Short-circuit current protection method for direct-current port of power electronic transformer

Abstract

The invention discloses a short-circuit current protection method for a direct-current port of a power electronic transformer, which relates to the technical field of power control, and comprises the following steps: when detecting that the short-circuit current of the direct-current port is abnormal, generating a control instruction, driving an ultrafast electronic switch to execute a cut-off action, and recording and uploading the fault occurrence time, position and abnormal parameters; generating a fault diagnosis result and a maintenance suggestion; when the fault diagnosis result and the maintenance suggestion confirm that recovery conditions exist, a recovery process is started, and the power electronic transformer is connected to the direct current power supply loop again; continuously collecting operation state data in real time, and dynamically correcting a short-circuit current response mechanism by adopting a self-adaptive updating algorithm in combination with an operation state data set and abnormal parameters recorded in a fault diagnosis result; by optimizing a protection mechanism and monitoring and analyzing faults in real time, risks and shutdown loss caused by equipment failure or external environment change are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of power control, and particularly to a method for protecting the short-circuit current of the DC port of a power electronic transformer. Background Art

[0002] As a new generation of distribution equipment, the power electronic transformer integrates various technologies such as power electronic conversion, communication, and intelligent control, and has wide applications in scenarios such as new energy access, flexible power distribution, rail transit, and data centers. Compared with traditional power frequency transformers, power electronic transformers have advantages such as small volume, light weight, diverse functions, and flexible regulation, and are gradually becoming key core equipment in the smart grid. Especially in high-voltage DC power distribution, the DC port of the power electronic transformer directly undertakes the power supply demand of the load side, and its operating state has an important impact on the overall stability and safety. Therefore, in order to ensure good anti-interference and fast protection capabilities in the event of a sudden short circuit, it is urgent to effectively monitor and handle the short-circuit fault of the DC port of the power electronic transformer.

[0003] Most of the existing short-circuit protection methods for power electronic transformers rely on traditional threshold-triggering strategies to achieve fault identification and protection actions through fixed current or voltage thresholds. This method has the risk of identification delay or misjudgment when facing situations such as changes in operating conditions and complex dynamic responses, and cannot meet the actual needs of rapid, accurate, and adaptive identification of DC short-circuit current. In addition, traditional protection schemes often lack an effective parameter evaluation and adaptive correction mechanism during the fault recovery stage, resulting in potential safety hazards during the recovery process, and even possibly triggering secondary faults in severe cases. The existing technologies still have obvious deficiencies in the real-time and accuracy of abnormal short-circuit current identification, which has become an important bottleneck restricting the safe operation of power electronic transformers in complex DC systems. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for protecting the short-circuit current of the DC port of a power electronic transformer to solve the problem that the short-circuit fault response mechanism in the prior art lacks real-time and self-adaptability.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for protecting the short-circuit current of the DC port of a power electronic transformer, which includes collecting current data, voltage data, and temperature data of the DC port to form an operating state data set; performing real-time analysis on the operating state data set using a density clustering algorithm to identify abnormal features related to the short-circuit current of the DC port and generating an abnormal intensity index; when an abnormal short-circuit current of the DC port is detected, generating a control instruction to drive an ultrafast electronic switch to perform a cut-off action, and recording and uploading the time, location, and abnormal parameters of the fault occurrence to generate a fault diagnosis result and maintenance suggestion; when the fault diagnosis result and maintenance suggestion confirm that the recovery condition is met, starting a recovery process to reconnect the power electronic transformer to the DC power supply circuit; continuously and real-time collecting operating state data, combining the operating state data set with the abnormal parameters recorded in the fault diagnosis result, and dynamically correcting the short-circuit current response mechanism using an adaptive update algorithm.

[0007] As a preferred solution of the method for protecting the short-circuit current of the DC port of the power electronic transformer according to the present invention, wherein: the current, voltage, and temperature data of the DC port are aligned by time stamp to generate a synchronous original data set, which is normalized to obtain a structured data table, and through periodic caching, an operating state data set is formed.

[0008] As a preferred solution of the method for protecting the short-circuit current of the DC port of the power electronic transformer according to the present invention, wherein: the specific steps for generating the abnormal intensity index are as follows. The operating state data set is processed using a sliding window, continuously intercepting data segments of a specific time length to form short-period data sequences, and extracting multi-dimensional features to obtain a set of short-period operating state feature vectors. The density clustering method is used to perform clustering analysis on the set of short-period operating state feature vectors to mark abnormal features. The confidence evaluation mechanism is used to calculate the credibility of the abnormal features to generate an abnormal intensity index.

[0009] As a preferred solution of the method for protecting the short-circuit current of the DC port of the power electronic transformer according to the present invention, wherein: the specific steps for generating the control instruction are as follows. Extract the current intensity index from the abnormal intensity index to generate a preliminary alarm signal, and combine it with the real-time current change rate to judge the fault level and generate a level identifier. Match the preset action strategy according to the level identifier, and determine the corresponding switch trigger method and action parameters to generate a control instruction.

[0010] As a preferred solution of the method for protecting the short-circuit current of the DC port of the power electronic transformer according to the present invention, wherein: the specific steps for generating the fault diagnosis result and maintenance suggestion are as follows. After receiving the control instruction, control the ultra-fast electronic switch to disconnect the circuit to complete the cut-off operation, record the fault occurrence time and location, and extract abnormal parameters at the same time; Summarize the fault time, location and abnormal parameters to form fault information, compare them using preset rules, generate a fault diagnosis result, and search for a maintenance process to generate maintenance suggestions.

[0011] As a preferred solution of the short-circuit current protection method for the DC port of the power electronic transformer described in the present invention, wherein: the recovery process has the following specific steps Define the recovery conditions according to the fault diagnosis result and maintenance suggestions; Check and verify the recovery determination parameters of the power electronic transformer to determine whether they meet the recovery conditions; When the recovery conditions are met, perform the power-off and reconnection operations of the power electronic transformer at the DC port.

[0012] As a preferred solution of the short-circuit current protection method for the DC port of the power electronic transformer described in the present invention, wherein: continuously and real-time collect the operation status data, combine the operation status data set with the abnormal parameters recorded in the fault diagnosis result, and use an adaptive update algorithm to dynamically correct the short-circuit current response mechanism. The specific steps are as follows Continuously and real-time collect the updated operation status data set, and maintain the timeliness of the data through a data synchronization mechanism; Match the updated operation status data set with the abnormal parameters in the fault diagnosis result, and extract the current short-circuit current response mechanism parameters; Use an adaptive update algorithm to dynamically adjust the current short-circuit current response mechanism parameters.

[0013] As a preferred solution of the short-circuit current protection method for the DC port of the power electronic transformer described in the present invention, wherein: use an adaptive update algorithm to dynamically adjust the current short-circuit current response mechanism parameters. The specific steps are as follows Obtain the current short-circuit current response mechanism parameters, compare them with the updated operation status data set, and obtain the parameter offset information; Identify the current short-circuit current response mechanism parameter items to be adjusted according to the parameter offset information; Based on the current operation status data, select the corresponding adaptive update algorithm and calculate the updated values of each parameter to be adjusted; Apply the updated values to the original response mechanism parameters to generate a new response mechanism parameter set.

[0014] Second aspect, the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and: when the computer program is executed by the processor, any step of the short-circuit current protection method for the DC port of the power electronic transformer as described in the first aspect of the present invention is implemented.

[0015] Third aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and: when the computer program is executed by the processor, any step of the short-circuit current protection method for the DC port of the power electronic transformer as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: By collecting the updated operation state data in real time and combining the abnormal parameters in the fault diagnosis results, an adaptive update algorithm is used to dynamically adjust the parameters of the short-circuit current response mechanism. This step realizes the automatic adjustment of the protection mechanism under different working conditions to ensure that the protection strategy can always adapt to the current operation state. The function is to optimize the protection mechanism in real time, with strong adaptability and response ability. It can not only reduce the misoperation and missed operation caused by parameter changes, but also continuously and effectively protect the power electronic transformer under different changes such as load, ambient temperature, and voltage fluctuation. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] Figure 1 It is a flowchart of the short-circuit current protection method for the DC port of the power electronic transformer in the present invention.

[0019] Figure 2 It is a flowchart of the generation of the abnormal intensity index in the present invention.

[0020] Figure 3 It is a flowchart of the fault diagnosis and recovery in the present invention.

[0021] Figure 4 It is a flowchart of the dynamic correction of the response mechanism in the present invention. Detailed Embodiments

[0022] To make the above objects, features, and advantages of the present invention more obvious and understandable, the detailed embodiments of the present invention will be described in detail below with reference to the drawings in the specification.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1 to 4 This embodiment provides a short-circuit current protection method for a DC port of a power electronic transformer, comprising the following steps: S1. Align the current, voltage and temperature data of the DC port by timestamp to generate a synchronized original data set, obtain a structured data table through normalization, and form an operating status data set through periodic caching.

[0026] It should be noted that the current data, voltage data and temperature data of the DC port are collected separately through a high-frequency sampling device, and each type of data is accompanied by a corresponding sampling timestamp when collected. After the collection is completed, the three types of data are sorted according to the timestamp field, and a unified reference time granularity is set, such as 100 milliseconds or 1 second as the alignment period, and extra samples are interpolated or discarded within each time granularity, so that different physical quantities form corresponding data values ​​at the same time node, and the synchronized original data sets of current data, voltage data and temperature data are obtained; Normalize each field in the synchronized original data set. The processing methods include but are not limited to maximum and minimum value normalization, mean variance normalization or Z-score normalization. The specific method can be determined according to the data fluctuation characteristics. For example, the value range of the current data is scaled to the [0, 1] interval through linear transformation, and the temperature data is standardized with zero mean and unit variance to ensure that the three types of data are comparable under the same scale. All normalization processes are performed synchronously on the basis of retaining the original timestamp field; After processing, the data is written into a structured data table in chronological order. The structured data table uses the timestamp as the primary key and has current, voltage, and temperature fields to form a standardized data structure. To support subsequent real-time analysis tasks, a fixed cache cycle is set, such as triggering a data refresh every 5 or 10 seconds, and periodically writing the data in the latest time period into the operating status data set, and removing historical data outside the cache window to ensure that the operating status data set is always up to date and the data volume is stable.

[0027] S2. Perform real-time analysis on the operating state dataset using an anomaly recognition algorithm to identify anomaly features related to the short-circuit current of the DC port and generate an anomaly intensity index.

[0028] Perform sliding window processing on the operating state dataset, continuously intercept data segments of a specific time length to form short-period data sequences, and extract multi-dimensional features to obtain a set of short-period operating state feature vectors.

[0029] Specifically, set the time length of the sliding window in the operating state dataset. For example, set the length of each window to 10 seconds and the step size to 1 second. Perform time series partitioning on the operating state dataset, and sequentially intercept data segments of each 10-second length from the beginning in chronological order. Each data segment contains all current data, voltage data, and temperature data within that time range. Each time the sliding window slides forward by 1 second, repeat the interception operation to form multiple consecutive short-period data sequences; Extract multi-dimensional features for each short-period data sequence. The extracted multi-dimensional features include statistical features, change features, and frequency domain features. Statistical features include the mean, standard deviation, range, maximum value, and minimum value of current data, voltage data, and temperature data, etc. For example, calculate the average value and standard deviation of the current values within each data segment. Change features include the mean and standard deviation of the first-order difference sequences of current data, voltage data, and temperature data, the number of maximum / minimum points, the change slope, etc.; Frequency domain features use the fast Fourier transform (FFT) to transform current data, voltage data, and temperature data respectively, and extract the frequency component with the largest amplitude as the frequency domain feature. For example, take the amplitudes of the first 5 frequencies with the largest energy distribution to form a spectrum feature vector; All the extracted features are combined according to current features, voltage features, and temperature features, spliced into a DC port operating state joint feature vector in a unified format, and associated with the start timestamp of the short-period data sequence. Repeat the above operations to finally obtain the corresponding set of short-period operating state feature vectors.

[0030] Use the density clustering method to perform clustering analysis on the set of short-period operating state feature vectors and mark the anomaly features.

[0031] Specifically, set the key parameters of the density clustering method, including the neighborhood radius threshold and the minimum number of neighborhood samples MinPts; then for each feature vector in the set of short-period operating state feature vectors , calculate the Euclidean distance from all other feature vectors , and the expression is: ; where, represents the short-period operating state feature vector and the short-period operating state feature vector The Euclidean distance between represents the dimensionality of the short-term operating state feature vector, indicating that distance component calculation is performed on the th dimension, representing the th short-term operating state feature vector in the set of short-term operating state feature vectors, representing the th short-term operating state feature vector in the set of short-term operating state feature vectors; Count the number of short-term operating state feature vectors that satisfy , denoted as representing the number of short-term operating state feature vectors contained in the neighborhood centered on the short-term operating state feature vector with a radius of ; when is satisfied, then the short-term operating state feature vector is used as a core point, and all short-term operating state feature vectors in the neighborhood are merged into the same cluster, and density-reachable feature vectors are continuously expanded outward until the cluster no longer expands; when a short-term operating state feature vector fails to be assigned to any cluster and is satisfied, then the short-term operating state feature vector is marked as an abnormal feature point. After completing the clustering assignment of all short-term operating state feature vectors, output all cluster subsets and the corresponding short-term operating state feature vectors, and at the same time mark all short-term operating state feature vectors that are not assigned to valid clusters as abnormal features.

[0032] Calculate the credibility of abnormal features through a confidence evaluation mechanism to generate an abnormal intensity index.

[0033] Specifically, calculate the average Euclidean distance, and the expression is: ; where represents the total number of short-term operating state feature vectors; represents the short-term operating state feature vector participating in the calculation and belonging to the same cluster subset as ; ; Construct a confidence function for quantifying the degree of abnormality, and the expression is: ; where represents the final abnormal intensity index of the short-term operating state feature vector , which is used to comprehensively evaluate the credibility of its being an abnormal feature; The weight coefficient representing the neighborhood density factor reflects the influence degree of neighborhood sparsity on the anomaly intensity, and its value range is usually ; The weight coefficient representing the Euclidean distance factor reflects the influence degree of the average distance from other feature vectors within the cluster on the anomaly intensity, and its value range is usually , and there is ; For all values, perform maximum-minimum normalization processing to generate the corresponding anomaly intensity index, and the expression is: ; Among them, represents the normalized anomaly intensity index of the short-cycle operation state feature vector , and its value range is , which is used to uniformly measure the anomaly degree between different short-cycle operation state feature vectors. The larger the value, the more abnormal it is; represents the minimum value of the anomaly intensity indexes of all short-cycle operation state feature vectors ; represents the maximum value of the anomaly intensity indexes of all short-cycle operation state feature vectors , represents the anomaly intensity index of the short-cycle operation state feature vector.

[0034] S3. When detecting an abnormal short-circuit current at the DC port, generate a control instruction.

[0035] Extract the current intensity index from the anomaly intensity index, generate a preliminary alarm signal, and combine it with the real-time current change rate to judge the fault level and generate a level identifier.

[0036] Specifically, extract the normalized anomaly intensity index corresponding to the current moment from the anomaly intensity index sequence generated in the previous stage ; then, compare this normalized anomaly intensity index with the preset alarm threshold. If , then generate a preliminary alarm signal, where is an example preset threshold, such as = 0.75; then, obtain the current value of the DC port and the current value at the previous moment from the structured data table, and calculate the current current change rate , and the expression is: ; Compare the normalized anomaly intensity index corresponding to the preliminary alarm signal with the current current change rate As a combined input, a multi - condition rule is used to judge the fault level. For example: when and it is marked as "Level - 1 fault"; when and it is marked as "Level - 2 fault"; when and it is marked as "Level - 3 fault"; the judgment result is output as the level identifier.

[0037] Match the preset action strategy according to the level identifier, and determine the corresponding switch trigger method and action parameters to generate a control instruction.

[0038] It should be noted that the fault level is read from the current fault level identifier, such as "Level - 1 fault", "Level - 2 fault" or "Level - 3 fault"; then the preset action strategy item corresponding to the fault level identifier is found in the action strategy. The action strategy item contains information such as switch type, trigger method, trigger delay parameter and action duration parameter. For example, the action strategy corresponding to "Level - 1 fault" is "trigger the circuit breaker to trip", "the trigger method is voltage drop", "the trigger delay is 100 ms", "the action duration is 500 ms"; then according to the switch type and trigger method in the action strategy, the control instruction fields are assembled, including the target switch address, action type field, trigger condition field and action parameter field; finally, the generated control instruction is written into the output instruction buffer queue and ready to enter the control execution process.

[0039] S4. Drive the ultrafast electronic switch to perform the cut - off action, record and upload the time, location and abnormal parameters of the fault occurrence, and generate the fault diagnosis result and maintenance suggestion.

[0040] After receiving the control instruction, control the ultrafast electronic switch to disconnect the circuit to complete the cut - off action, record the fault occurrence time and location, and extract the abnormal parameters at the same time.

[0041] It should be noted that the target switch address and action type field in the control instruction are read, and the action type is parsed as "disconnect" and the target device is the ultrafast electronic switch; then a disconnection trigger signal is sent to the target ultrafast electronic switch to start its internal opening logic and complete the rapid cut - off of the circuit; at the same time when the action is completed, the current timestamp is recorded by the clock module as the fault occurrence time, and the fault occurrence location is located according to the line number or sensor identifier included in the control instruction; finally, multi - dimensional abnormal parameters within a set time window before and after the fault occurrence moment are extracted from the currently collected original waveform data such as current and voltage, such as current peak value, voltage drop amplitude, current change rate, etc., and summarized to form abnormal parameters.

[0042] Summarize the fault time, location, and abnormal parameters to form fault information, compare it using preset rules, generate a fault diagnosis result, and search for the maintenance process to generate maintenance suggestions.

[0043] Specifically, the fault occurrence time, fault location identifier, and currently extracted abnormal parameters (such as current peak value, voltage drop amplitude, current change rate) are formed into structured fault information in a unified format; subsequently, the preset rules are called to compare the structured fault information. The preset rules are constructed through statistical analysis of historical fault cases, and the rule content includes mapping relationships such as abnormal parameter threshold ranges, fault time period characteristics, and typical fault types corresponding to locations. For example, it is set that when the current peak value is greater than 300A and the current change rate exceeds 500A / ms, it is judged as a short - circuit fault; during the comparison process, each rule is condition - matched, and the rule item with the highest matching degree is selected to generate the corresponding fault diagnosis result; after the diagnosis result is generated, based on the current diagnosis result, the matching maintenance suggestions are retrieved from the preset maintenance process database. The maintenance process database establishes maintenance step entries indexed by fault types. For example, the maintenance suggestions for a short - circuit fault include: power - off inspection, insulation test, fuse replacement, and line re - inspection; finally, the fault diagnosis result and the corresponding maintenance suggestions are output.

[0044] S5. When it is confirmed that the fault diagnosis result and the maintenance suggestions meet the recovery conditions, start the recovery process and reconnect the power electronic transformer to the DC power supply circuit.

[0045] Define the recovery conditions according to the fault diagnosis result and the maintenance suggestions.

[0046] Specifically, collect common fault types, such as "bus grounding", "feeder short circuit", "transformer overload", etc., and sort out the typical recovery operation processes corresponding to each fault type. Extract the maintenance steps and judgment conditions from them, such as the completion status of operations like "replace faulty components", "detect insulation value", "confirm load drop", etc. as binary flag bits; then, using the fault type as the primary key, construct the corresponding recovery condition expression. For example, for the "feeder short circuit" fault, the recovery condition expression is "fuse replacement completion flag = 1 and cable insulation detection pass flag = 1 and current recovery to rated value flag = 1"; then, organize all fault types and their corresponding recovery conditions to form a fault recovery logic rule table, and continuously supplement and optimize it during subsequent operation and maintenance processes; during application, extract the fault type label from the fault diagnosis result, such as "short circuit fault", "grounding anomaly", or "overload operation", and match the maintenance operation items listed in the maintenance suggestions. Extract the operation completion status flag bits for each item, such as "fuse replacement completion flag = 1", "insulation resistance detection pass flag = 1", "line voltage recovery to normal range flag = 1"; finally, based on the recovery condition expression corresponding to the current fault type in the fault recovery logic rule table, perform a logical combination judgment on each flag bit. If the logical judgment result is true, mark it as "meeting the recovery conditions", otherwise maintain the fault isolation state.

[0047] Check and verify the recovery judgment parameters of the power electronic transformer to determine whether they meet the recovery conditions.

[0048] Preferably, extract the fault type label from the fault diagnosis result and determine whether it belongs to the power electronic transformer type of fault, such as "DC side bus undervoltage", "AC side overcurrent", "power module overtemperature", etc.; then, according to the power electronic transformer recovery condition expression, extract the preset parameter items, such as "bus voltage value", "power module temperature", "input and output current values", etc.; then, based on the equipment factory technical parameter specification, industry standard documents, and operation and maintenance experience data of the power electronic transformer, set the recovery judgment thresholds for each parameter item respectively. For example, compare the current bus voltage value with the lower limit of 300V, the power module temperature with the upper limit of 85°C, and the output current value with the rated value; then, perform a logical combination of the comparison results according to the power electronic transformer recovery condition expression, such as "bus voltage value ≥ 300V and power module temperature ≤ 85°C and output current value within the rated range"; finally, judge whether it meets the recovery conditions of the power electronic transformer according to the logical combination result. If all conditions are met, the result is "meeting the recovery conditions", and if the conditions are not met, the result is "not meeting the recovery conditions".

[0049] When the recovery conditions are met, perform the power-off and reconnection operations on the DC port power electronic transformer.

[0050] Preferably, a control instruction is issued to turn off the DC-side input switch of the DC-port power electronic transformer, disconnect the main circuit power supply, and record the timestamp of the disconnection operation; then, it is confirmed through the acquisition device that the DC voltage has dropped to a safe level, for example, it is confirmed that the voltage is lower than 50V, and the states of key devices are monitored to be idle or non-excited; then, a reconnect preparation process is executed, including voltage pre-charging of the main circuit of the DC-port power electronic transformer, buffer charging of the bus capacitor, gate drive initialization, etc.; after all preparation actions are completed, a control instruction is sent to close the DC-side input switch and reconnect the power electronic transformer to the DC power supply; finally, the initial operation data after reconnection is read and recorded, for example, key parameters such as voltage, current, and temperature within 5 seconds after reconnection are recorded for subsequent stability analysis or comparison.

[0051] S6. Continuously and real-time collect operation status data, and combine the operation status data set with the abnormal parameters recorded in the fault diagnosis result, and use an adaptive update algorithm to dynamically correct the short-circuit current response mechanism.

[0052] Continuously and real-time collect the updated operation status data set, and maintain the timeliness of the data through a data synchronization mechanism.

[0053] It should be noted that a collection instruction is configured to regularly read operation status data such as voltage, current, power, temperature, and insulation resistance value at a sampling period of 100ms and write them into the operation status data set; then, a queue cache mechanism based on the timestamp is established to bind the collection time and the data value, for example, stored in the format of "202X-04-15 10:30:00.100 - voltage value = 420V"; then, an existing time synchronization protocol such as the Network Time Protocol (NTP) is used to align the clocks of the collection terminals to ensure the time consistency among multiple data sources; then, the new data in the operation status data set is directionally synchronized to the target processing end through a publish-subscribe mechanism or HTTP push, etc., to ensure that the latest data can be reached within 200ms; finally, the synchronization result is verified, including comparing the continuity of the timestamp, the integrity of the fields, and the legality of the data value, for example, it is judged that the synchronized voltage value should be between 300 - 500V, and an abnormal record is triggered and re-synchronized if it exceeds the range.

[0054] Match the updated operation status data set with the abnormal parameters in the fault diagnosis result, and extract the current short-circuit current response mechanism parameters.

[0055] It should be noted that in the fault diagnosis result, the fault type label to which the abnormal parameter belongs is "short - circuit fault"; then, extract the characteristic items related to the short - circuit fault from the abnormal parameters, such as "maximum short - circuit current value", "short - circuit current duration", "current rise rate", etc.; then, search for the current data sequence within the corresponding time - stamp interval in the updated operating state data set, and extract the current change curve within 500 ms before and after the fault occurs; then, extract the corresponding characteristic values by name, for example, calculate the maximum short - circuit current value as the current peak value within this interval, the short - circuit current duration as the time - period length during which the current is greater than twice the rated value, and the current rise rate as the maximum slope of the current rising from the steady state to the peak value; finally, mark the "maximum short - circuit current value", "short - circuit current duration", and "current rise rate" extracted as the current short - circuit current response mechanism parameters.

[0056] Use an adaptive update algorithm to dynamically adjust the current short - circuit current response mechanism parameters.

[0057] Furthermore, obtain the current short - circuit current response mechanism parameters, compare them with the updated operating state data set, and obtain the parameter offset information.

[0058] It should be noted that extract each value in the current short - circuit current response mechanism parameters, including the maximum short - circuit current value, short - circuit current duration, and current rise rate, for example, "maximum short - circuit current value = 420 A", "short - circuit current duration = 180 ms", "current rise rate = 2800 A / s"; then, in the updated operating state data set, find the historical statistical reference value or rated threshold of the corresponding parameter, such as "maximum short - circuit current value reference value = 400 A", "short - circuit current duration reference value = 150 ms", "current rise rate reference value = 2500 A / s"; then, calculate the offset of each parameter item by item, for example, offset = current value−reference value, and obtain "maximum short - circuit current value offset = 20 A", "short - circuit current duration offset = 30 ms", "current rise rate offset = 300 A / s"; finally, form the parameter offset information with each offset and output it, including the offset direction and the magnitude of the offset value.

[0059] Identify the current short - circuit current response mechanism parameter items that need to be adjusted according to the parameter offset information.

[0060] It should be noted that the offset values and offset directions of each parameter in the read parameter offset information are read, for example, "Maximum short-circuit current value offset = +20A", "Short-circuit current duration offset = +30ms", "Current rise rate offset = +300A / s"; then the parameter offset threshold range is set, for example, "Allowable offset range of maximum short-circuit current value ±10A", "Allowable offset range of short-circuit current duration ±20ms", "Allowable offset range of current rise rate ±200A / s"; then the offset of each parameter is compared with its corresponding allowable offset range to identify the parameter items that exceed the threshold range. For example, it is determined that "Maximum short-circuit current value offset = +20A > +10A" is out of limit, "Short-circuit current duration offset = +30ms > +20ms" is out of limit, "Current rise rate offset = +300A / s > +200A / s" is out of limit; finally, the parameter items that exceed the allowable offset range are marked as the response mechanism parameter items to be adjusted.

[0061] Based on the current operating state data, select the corresponding adaptive update algorithm and calculate the update values of each parameter to be adjusted.

[0062] It should be noted that the current operating state data is analyzed to extract the data items related to the parameters to be adjusted, such as "Current load current", "Voltage fluctuation", "Temperature change", etc.; then an adaptive update algorithm suitable for the current situation is selected, such as "Least mean square error algorithm", "Recursive least squares algorithm", etc.; then the extracted data items are used as the input of the algorithm and calculated according to the selected adaptive update algorithm. For example, for the least mean square error algorithm, the error values of each parameter to be adjusted (such as the updated maximum short-circuit current value, short-circuit duration, and current rise rate) are calculated, and the parameters are adjusted according to the error values; then according to the calculation results, the update values of each parameter to be adjusted are obtained, and finally the calculated update values of each parameter to be adjusted are applied to the current operation to ensure the adaptability of the response mechanism.

[0063] Apply the update values to the original response mechanism parameters to generate a new set of response mechanism parameters.

[0064] It should be noted that real-time acquisition data from the initial configuration and operating status of the device is obtained, including response mechanism parameters such as "maximum short-circuit current value", "response time", "voltage fluctuation range", etc. Then, the updated values of each parameter to be adjusted calculated by the adaptive update algorithm are obtained, such as "current rise rate", "current threshold", etc.; Next, each updated value is respectively replaced with the corresponding item of the original response mechanism parameter to ensure that each value in the original response mechanism parameter is replaced by the updated value; Subsequently, the integrity and consistency of the new response mechanism parameter set are verified to ensure that all parameter items have been successfully updated and there is no missing or incorrect data; Finally, a new response mechanism parameter set is generated, which will be stored as the latest response mechanism parameter set used in the current operation and used for subsequent calculations or adjustments.

[0065] This embodiment also provides a computer device applicable to the case of the DC port short-circuit current protection method of a power electronic transformer, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the DC port short-circuit current protection method of the power electronic transformer proposed in the above embodiment.

[0066] The computer device can be a terminal. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, a carrier network, NFC (near-field communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer covering the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0067] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the method for protecting the short-circuit current of the DC port of the power electronic transformer as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, abbreviated as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, abbreviated as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, abbreviated as EPROM), programmable read-only memory (Programmable Red-Only Memory, abbreviated as PROM), read-only memory (Read-Only Memory, abbreviated as ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0068] In summary, the present invention: real-time collects the updated operation state data and combines the abnormal parameters in the fault diagnosis result, and uses an adaptive update algorithm to dynamically adjust the short-circuit current response mechanism parameters. This step realizes the automatic adjustment of the protection mechanism under different working conditions to ensure that the protection strategy can always adapt to the current operation state. The function is to optimize the protection mechanism in real time, with strong adaptability and response ability. It can not only reduce the misoperation and missed operation caused by parameter changes, but also continuously and effectively protect the power electronic transformer under different load, ambient temperature and voltage fluctuation conditions.

[0069] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A method for protecting the short-circuit current of the DC port of a power electronic transformer, characterized in that: including, collecting current data, voltage data and temperature data of the DC port to form an operating state data set; performing real-time analysis on the operating state data set using a density clustering algorithm to identify abnormal features related to the short-circuit current of the DC port and generate an abnormal intensity index; when detecting an abnormal short-circuit current at the DC port, generating a control instruction to drive the ultrafast electronic switch to perform a cut-off action, and recording and uploading the time, location and abnormal parameters of the fault occurrence, generating a fault diagnosis result and a maintenance suggestion; when the fault diagnosis result and the maintenance suggestion confirm that the recovery condition is met, starting the recovery process to reconnect the power electronic transformer to the DC power supply circuit; continuously and real-time collecting operating state data, combining the operating state data set with the abnormal parameters recorded in the fault diagnosis result, and dynamically correcting the short-circuit current response mechanism using an adaptive update algorithm.

2. The short-circuit current protection method for the DC port of the power electronic transformer according to claim 1, characterized in that: Aligning the current, voltage and temperature data of the DC port according to the time stamp to generate a synchronized original data set, obtaining a structured data table through normalization processing, and forming an operating state data set through periodic caching.

3. The DC port short-circuit current protection method for a power electronic transformer according to claim 1, characterized in that: The specific steps for generating the abnormal intensity index are as follows: Performing a sliding window process on the operating state data set, continuously intercepting data segments of a specific time length to form short-period data sequences, and extracting multi-dimensional features to obtain a set of short-period operating state feature vectors; Using the density clustering method to perform clustering analysis on the set of short-period operating state feature vectors and marking abnormal features; Calculating the credibility of the abnormal features through a confidence evaluation mechanism to generate an abnormal intensity index.

4. The DC port short-circuit current protection method for a power electronic transformer according to claim 1, characterized in that: The specific steps for generating the control instruction are as follows: Extracting the current intensity index from the abnormal intensity index to generate a preliminary alarm signal, and combining the real-time current change rate to judge the fault level and generate a level identifier; Matching the preset action strategy according to the level identifier, determining the corresponding switch trigger mode and action parameters, and generating a control instruction.

5. The DC port short-circuit current protection method for a power electronic transformer according to claim 1, characterized in that: The specific steps for generating the fault diagnosis result and the maintenance suggestion are as follows: After receiving the control instruction, controlling the ultrafast electronic switch to disconnect the circuit to complete the cut-off action, recording the fault occurrence time and location, and extracting abnormal parameters at the same time; Summarizing the fault time, location and abnormal parameters to form fault information, comparing them using preset rules to generate a fault diagnosis result, and searching for the maintenance process to generate a maintenance suggestion.

6. The short-circuit current protection method for the DC port of the power electronic transformer according to claim 5, characterized in that: The specific steps of the recovery process are as follows: Defining the recovery condition according to the fault diagnosis result and the maintenance suggestion; Checking and verifying the recovery determination parameters of the power electronic transformer to judge whether they meet the recovery condition; When the recovery condition is met, performing the power-off and reconnection operations of the power electronic transformer at the DC port.

7. The short-circuit current protection method for the DC port of the power electronic transformer according to claim 6, characterized in that: The specific steps for continuously and real-time collecting operating state data, combining the operating state data set with the abnormal parameters recorded in the fault diagnosis result, and dynamically correcting the short-circuit current response mechanism using an adaptive update algorithm are as follows: Continuously and real-time collecting the updated operating state data set and maintaining the timeliness of the data through a data synchronization mechanism; Matching the updated operating state data set with the abnormal parameters in the fault diagnosis result to extract the current short-circuit current response mechanism parameters; The adaptive update algorithm is used to dynamically adjust the parameters of the current short - circuit current response mechanism.

8. The DC port short-circuit current protection method for a power electronic transformer according to claim 7, characterized in that: The adaptive update algorithm is used to dynamically adjust the parameters of the current short - circuit current response mechanism. The specific steps are as follows: Obtain the parameters of the current short - circuit current response mechanism, compare them with the updated operating state data set, and obtain the parameter offset information; According to the parameter offset information, identify the current short - circuit current response mechanism parameter items to be adjusted; Based on the current operating state data, select the corresponding adaptive update algorithm and calculate the updated values of each parameter to be adjusted; Apply the updated values to the original response mechanism parameters to generate a new response mechanism parameter set.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the short - circuit current protection method for the DC port of the power electronic transformer according to any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the short - circuit current protection method for the DC port of the power electronic transformer according to any one of claims 1 to 8.

Images