terminal of the DC-DC boost chip to obtain the current actual voltage value, and calculate the difference between the actual voltage and the expected voltage.

[0015] S33. The main control center adjusts the duty cycle based on the difference. The duty cycle adjustment includes... , and :

[0016] Proportion adjustment It is proportional to the current error. , for Adjust the weights; dif represents the difference.

[0017] Points adjustment It is directly proportional to the accumulated difference over a period of time. , for Adjusting the weights;

[0018] Differential adjustment It is proportional to the rate of change of the difference. , for Adjusting the weights;

[0019] The total output of the main control center is a proportional adjustment. +Points Adjustment +Differential Adjustment , is a numerical value, where , and The sum is 1;

[0020] S34. The total output of the main control center is input to the PWM waveform generator to change the duty cycle of the PWM.

[0021] S35. Repeat steps S32-S34 to ensure the accuracy of slope adjustment.

[0022] In some embodiments, in step S3, the first slope is the fast slope. The voltage boost rate is 3-5 kV / s; the second slope is the medium-speed slope. The voltage boost rate is 1-2 kV / s; the third slope is the low-speed slope. The boost rate is 0.1-0.3 kV / s; the first set threshold is 50-100 kV / s. The second threshold is set at 800-900. The target threshold for leakage current is 1mA.

[0023] In some embodiments, since poor contact may occur during the connection between the current sensing terminal of the DC detector and the gap terminal of the surge arrester, an early warning mechanism is set up: during the DC-DC boost process, the DC detector continuously samples the leakage current and calculates the short-time variance of the leakage current in real time. When the variance suddenly increases and exceeds a set threshold, an alarm signal for poor contact is sent to the display terminal of the DC detector. Simultaneously, in terms of data processing, the generated UI curve is filtered to remove jump points using the median filtering method, generating a smooth UI curve for calculating U1mA and... .

[0024] In some embodiments, the specific steps for establishing the DBSCAN clustering model in step S4 are as follows:

[0025] S41. Establish a dataset: Detect n surge arresters in a high-voltage power grid, collecting data including leakage current under a DC reference voltage U1mA and a voltage of 0.75U1mA. and the average slope of the pressure rise Create an n x 3 matrix;

[0026] S42. Data Preprocessing: Preprocessing DC reference voltage U1mA and leakage current... and the average slope of the pressure rise Z-score standardization effectively preserves the original shape of the data while making each row in the data matrix a data point with a three-dimensional spatial location.

[0027] S43. Run the DBSCAN algorithm: Input the standardized data and selected parameters into the DBSCAN algorithm. The selected parameters include the neighborhood radius. Given a minimum number of samples, all data points are set to an unvisited state. Each data point is then iterated over to... Clustering is performed based on the neighborhood radius. If the number of data points in a cluster reaches the minimum sample size, a cluster is generated and its core points are calculated. This process generates multiple clusters, while clusters not located within the neighborhood radius are not clustered. The data points within are set as noise points, and the selected parameters are iteratively debugged and updated.

[0028] S44. Fault Classification: Based on the actual situation of the surge arrester, labels are added to multiple clusters and noise points generated in the dataset to establish fault type classification.

[0029] In some embodiments, in step S41, the leakage current under the condition of DC reference voltage U1mA and 0.75U1mA is... The average pressure ramp rate was obtained through direct measurement. The first slope Second slope and the third slope The mean is obtained by adding the ingredients together and then dividing by 3.

[0030] In some embodiments, in step S43, for a data point in three-dimensional space, the neighborhood radius The initial setup uses a k-distance graph, which involves calculating the distance between each point and its k-th nearest neighbor, sorting all distances in ascending order, plotting the graph, and then finding the distance value corresponding to the inflection point of the curve as the neighborhood radius. The initial value, after the first clustering by the DBSCAN algorithm, corresponds to the neighborhood radius. This will cause a change, affecting the neighborhood radius. The data is then input into the model again for clustering, and after multiple iterations, the final model is generated.

[0031] In some embodiments, in step S4, a three-dimensional data point is established based on the current data currently acquired by the DC detector. The Euclidean distance between the current data point and the core points of each cluster in the DBSCAN model is calculated and a threshold is set. When the spatial distance between two points is within the threshold, it is determined that the fault type is the same as that of the cluster, and the fault type is output on the display end of the DC detector.

[0032] By adopting the above technical solution, the beneficial effects of the present invention are:

[0033] This invention sets various boost slopes for the DC detector by adjusting the duty cycle of the PWM, thereby enabling rapid voltage boosting and detection of surge arresters with gaps. In the initial stage of voltage boosting, the slope is adjusted for rapid voltage boosting. When approaching the critical detection point, the voltage boosting rate is slowed down to ensure the accuracy of the detection results. At the same time, a DBSCAN clustering model is established based on the voltage boosting slope to determine the possible fault type of the current surge arrester, thereby deciding whether to stop the voltage boosting process.

[0034] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure.

[0035] Undoubtedly, such and other objects of the present invention will become more apparent after the following detailed description of the preferred embodiments, which are illustrated in various accompanying drawings and figures.

[0036] To make the above and other objects, features and advantages of the present invention more apparent and understandable, one or more preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0037] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0038] In the accompanying drawings, the same parts use the same reference numerals, and the drawings are schematic and not necessarily drawn to actual scale.

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only one or more embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on such drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of a DC detector performing live testing on a gapped surge arrester in some embodiments of the present invention;

[0041] Figure 2 This is a schematic diagram of the overall flow of the control method in some embodiments of the present invention;

[0042] Figure 3 This is a schematic diagram of the external connection of the DC-DC boost chip in some embodiments of the present invention;

[0043] Figure 4 This is a schematic diagram of the internal circuit of the DC-DC boost chip in some embodiments of the present invention;

[0044] Figure 5 This is a schematic diagram of the process of adjusting the duty cycle to the target slope in some embodiments of the present invention;

[0045] Figure 6 This is a broken line diagram illustrating the effect of arrester failure on voltage ramp rate in some embodiments of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0047] Furthermore, in the description of this invention, it should be understood that the terms "center," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0048] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral unit; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. However, specifying a direct connection indicates that the two main bodies are not connected through a transitional structure, but rather formed as a whole through a connecting structure. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0049] In this invention, unless otherwise expressly specified and limited, the first feature "on" or "below" the second feature may be in direct contact with the first and second features, or indirect contact through an intermediate medium. In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0050] Reference Figures 1-2 , Figure 1 This is a schematic diagram of a DC detector performing live testing on a gapped surge arrester in some embodiments of the present invention; Figure 2 This is a schematic diagram of the overall flow of the control method in some embodiments of the present invention.

[0051] According to some embodiments of the present invention, the present invention provides a continuously adjustable boost control method relating to the detection of a gapped surge arrester, comprising:

[0052] S1. DC detector installation: For surge arresters with gaps, connect the current detection terminal of the DC detector to the gap terminal of the surge arrester. Ground the negative terminal of the DC detector and take readings from the DC detector.

[0053] The current detection terminal of the DC detector is cap-shaped and has an extended connecting wire. The inner side of the cap has a groove that can be energized and is engaged with the gapped end of the surge arrester. It is used to detect the real-time data of the gapped surge arrester and transmit it to the DC detector body. The outer side of the cap is made of insulating material to prevent the gapped end from forming a path with the top support through the current detection terminal, which would cause the surge arrester body to be energized and broken down. In actual use, the testing personnel climb the high-voltage tower and engage the current detection terminal with the gapped end of the surge arrester through the support. The negative terminal of the DC detector is suspended on the power pole to ground it. Another testing personnel on the ground remotely receive the data from the DC detector through a mobile terminal and check the reading of the surge arrester.

[0054] Because poor contact can cause disconnections during the connection between the current sensing terminal of the DC detector and the gap terminal of the surge arrester, an early warning mechanism is implemented: During the DC-DC boost process, the DC detector continuously samples the leakage current and calculates the short-time variance of the leakage current in real time. When the variance suddenly increases and exceeds a set threshold, an alarm signal for poor contact is sent to the display terminal of the DC detector. Simultaneously, in terms of data processing, the generated UI curve is filtered to remove jump points using median filtering, generating a smooth UI curve for calculating U1mA and... .

[0055] Reference Figures 3-4 , Figure 3 This is a schematic diagram of the external connection of the DC-DC boost chip in some embodiments of the present invention; Figure 4 This is a schematic diagram of the internal circuit of a DC-DC boost chip in some embodiments of the present invention.

[0056] S2. Internal circuit of DC detector: The DC detector is equipped with a battery and uses a DC-DC boost chip to convert the low voltage powered by the lithium battery into a high voltage corresponding to the high voltage grid, obtain the DC reference voltage U1mA and leakage current data under the condition of 0.75U1mA, and determine the fault status of the gap arrester.

[0057] Reference Figure 3 The GND pin of the DC-DC boost chip is grounded, the VSYS pin is connected to a capacitor and then grounded, the LX pin is connected to an inductor and a capacitor in sequence and then grounded, the VIN pin is connected between the capacitor and the inductor on the LX pin, and a diode is connected between the VSYS pin and the LX pin. The high voltage output terminal of the VSYS pin of the DC-DC boost chip is sampled to obtain the current actual voltage value for subsequent calculations. Figure 4 This is an example of the internal structure of a DC-DC boost chip. For specific details, please refer to existing technologies. No specific limitations are made to the DC-DC boost circuit here.

[0058] Reference Figure 5 , Figure 5 This is a schematic diagram illustrating the process of adjusting the duty cycle to the target slope in some embodiments of the present invention.

[0059] S3, Continuous Boost Mode: During the DC-DC boost process, a continuous boost mode is adopted. By adjusting the duty cycle of the PWM input, multiple boost slopes are set. When the DC detector is turned on, the first slope is used for rapid continuous boost, and the leakage current of the surge arrester is monitored. When the leakage current exceeds the first set threshold, the second slope is switched to medium-speed continuous boost, until the leakage current exceeds the second set threshold, and then the third slope is switched to slow-speed continuous boost, thereby accurately boosting the voltage to the high voltage corresponding to the high voltage grid.

[0060] In the low-current linear region, the current change is minimal. An excessively long voltage boosting process would result in long waiting times for testing personnel on the high-voltage tower. Therefore, a rapid voltage boosting with the first slope is used to shorten the time to a few seconds. The speed is reduced at the intermediate voltage. At the most critical inflection point, namely the 980-1020 microamp range, the voltage is boosted at an extremely slow speed to extend the sampling time and increase the number of samples, accurately locating the 1mA leakage current target threshold for monitoring.

[0061] The specific steps to generate various boost slopes by adjusting the duty cycle of the PWM input are as follows:

[0062] S31. Set the target slope and calculate the desired voltage, which is the voltage value that should be reached in one cycle. The target slope includes the first slope, the second slope and the third slope.

[0063] S32. Sample the high voltage output terminal of the DC-DC boost chip to obtain the current actual voltage value, and calculate the difference between the actual voltage and the expected voltage.

[0064] S33. The main control center adjusts the duty cycle based on the difference. The duty cycle adjustment includes... , and :

[0065] Proportion adjustment It is proportional to the current error. , for Adjust the weights; dif represents the difference. As the main adjustment item for duty cycle, the larger the difference, the greater the correction.

[0066] Points adjustment It is directly proportional to the accumulated difference over a period of time. , for Adjusting the weights; To eliminate static error, when the voltage is always slightly lower than the target voltage value, the integral term will gradually increase to make up for the entire gap.

[0067] Differential adjustment It is proportional to the rate of change of the difference. , for The adjustment weights are adjusted; if the voltage is rapidly approaching the target value, the differential term will reduce the correction magnitude in advance to prevent overshoot;

[0068] The total output of the main control center is a proportional adjustment. +Points Adjustment +Differential Adjustment , is a numerical value, where , and The sum is 1; preferably, , and The initial settings were 0.7, 0.2, and 0.1, respectively.

[0069] S34. The main output of the control center is input to the PWM waveform generator to change the duty cycle of the PWM. A high duty cycle will cause the voltage to rise faster and the slope to increase, while a low duty cycle will cause the voltage to rise slower and the slope to decrease.

[0070] S35. Repeat steps S32-S34 to ensure the accuracy of slope adjustment. When the actual voltage lags behind the target voltage, the main control center will increase the duty cycle to accelerate the voltage rise. When the actual voltage is about to exceed the target voltage, the main control center will decrease the duty cycle to reduce the voltage rise. Through this repeated adjustment process, the actual voltage of the surge arrester under normal conditions will conform to the preset voltage rise curve, thereby achieving an adjustable and precise voltage rise curve.

[0071] In actual boosting processes, the first slope is the rapid slope. The voltage boost rate is 3-5 kV / s; the second slope is the medium-speed slope. The voltage boost rate is 1-2 kV / s; the third slope is the low-speed slope. The boost rate is 0.1-0.3 kV / s; the first set threshold is 50-100 kV / s. The second threshold is set at 800-900. The target threshold for leakage current is 1mA.

[0072] S4. Fault Judgment: For the gap arrester under test, the leakage current data under the DC reference voltage U1mA and 0.75U1mA are sent to the main control unit. The main control unit is equipped with a DBSCAN clustering model, which determines the fault type by judging the distance to each core point and outputs it to the display terminal of the DC detector.

[0073] The specific steps for establishing a DBSCAN clustering model are as follows:

[0074] S41. Establish a dataset: Detect n surge arresters in a high-voltage power grid, collecting data including leakage current under a DC reference voltage U1mA and a voltage of 0.75U1mA. and the average slope of the pressure rise Establish an n x 3 matrix; leakage current under DC reference voltage U1mA and 0.75U1mA conditions. The average slope of the pressure rise can be obtained through direct measurement. The first slope Second slope and the third slope The mean is obtained by adding the ingredients together and then dividing by 3.

[0075] S42. Data Preprocessing: Preprocessing DC reference voltage U1mA and leakage current... and the average slope of the pressure rise Z-score standardization effectively preserves the original shape of the data while each row in the data matrix represents a data point with a three-dimensional spatial location. The standardized data have a mean of 0 and a standard deviation of 1, with all features on the same order of magnitude, making distance calculation more equitable.

[0076] S43. Run the DBSCAN algorithm: Input the standardized data and selected parameters into the DBSCAN algorithm. The selected parameters include the neighborhood radius. Given a minimum number of samples, all data points are set to an unvisited state. Each data point is then iterated over to... Clustering is performed based on the neighborhood radius. If the number of data points in a cluster reaches the minimum sample size, a cluster is generated and its core points are calculated. This process generates multiple clusters, while clusters not located within the neighborhood radius are not clustered. The data points within are set as noise points, and the selected parameters are iteratively debugged and updated. Unlike algorithms such as K-MEANS that require the number of clusters to be preset, DBSCAN does not require the user to input the number of clusters to be divided. It can effectively handle noise points and connect adjacent areas with sufficient density, which is consistent with the situation in the actual detection process of surge arresters where the number of specific fault types is uncertain.

[0077] For data points in three-dimensional space, the neighborhood radius The initial setup uses a k-distance graph, which involves calculating the distance between each point and its k-th nearest neighbor, sorting all distances in ascending order, plotting the graph, and then finding the distance value corresponding to the inflection point of the curve as the neighborhood radius. The initial value, after the first clustering by the DBSCAN algorithm, corresponds to the neighborhood radius. This will cause a change, affecting the neighborhood radius. The data is input into the model again for clustering, and the final model is generated after multiple iterations.

[0078] Neighborhood radius Draw a circle with radius centered at the current data point. The clustering method, which uses a sphere to represent all data points within it, effectively reduces the computational load of the model. Compared to the conventional method of calculating neighboring points using Euclidean distance, this improves the model building speed and reduces the impact of high dimensionality on the DBSCAN algorithm. Furthermore, since practical applications only require determining the approximate fault type for surge arresters, minor deviations in the data will not cause abrupt changes in the fault type. Therefore, during the model building stage, the neighborhood radius can be used. Perform rapid cluster partitioning.

[0079] Reference Figure 6 , Figure 6 This is a broken line diagram illustrating the effect of arrester failure on voltage ramp rate in some embodiments of the present invention.

[0080] S44. Fault Classification: Based on the actual situation of the surge arrester, labels are added to multiple clusters and noise points generated in the dataset to establish fault type classification.

[0081] Based on the actual situation of surge arresters, add corresponding fault labels to the clusters in the model:

[0082] For the largest cluster in the model, which is generally the surge arrester under normal conditions, the core points of the cluster can be labeled based on the judgment of the actual staff to mark the surge arrester data under normal conditions. The current-voltage characteristic curve of the surge arrester under normal conditions rises sharply when the current is large, and the nonlinear characteristics are obvious.

[0083] Noise Point 1 / Cluster 1: Aging and deteriorating cluster, DC reference voltage U1mA decreases; when the leakage current reaches 1mA, the obtained voltage value will be significantly lower than the factory setting value, and due to nonlinear variation, the leakage current at 0.75U1mA increases significantly. More current will flow at lower voltages, causing the overall volt-ampere characteristic curve to shift to the left. However, the overall measurement process can still be completed. Add a label to the core point of this point / cluster to mark the aging and deterioration data of the surge arrester.

[0084] Noise Point 2 / Cluster 2: Severely Moisturized Cluster. In the initial stage of voltage rise, the current increases sharply with a slight increase in voltage. When the voltage reaches just a few kV, the current exceeds 1mA. Moreover, the volt-ampere characteristic curve becomes very flat, the linearity increases, the voltage is difficult to rise, and it loses its nonlinear characteristics. Add a label to the core point of this point / cluster to mark the severe moisture data of the surge arrester.

[0085] Noise Point 3 / Cluster 3: Internal fracture cluster. When there is an internal fracture, the U1mA condition is similar to that of aging and deterioration, but the overall threshold voltage is reduced, and jump steps are likely to appear in the boost curve. The jump steps are due to the sparks generated by internal fragments under high voltage. Add a label to the core point of this point / cluster to mark the internal fracture data of the surge arrester.

[0086] Based on the current data acquired by the DC detector, a three-dimensional data point is established. The Euclidean distance between the current data point and the core points of each cluster in the DBSCAN model is calculated and a threshold is set. When the spatial distance between two points is within the threshold, it is determined that the fault type is the same as that of the cluster, and the fault type is output on the display end of the DC detector.

[0087] To ensure the accuracy of fault type determination, Euclidean distance calculation is used for the distance calculation between the current data point and the core point. The process of comparing a single data point with the core points in multiple clusters is relatively simple and fast, which can quickly determine the fault type and avoid the inspection personnel staying on the high-voltage tower for too long.

[0088] During the testing process, if the DC detector has determined the current fault type of the surge arrester, the voltage boosting process will be stopped accordingly, and the testing personnel will be notified to replace the surge arrester. If the surge arrester is found to be severely damp and the current is too high in the initial stage of voltage boosting, the DC detector will stop the voltage boosting process and activate the overcurrent protection, displaying the corresponding fault type on the display terminal and reminding the user to replace the surge arrester. If an internal break is detected in the surge arrester, it is very dangerous for the entire high-voltage tower. When a step change occurs, the DC detection voltage boosting process will be quickly cut off, and the internal break will be displayed on the display terminal. At the same time, it is necessary to consider whether to activate the drop-out fuse.

[0089] It should be understood that the embodiments disclosed herein are not limited to the specific processing steps or materials disclosed herein, but should be extended to equivalent substitutions of such features as understood by those skilled in the art. It should also be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting.

[0090] The term "embodiment" in this specification refers to a specific feature or characteristic described in connection with an embodiment that is included in at least one embodiment of the invention. Therefore, phrases or "embodiments" appearing in various places throughout the specification do not necessarily refer to the same embodiment.

[0091] Furthermore, the described features or characteristics can be incorporated into one or more embodiments in any other suitable manner. In the above description, specific details, such as thickness, quantity, etc., are provided to provide a comprehensive understanding of embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented without the aforementioned specific details or may be implemented using other methods, components, materials, etc.

Claims

1. A continuously adjustable boost control method involving gap arrester detection, characterized by, Comprising S1, DC detector installation: for the gap arrester, the current detection end of the DC detector is buckled with the gap end of the arrester, the negative end of the DC detector is grounded, and the DC detector is read; S2, DC detector internal circuit: the DC detector is provided with a battery, a DC-DC boost chip is used to convert the low voltage of the lithium battery power supply into the high voltage corresponding to the high voltage power grid, the DC reference voltage U1mA and the leakage current data under the condition of 0.75U1mA are obtained, and the fault condition of the gap arrester is determined; S3, continuous boost mode: in the DC-DC boost process, a continuous boost mode is used, the duty cycle of PWM input is adjusted, a plurality of boost slopes are set, when the DC detector is turned on, first, a first slope is used for fast continuous boost, and the arrester leakage current is monitored, when the leakage current exceeds the first set threshold, a second slope is switched to medium-speed continuous boost, and when the leakage current exceeds the second set threshold, a third slope is switched to slow-speed continuous boost, so as to accurately boost to the high voltage corresponding to the high voltage power grid; By adjusting the duty cycle of PWM input, the specific steps of forming a plurality of boost slopes are: S31, set the target slope and calculate the expected voltage, that is, the voltage value that should be reached in a period, the target slope includes the first slope, the second slope and the third slope; S32, sample the high voltage output end of the DC-DC boost chip to obtain the current actual voltage value, and calculate the difference between the actual voltage and the expected voltage; S33, the master control center adjusts the duty cycle according to the difference, the duty cycle adjustment part includes 、 and : Proportion adjustment in proportion to the current error, , is adjusting the weight, dif is the difference value; Integral adjustment proportional to the difference over time, an adjustment weight of​​ Differential adjustment It is proportional to the rate of change of the difference. , for Adjusting the weights; the total output of the master center is proportionally adjusted + integral adjustment + derivative adjustment is a value, wherein , and the sum of which is 1; S34, the total output of the main control center is input into the PWM waveform generator to change the duty cycle of PWM; S35, repeat the steps of S32-S34 to ensure the accuracy of the slope adjustment; S4, fault determination: for the gap arrester under test, the leakage current data under the conditions of DC reference voltage U1mA and 0.75U1mA are sent to the main control unit, the DBSCAN clustering model is set in the main control unit, the distance from each core point is determined to determine the fault type, and the output is output to the DC detector display end.

2. The continuously adjustable boost control method involving gap arrester detection of claim 1, wherein, In S2, the GND pin of the DC-DC boost chip is grounded, the VSYS pin is connected to a capacitor and then grounded, the LX pin is sequentially connected to an inductor and a capacitor and then grounded, the VIN pin is connected between the capacitor and the inductor of the LX pin, and a diode is connected between the VSYS pin and the LX pin.

3. The continuously adjustable boost control method involving gap arrester detection of claim 1, wherein, In the S3 step, the first slope is a fast slope with a ramp-up rate of 3-5 kV / s; the second slope is a medium slope with a ramp-up rate of 1-2 kV / s; the third slope is a low slope with a ramp-up rate of 0.1-0.3 kV / s; the first set threshold is 50-100 ; the second set threshold is 800-900 ; and the leakage current target threshold is 1 mA.

4. The continuously adjustable boost control method involving gap arrester detection of claim 1, wherein, Because the process of the current detection end of the DC detector and the gap end of the lightning arrester is connected, the disconnection will occur due to poor contact, so the early warning mechanism is set: during the DC-DC voltage boosting process, the DC detector continuously samples the leakage current, and calculates the short-time variance of the leakage current in real time, when the variance suddenly increases and exceeds the set threshold, an alarm signal of poor contact is sent to the display end of the DC detector; at the same time, in the aspect of data processing, the generated U-I curve is filtered by the median filtering method to remove the jumping points, and a smooth U-I curve is generated to calculate U1mA and .

5. The method for continuously adjustable boost control involving gap arrester detection according to claim 1, wherein, In S4, the specific steps of establishing the DBSCAN clustering model are: S41, establish a data set: detecting n arrester in high-voltage power grid, collecting data including DC reference voltage U1mA, leakage current under 0.75U1mA And boost slope mean , establishing a data matrix of n rows and 3 columns; S42, data preprocessing: on the direct current reference voltage U1mA, leakage current And the boost slope mean Z-score standardization is performed, each row in the data matrix is a data point with three-dimensional spatial position while effectively preserving the original shape of the data. S43, running DBSCAN algorithm: inputting the standardized data and selected parameters, including neighborhood radius and minimum sample number, into the DBSCAN algorithm, setting all data points as unvisited, traversing each data point to cluster in the neighborhood radius, generating a cluster and calculating the core point if the number of clustered data points reaches the minimum sample number, thereby generating multiple clusters, setting data points not within the neighborhood radius as noise points, and iteratively debugging and updating the selected parameters; ;​​ S44, fault classification: add labels to a plurality of clusters and noise points generated in the data set according to the actual situation of the arrester, and establish a fault type classification.

6. The continuously adjustable boost control method involving gap arrester detection of claim 5, wherein, In step S41, the direct current reference voltage Ui mA, the leakage current in the case of 0.75 Ui mA obtained by direct measurement, average of the boost slope is the first slope , the second slope and the third slope added and divided by 3 to obtain the average.

7. The method for continuously adjustable boost control involving gap arrester detection of claim 5, wherein, In step S43, for the three-dimensional spatial data points, the neighborhood radius The initial setting is made by k-distance graph, that is, the distance of each point to its kth nearest neighbor is calculated, and all distances are sorted from small to large and plotted to find the distance value corresponding to the inflection point in the curve as the initial value of the neighborhood radius After the initial clustering of the DBSCAN algorithm, the corresponding neighborhood radius will change, and the neighborhood radius will be input into the model again for clustering, and the final model is generated after multiple iterations.

8. The method for continuously adjustable boost control involving gap arrester detection of claim 1, wherein, In S4, according to the current current data obtained by the DC detector, a three-dimensional data point is established, the Euclidean distance between the current data point and each cluster core point in the DBSCAN model is calculated and a threshold is set, when the spatial distance between two points is within the threshold, it is determined that the fault type of the cluster is the same, and the determined fault type is output to the DC detector display end.

Images