A transformer winding state evaluation method and device, a storage medium and equipment

Abstract

The present application relates to the technical field of winding fault detection, and discloses a transformer winding state evaluation method, device, storage medium and equipment, comprising: through real-time acquisition of the current vibration signal of the transformer to be tested and comparison with the health baseline, quantifying the signal characteristics by using LZ complexity, combining the relative deviation rate calculation and the preset range criterion, realizing the on-line evaluation of the winding state under voltage, which breaks through the limitation of traditional power-off detection, effectively shortens the detection period and improves the detection real-time performance, reduces the environmental noise interference through vibration signal feature extraction and quantitative analysis, improves the detection accuracy and adaptability, simplifies the detection process, reduces the dependence on professional detection personnel, realizes the rapid, accurate and automatic evaluation of the transformer winding state, provides efficient and reliable technical support for transformer operation and maintenance, and significantly improves the equipment operation safety and maintenance efficiency.

Description

Technical Field

[0001] This invention relates to the field of winding fault detection technology, and in particular to a method, apparatus, storage medium and device for evaluating the condition of transformer windings. Background Technology

[0002] Transformer windings are susceptible to transient overvoltages or high short-circuit currents during operation, which can lead to a decrease in short-circuit withstand capability or even damage to the windings, causing major equipment accidents.

[0003] Traditional winding fault detection techniques, such as the short-circuit impedance method, frequency response method, and low-voltage pulse method, all have significant drawbacks: First, they require detection to be carried out under power outage conditions, resulting in long detection cycles and poor real-time performance, making it difficult to meet the continuous monitoring needs of transformers under energized operation. Second, the detection results are easily affected by environmental interference, equipment aging, and other factors, resulting in insufficient accuracy and adaptability. Third, the detection process requires professional personnel to operate complex equipment and analyze data, which places high demands on the technical skills of the detection personnel and increases the difficulty of operation and maintenance. Summary of the Invention

[0004] Based on this, it is necessary to address the above-mentioned problems by proposing a method, device, storage medium, and equipment for evaluating the condition of transformer windings. This method overcomes the limitations of traditional power outage detection, effectively shortens the detection cycle, and improves the real-time performance of detection. Through vibration signal feature extraction and quantitative analysis, it reduces environmental noise interference, improves detection accuracy and adaptability, simplifies the detection process, reduces reliance on professional testing personnel, and achieves rapid, accurate, and automated evaluation of transformer winding conditions. This provides efficient and reliable technical support for transformer operation and maintenance, and significantly improves equipment operation safety and maintenance efficiency.

[0005] To achieve the above objectives, the present invention provides a method for evaluating the condition of transformer windings in a first aspect, the method comprising: Acquire the current vibration signal and health baseline of the transformer under test; Determine the LZ complexity based on the current vibration signal; The relative deviation rate is determined based on the LZ complexity and the health baseline. The winding state of the transformer under test is determined based on the relative deviation rate and the preset range.

[0006] Optionally, determining the LZ complexity based on the current vibration signal includes: The current vibration signal is subjected to binary coarsening processing to obtain a binary sequence; Convert the binary sequence into a binary string; The difference between the substring consisting of the first r characters of the binary string and the dictionary is taken as the current string, where the initial value of r is 1, and the initial value of the dictionary is the substring consisting of the first 1 character of the binary string; If the dictionary contains the current string, increment the complexity by 0; the initial value of the complexity is 0. If the dictionary does not contain the current string, let the dictionary be equal to the substring consisting of the first r characters of the binary string, and increase the complexity by 1; Let r = r + 1, return to the step of taking the difference between the substring composed of the first r characters in the binary string and the dictionary as the current string, until r is greater than the total number of characters in the binary string, and take the complexity as the target complexity; The LZ complexity is determined based on the target complexity and the total number of characters in the binary string.

[0007] Optionally, determining the relative deviation rate based on the LZ complexity and the health baseline includes: The absolute value of the difference between the LZ complexity and the healthy baseline is taken as the LZ complexity deviation; The quotient between the LZ complexity deviation and the healthy baseline is used as the relative deviation rate.

[0008] Optionally, determining the winding state of the transformer under test based on the relative deviation rate and a preset range includes: If the relative deviation rate is less than or equal to the lower limit of the preset range, the winding state of the transformer under test is determined to be normal. If the relative deviation rate is greater than the lower limit of the preset range and less than or equal to the upper limit of the preset range, the winding state of the transformer under test is determined to be an alarm state. If the relative deviation rate is greater than the upper limit of the preset range, the winding state of the transformer under test is determined to be a fault state.

[0009] Optionally, obtain the health baseline of the transformer under test, including: When the transformer under test is in a healthy state, acquire multiple sets of healthy vibration signals of the transformer under test; Based on each set of healthy vibration signals, determine the health LZ complexity corresponding to each set of healthy vibration signals; The average of all healthy LZ complexities is used as the health baseline.

[0010] Optionally, before determining the winding state of the transformer under test based on the relative deviation rate and the preset range, the method further includes: Determine the sample entropy based on the current vibration signal; The correction value is determined based on the sample entropy and the LZ complexity; The relative deviation rate is corrected based on the correction value to obtain the corrected relative deviation rate.

[0011] Optionally, before determining the winding state of the transformer under test based on the relative deviation rate and the preset range, the method further includes: The LZ complexity is input into a preset complexity state probability prediction model to obtain the complexity state probability of the LZ complexity. Based on the complexity state probability, the preset range is optimized and adjusted to obtain the optimized preset range.

[0012] To achieve the above objectives, the present invention provides a transformer winding condition assessment device in a second aspect, the device comprising: The acquisition module is used to acquire the current vibration signal of the transformer under test; The first determining module is used to determine the LZ complexity based on the current vibration signal; The second determining module is used to determine the relative deviation rate based on the LZ complexity and the health baseline; The third determining module is used to determine the winding state of the transformer under test based on the relative deviation rate and the preset range.

[0013] To achieve the above objectives, the present invention provides, in a third aspect, a computer-readable storage medium storing a computer program, which, when executed by a processor, causes the processor to perform the transformer winding state assessment method as described in any one of the first aspects.

[0014] To achieve the above objectives, the present invention provides a computer device in a fourth aspect, including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform the transformer winding state assessment method as described in any one of the first aspects.

[0015] The embodiments of the present invention have the following beneficial effects: The above method acquires the current vibration signal and healthy baseline of the transformer under test, then determines the LZ complexity based on the current vibration signal, then determines the relative deviation rate based on the LZ complexity and the healthy baseline, and finally determines the winding state of the transformer under test based on the relative deviation rate and a preset range; that is, by acquiring the current vibration signal of the transformer under test in real time and comparing it with the healthy baseline, using the LZ complexity to quantify the signal characteristics, and combining the relative deviation rate calculation and the preset range criterion, the energized online assessment of the winding state is realized. This method breaks through the limitations of traditional power outage detection, effectively shortens the detection cycle and improves the real-time performance of the detection. Through vibration signal feature extraction and quantitative analysis, environmental noise interference is reduced, and the detection accuracy and adaptability are improved. At the same time, the detection process is simplified, the dependence on professional testing personnel is reduced, and a rapid, accurate, and automated assessment of the transformer winding state is realized, providing efficient and reliable technical support for transformer operation and maintenance, and significantly improving equipment operation safety and maintenance efficiency. Attached Figure Description

[0016] 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 some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] in: Figure 1 This is a schematic diagram of a transformer winding condition assessment method according to an embodiment of this application; Figure 2 This is a schematic diagram of a transformer winding condition assessment device according to an embodiment of this application; Figure 3 This is a diagram showing the internal structure of a computer device in some embodiments. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Transformer windings are susceptible to transient overvoltages or high short-circuit currents during operation, which can lead to a decrease in short-circuit withstand capability or even damage to the windings, causing major equipment accidents.

[0020] Traditional winding fault detection techniques, such as the short-circuit impedance method, frequency response method, and low-voltage pulse method, all have significant drawbacks: First, they require detection to be carried out under power outage conditions, resulting in long detection cycles and poor real-time performance, making it difficult to meet the continuous monitoring needs of transformers under energized operation. Second, the detection results are easily affected by environmental interference, equipment aging, and other factors, resulting in insufficient accuracy and adaptability. Third, the detection process requires professional personnel to operate complex equipment and analyze data, which places high demands on the technical skills of the detection personnel and increases the difficulty of operation and maintenance.

[0021] To address the aforementioned issues, this application proposes a method, apparatus, storage medium, and device for assessing the condition of transformer windings. This method overcomes the limitations of traditional power outage detection, effectively shortens the detection cycle, and improves real-time performance. Through vibration signal feature extraction and quantitative analysis, it reduces environmental noise interference, enhances detection accuracy and adaptability, simplifies the detection process, reduces reliance on professional testing personnel, and achieves rapid, accurate, and automated assessment of transformer winding conditions. This provides efficient and reliable technical support for transformer operation and maintenance, significantly improving equipment safety and maintenance efficiency. The specific implementation principles will be detailed in the following embodiments.

[0022] This application provides a method for evaluating the condition of transformer windings in its first aspect.

[0023] Please see Figure 1 This is a schematic diagram of a transformer winding condition assessment method according to an embodiment of this application. The method includes: Step 110: Obtain the current vibration signal and health baseline of the transformer under test.

[0024] The health baseline can be set in advance by the operator based on extensive experience, experiments, or statistics. Alternatively, it can be set in advance by the operator according to actual needs.

[0025] Regarding the current method of acquiring vibration signals, in some embodiments, it can be based on installing a vibration sensor on the housing of the transformer under test to collect the current vibration signal of the transformer under test through the vibration sensor.

[0026] Furthermore, regarding the installation location of the vibration sensor, in some embodiments, when the transformer under test is a two-winding transformer, the vibration sensor can be installed on the low-voltage side; when the transformer under test is a three-winding transformer, the vibration sensor can be installed on the medium-voltage side.

[0027] It should be noted that because the current in the low-voltage and medium-voltage windings is relatively large, the vibration generated is also relatively large. Therefore, vibration sensors can be installed on the low-voltage and medium-voltage sides. The vibration sensors should be installed directly opposite the windings. Two vibration sensors can be installed at 1 / 4 and 3 / 4 height of each winding, respectively. The vibration sensor should collect vibration signals for more than 1 hour to obtain enough data and improve the accuracy of the evaluation.

[0028] Step 120: Determine the LZ complexity based on the current vibration signal.

[0029] For LZ complexity, in some embodiments, existing methods for determining LZ complexity can be used.

[0030] Step 130: Determine the relative deviation rate based on LZ complexity and healthy baseline.

[0031] Step 140: Determine the winding state of the transformer under test based on the relative deviation rate and the preset range.

[0032] The preset range can be set in advance by the operator based on a large amount of experience, experiments or statistics. Of course, it can also be set in advance by the operator according to actual needs.

[0033] Regarding the value of the preset range, in some embodiments, this application preferably sets the preset range to (0.1, 0.25).

[0034] In this embodiment, the current vibration signal and healthy baseline of the transformer under test are acquired. Then, the LZ complexity is determined based on the current vibration signal. Next, the relative deviation rate is determined based on the LZ complexity and the healthy baseline. Finally, the winding state of the transformer under test is determined based on the relative deviation rate and a preset range. That is, by acquiring the current vibration signal of the transformer under test in real time and comparing it with the healthy baseline, using the LZ complexity to quantify the signal characteristics, and combining the relative deviation rate calculation and preset range criteria, the energized online assessment of the winding state is realized. This method breaks through the limitations of traditional power outage detection, effectively shortens the detection cycle and improves the real-time performance of the detection. Through vibration signal feature extraction and quantitative analysis, environmental noise interference is reduced, and the detection accuracy and adaptability are improved. At the same time, the detection process is simplified, the dependence on professional testing personnel is reduced, and a rapid, accurate, and automated assessment of the transformer winding state is realized. This provides efficient and reliable technical support for transformer operation and maintenance, and significantly improves equipment operation safety and maintenance efficiency.

[0035] In addition to the aforementioned beneficial effects, this transformer winding condition assessment method also has the following advantages: Early fault warning: This method can monitor the transformer winding condition in real time and, by continuously tracking the changing trend of the relative deviation rate, promptly issue early warning signals when the winding condition is faulty. This allows maintenance personnel to take measures before the fault develops to a serious level, preventing further deterioration and reducing economic losses and power outage time caused by equipment damage; Preventive maintenance decision support: Based on long-term monitoring data and winding condition assessment results, more scientific and reasonable preventive maintenance plans can be formulated, such as based on the frequency and trend of changes in the winding condition. By scheduling maintenance time and replacing parts in advance, the targetedness and effectiveness of maintenance work can be improved, extending the service life of transformers. Big data accumulation: This method continuously collects a large amount of transformer vibration signal data during operation and calculates the corresponding LZ complexity and relative deviation rate. This data constitutes a rich database, providing valuable information for subsequent research and analysis. In-depth mining and analysis of this large amount of data can further reveal the mechanism and laws of transformer winding faults, providing a basis for improving transformer design and manufacturing processes. Trend analysis: The accumulated data can be used for trend analysis to predict future changes in the state of transformer windings, such as by analyzing phase... The deviation rate over time curve can determine whether the winding condition is gradually deteriorating and at what rate. This helps in taking preventative measures and ensuring the stable operation of the power system. It is compatible with various transformer types: this method is not only applicable to two-winding and three-winding transformers, but also to other types, such as autotransformers. By appropriately adjusting the installation location and number of vibration sensors and the corresponding data processing algorithm, the winding condition can be assessed. This demonstrates the method's strong compatibility and ability to meet the monitoring needs of different transformer types. It reduces operation and maintenance costs: because this method achieves online energized assessment of the transformer winding condition, it reduces... This reduces the frequency and duration of power outage inspections, thereby decreasing economic losses caused by power outages. It also simplifies the inspection process, reduces reliance on specialized inspection personnel, and lowers labor costs. Furthermore, early fault warnings and preventative maintenance can avoid expensive repairs and replacements due to severe equipment damage. It also improves power supply reliability: timely and accurate winding condition assessment helps quickly identify and resolve potential problems, reducing the probability of transformer failures and thus improving power supply reliability. A stable power supply is crucial for ensuring the normal operation of social production and life, and can prevent losses such as production interruptions and data loss caused by power outages, resulting in significant socio-economic benefits.

[0036] In one feasible implementation, step 120 in the above embodiment, determining the LZ complexity based on the current vibration signal, includes: performing binary coarse-grained processing on the current vibration signal to obtain a binary sequence; converting the binary sequence into a binary string; using the difference between the substring composed of the first r characters in the binary string and the dictionary as the current string, where r is initially set to 1 and the dictionary is initially set to the substring composed of the first 1 character in the binary string; if the dictionary contains the current string, incrementing the complexity by 0, with the initial value of the complexity being 0; if the dictionary does not contain the current string, setting the dictionary equal to the substring composed of the first r characters in the binary string, and incrementing the complexity by 1; setting r = r + 1, and returning to execute the step of using the difference between the substring composed of the first r characters in the binary string and the dictionary as the current string, until r is greater than the total number of characters in the binary string, and using the complexity as the target complexity; determining the LZ complexity based on the target complexity and the total number of characters in the binary string.

[0037] It should be noted that the difference in this application refers to the difference between strings, not the difference between sets. For example, if the substring formed by the first r characters of a binary string is 0111 and the dictionary is 011, then the difference between the substring formed by the first r characters of the binary string and the dictionary is 1, which means the current string is 1. If the substring formed by the first r characters of a binary string is 0 and the dictionary is 0, then the difference between the substring formed by the first r characters of the binary string and the dictionary is NULL (i.e., empty), which means the current string is NULL.

[0038] For binary coarsening, in some embodiments, the average value of all data points in the current vibration signal can be determined first, and then data points greater than the average value can be set to 1, and data points less than the average value can be set to 0, so as to obtain a binary sequence.

[0039] In some embodiments, the LZ complexity can be determined using a formula. Determine the complexity of LZ; where, For LZ's complexity, For the target complexity, It is the logarithm to the base 2. This represents the total number of characters in the binary string.

[0040] In the embodiments of this application, the LZ complexity determination method provides an accurate and scientific quantitative means for evaluating the state of transformer windings, thereby enhancing the reliability and effectiveness of the evaluation.

[0041] Understandably, this method achieves several key benefits: First, it accurately quantifies signal characteristics. This involves coarsely processing the current vibration signal into a binary sequence, converting it into a binary string, and then progressively determining the target complexity according to specific rules. Finally, the LZ complexity is determined by combining the total number of characters in the binary string. This series of steps precisely extracts key features from the vibration signal, transforming complex vibration signals into quantifiable indicators, providing an accurate data foundation for subsequent winding condition assessment. Second, it enhances assessment reliability. This method of determining LZ complexity based on rigorous mathematical logic and rules avoids the subjective interference that may exist in traditional methods, making the assessment results more objective and reliable. Regardless of different environmental conditions or various types of transformers, it can accurately calculate LZ complexity according to a unified standard, thus providing a stable and reliable basis for winding condition assessment. Third, it improves assessment effectiveness. The LZ complexity determined by this method truly reflects the actual operating state of the transformer windings. By comparing it with a healthy baseline, combined with relative deviation rate calculation and preset range criteria, it can promptly detect subtle changes in the winding state, effectively determine whether there are potential faults in the windings, improve the effectiveness of transformer winding condition assessment, and help take preventative measures to ensure the safe and stable operation of the transformer.

[0042] In one feasible implementation, step 130 in the above embodiment, which determines the relative deviation rate based on the LZ complexity and the healthy baseline, includes: taking the absolute value of the difference between the LZ complexity and the healthy baseline as the LZ complexity deviation; and taking the quotient between the LZ complexity deviation and the healthy baseline as the relative deviation rate.

[0043] In the embodiments of this application, the method for determining the relative deviation rate provides a key basis for accurately assessing the state of transformer windings, effectively improving the accuracy and practicality of the assessment.

[0044] Understandably, this method achieves several benefits: First, it precisely quantifies the difference in winding condition by calculating the absolute value of the difference between the LZ complexity and the healthy baseline. This yields the LZ complexity deviation, and then the relative deviation rate is calculated by dividing this deviation by the healthy baseline. This method accurately quantifies the difference between the current winding condition and the healthy condition, transforming complex winding condition changes into intuitive numerical values, providing strong support for accurate subsequent judgment of the winding condition. Second, it enhances assessment accuracy by using rigorous mathematical calculations to determine the relative deviation rate, avoiding errors caused by subjective judgment. Regardless of operating conditions or environmental conditions, it can accurately calculate based on a unified standard, making the assessment of the winding condition more accurate and reliable, and helping to promptly identify potential faults. Third, it improves the practicality of the assessment by using the relative deviation rate as a quantitative indicator to intuitively reflect the degree of deviation in the winding condition. Maintenance personnel can quickly understand the winding condition based on this indicator and make reasonable decisions based on preset ranges, such as whether further maintenance is needed. This improves the practicality of the assessment results in actual operation and maintenance, ensuring the safe and stable operation of the transformer.

[0045] In one feasible implementation, step 140 in the above embodiment, which determines the winding state of the transformer under test based on the relative deviation rate and a preset range, includes: determining the winding state of the transformer under test to be normal when the relative deviation rate is less than or equal to the lower limit of the preset range; determining the winding state of the transformer under test to be alarm when the relative deviation rate is greater than the lower limit of the preset range and less than or equal to the upper limit of the preset range; and determining the winding state of the transformer under test to be faulty when the relative deviation rate is greater than the upper limit of the preset range.

[0046] In the embodiments of this application, the winding state determination method provides a clear and effective judgment standard for the accurate classification and evaluation of transformer winding state, which greatly improves the accuracy and practicality of state evaluation.

[0047] Understandably, precise classification and assessment: By meticulously comparing the relative deviation rate with the lower and upper limits of a preset range, three states—normal, alarm, and fault—are clearly defined. This classification method can accurately locate the current state of the transformer windings, avoiding ambiguity in state judgment and providing a clear basis for subsequent operation and maintenance decisions. Enhanced assessment accuracy: Based on rigorous numerical comparison logic, the state is determined according to the precise relationship between the relative deviation rate and the preset range, avoiding the influence of subjective judgment or experience errors. Regardless of the transformer's operating environment or conditions, accurate assessment can be performed according to a unified standard, ensuring objective and reliable assessment results and timely detection of potential fault hazards. Improved assessment practicality: Clear state classification results allow operation and maintenance personnel to quickly and intuitively understand the transformer winding status. For normal states, normal monitoring can continue; for alarm states, attention can be paid in advance and further inspections can be arranged; for fault states, immediate measures can be taken. This highly practical assessment method helps to rationally allocate operation and maintenance resources and ensure the safe and stable operation of the transformer.

[0048] In one feasible implementation, step 110 in the above embodiment, obtaining the health baseline of the transformer under test, includes: acquiring multiple sets of healthy vibration signals of the transformer under test when the transformer under test is in a healthy state; determining the health LZ complexity corresponding to each set of healthy vibration signals based on each set of healthy vibration signals; and using the average value of all health LZ complexities as the health baseline.

[0049] It should be noted that each time the winding condition is assessed, the health baseline of the transformer under test needs to be redefined. The multiple sets of health vibration signals used for subsequent re-determination include the LZ complexity of assessing the winding condition as normal before the last winding condition assessment.

[0050] In this embodiment of the application, the method of obtaining the health baseline provides a scientific and reasonable benchmark for evaluating the condition of transformer windings, thereby enhancing the accuracy and reliability of the evaluation.

[0051] Understandably, acquiring multiple sets of healthy vibration signals while the transformer under test is in a healthy state comprehensively reflects the vibration characteristics of the transformer under normal operating conditions, ensuring the authenticity and representativeness of the collected data and laying a solid foundation for determining the health baseline. Subsequently, the corresponding health LZ complexity is determined based on each set of healthy vibration signals. By quantifying the characteristics of the vibration signals, the complex vibration information is transformed into comparable numerical indicators, facilitating a precise characterization of the health state. Finally, the average value of all health LZ complexities is used as the health baseline. This approach comprehensively considers multiple sets of data, avoiding the randomness and errors that may exist in a single set of data, making the health baseline more stable and reliable. It can more accurately reflect the vibration characteristics of the transformer in a healthy state, providing a scientific and reasonable benchmark for subsequent comparison with the LZ complexity of the current vibration signal, thereby improving the accuracy and reliability of transformer winding condition assessment.

[0052] In one feasible implementation, before determining the winding state of the transformer under test based on the relative deviation rate and the preset range in step 140 of the above embodiment, the method further includes: determining the sample entropy based on the current vibration signal; determining the correction value based on the sample entropy and LZ complexity; and correcting the relative deviation rate based on the correction value to obtain the corrected relative deviation rate.

[0053] In some embodiments, the correction value can be determined using a formula. Determine the correction value; where, This is a correction value. For sample entropy, and These are the first constant and the second constant, respectively. Let LZ be the complexity.

[0054] In the embodiments of this application, by correcting the relative deviation rate, the accuracy and reliability of winding condition assessment can be improved.

[0055] Understandably, before determining the winding state based on the relative deviation rate and preset range, the sample entropy is first determined based on the current vibration signal. The correction value is then determined by combining the sample entropy and LZ complexity. The relative deviation rate is then corrected to obtain the corrected relative deviation rate. The sample entropy can reflect the complexity and regularity of the signal. Combining it with LZ complexity to determine the correction value can comprehensively consider multiple characteristics of the vibration signal. Correcting the relative deviation rate eliminates the influence of some interference factors, making the final relative deviation rate used to judge the winding state more accurately reflect the true state of the winding, thereby improving the accuracy and reliability of the assessment.

[0056] In one feasible implementation, before determining the winding state of the transformer under test based on the relative deviation rate and the preset range in step 140 of the above embodiment, the method further includes: inputting the LZ complexity into the preset complexity state probability prediction model to obtain the complexity state probability of the LZ complexity; and optimizing and adjusting the preset range based on the complexity state probability to obtain the optimized preset range.

[0057] Here, the preset complexity state probability prediction model refers to a pre-trained model used to predict the output complexity state probability based on the LZ complexity of the input.

[0058] In some embodiments, a large number of healthy LZ complexities and fault LZ complexities can be obtained for training the pre-complexity state probability prediction model. Then, these large numbers of healthy LZ complexities and fault LZ complexities are input into the initial complexity state probability prediction model for training. Once the model has been trained to a certain extent, the pre-complexity state probability prediction model can be obtained.

[0059] In this embodiment of the application, the accuracy and adaptability of winding state assessment are improved by optimizing and adjusting the preset range.

[0060] Understandably, the process involves first obtaining a pre-defined complexity state probability prediction model, inputting the LZ complexity into the model to obtain the complexity state probability, and then optimizing and adjusting the pre-defined range based on this probability. The sample entropy reflects the signal characteristics, and the correction value is determined in conjunction with the LZ complexity. This approach comprehensively considers multiple features, eliminates some interference factors, and makes the relative deviation rate more accurately reflect the true state of the winding. Furthermore, optimizing the pre-defined range based on the complexity state probability makes the judgment criteria more closely aligned with the actual situation. The combined effect of these two approaches improves the accuracy and adaptability of the winding state assessment.

[0061] In a second aspect, this application provides a transformer winding condition assessment device.

[0062] Please see Figure 2 This is a schematic diagram of a transformer winding condition assessment device according to an embodiment of this application. The device 210 includes: The acquisition module 211 is used to acquire the current vibration signal of the transformer under test; The first determining module 212 is used to determine the LZ complexity based on the current vibration signal; The second determining module 213 is used to determine the relative deviation rate based on the LZ complexity and the health baseline; The third determining module 214 is used to determine the winding state of the transformer under test based on the relative deviation rate and the preset range.

[0063] In this embodiment of the application, the relevant contents of the above-mentioned acquisition module 211, first determination module 212, second determination module 213 and third determination module 214 can be found in the following references. Figure 1 The contents of the illustrated embodiments will not be repeated here.

[0064] It should be noted that the device 210 of this application also includes other modules. It can be understood that the method of this application and the device 210 have a one-to-one correspondence. Therefore, the other modules of the device 210 of this application are the contents corresponding to the method of this application in the above embodiments.

[0065] In this embodiment, the current vibration signal and healthy baseline of the transformer under test are acquired. Then, the LZ complexity is determined based on the current vibration signal. Next, the relative deviation rate is determined based on the LZ complexity and the healthy baseline. Finally, the winding state of the transformer under test is determined based on the relative deviation rate and a preset range. That is, by acquiring the current vibration signal of the transformer under test in real time and comparing it with the healthy baseline, using the LZ complexity to quantify the signal characteristics, and combining the relative deviation rate calculation and preset range criteria, the energized online assessment of the winding state is realized. This device breaks through the limitations of traditional power outage detection, effectively shortens the detection cycle and improves the real-time performance of the detection. Through vibration signal feature extraction and quantitative analysis, environmental noise interference is reduced, and the detection accuracy and adaptability are improved. At the same time, the detection process is simplified, the dependence on professional testing personnel is reduced, and a rapid, accurate, and automated assessment of the transformer winding state is realized. This provides efficient and reliable technical support for transformer operation and maintenance, and significantly improves equipment operation safety and maintenance efficiency.

[0066] In addition to the aforementioned beneficial effects, this transformer winding condition assessment device also has the following advantages: Early fault warning: The device can monitor the transformer winding condition in real time and, by continuously tracking the changing trend of the relative deviation rate, promptly issue an early warning signal when the winding condition is faulty. This allows maintenance personnel to take measures before the fault develops to a serious level, preventing further deterioration and reducing economic losses and power outage time caused by equipment damage; Preventive maintenance decision support: Based on long-term monitoring data and winding condition assessment results, more scientific and reasonable preventive maintenance plans can be formulated, such as predicting future faults based on the frequency and trend of changes in the winding condition. Scheduling maintenance time and replacing parts improves the targetedness and effectiveness of maintenance work, extending the service life of transformers. Big data accumulation: During operation, the device continuously collects a large amount of transformer vibration signal data and calculates the corresponding LZ complexity and relative deviation rate. This data constitutes a rich database, providing valuable information for subsequent research and analysis. In-depth mining and analysis of this large amount of data can further reveal the mechanism and laws of transformer winding faults, providing a basis for improving transformer design and manufacturing processes. Trend analysis: The accumulated data can be used for trend analysis to predict future changes in the transformer winding condition, such as by analyzing relative... The deviation rate curve over time can determine whether the winding condition is gradually deteriorating and at what rate. This helps in taking preventative measures and ensuring the stable operation of the power system. It is compatible with various transformer types: the device is not only suitable for two-winding and three-winding transformers, but also for other types, such as autotransformers. By appropriately adjusting the installation position and number of vibration sensors and the corresponding data processing algorithm, the winding condition can be assessed. This demonstrates the device's strong compatibility and ability to meet the monitoring needs of different types of transformers. It reduces operation and maintenance costs: because the device enables online energized assessment of the transformer winding condition, it reduces... The frequency and duration of power outage inspections are reduced, thereby minimizing economic losses caused by power outages. Simultaneously, the inspection process is simplified, reducing reliance on specialized inspection personnel and lowering labor costs. Furthermore, early fault warnings and preventative maintenance can avoid expensive repair and replacement costs due to severe equipment damage. Power supply reliability is improved: timely and accurate winding condition assessments help quickly identify and resolve potential problems, reducing the probability of transformer failures and thus enhancing power supply reliability. A stable power supply is crucial for ensuring the normal operation of social production and life, preventing losses such as production interruptions and data loss caused by power outages, resulting in significant socio-economic benefits.

[0067] In a third aspect, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform a transformer winding state assessment method as described in any of the first aspects.

[0068] This application provides a computer device in a fourth aspect, including a memory and a processor, the memory storing a computer program, which, when executed by the processor, causes the processor to perform a transformer winding state assessment method as described in any of the first aspects.

[0069] Figure 3 The diagram illustrates the internal structure of a computer device in some embodiments. This computer device may specifically be a terminal, a server, or a gateway. Figure 3 As shown, the computer device includes a processor, memory, and network interface connected via a system bus.

[0070] The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When executed by a processor, this computer program causes the processor to perform the steps in the above method embodiments. The internal memory may also store a computer program, which, when executed by a processor, causes the processor to perform the steps in the above method embodiments. Those skilled in the art will understand that... Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0071] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods.

[0072] Any references to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0073] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0074] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for evaluating the condition of transformer windings, characterized in that, The method includes: Acquire the current vibration signal and health baseline of the transformer under test; Determine the LZ complexity based on the current vibration signal; The relative deviation rate is determined based on the LZ complexity and the health baseline. The winding state of the transformer under test is determined based on the relative deviation rate and the preset range.

2. The transformer winding condition assessment method according to claim 1, characterized in that, The determination of LZ complexity based on the current vibration signal includes: The current vibration signal is subjected to binary coarsening processing to obtain a binary sequence; Convert the binary sequence into a binary string; The difference between the substring consisting of the first r characters of the binary string and the dictionary is taken as the current string, where the initial value of r is 1, and the initial value of the dictionary is the substring consisting of the first 1 character of the binary string; If the dictionary contains the current string, increment the complexity by 0; the initial value of the complexity is 0. If the dictionary does not contain the current string, let the dictionary be equal to the substring consisting of the first r characters of the binary string, and increase the complexity by 1; Let r = r + 1, return to the step of taking the difference between the substring composed of the first r characters in the binary string and the dictionary as the current string, until r is greater than the total number of characters in the binary string, and take the complexity as the target complexity; The LZ complexity is determined based on the target complexity and the total number of characters in the binary string.

3. The transformer winding condition assessment method according to claim 1, characterized in that, The determination of the relative deviation rate based on the LZ complexity and the health baseline includes: The absolute value of the difference between the LZ complexity and the healthy baseline is taken as the LZ complexity deviation; The quotient between the LZ complexity deviation and the healthy baseline is used as the relative deviation rate.

4. The transformer winding condition assessment method according to claim 1, characterized in that, The step of determining the winding state of the transformer under test based on the relative deviation rate and the preset range includes: If the relative deviation rate is less than or equal to the lower limit of the preset range, the winding state of the transformer under test is determined to be normal. If the relative deviation rate is greater than the lower limit of the preset range and less than or equal to the upper limit of the preset range, the winding state of the transformer under test is determined to be an alarm state. If the relative deviation rate is greater than the upper limit of the preset range, the winding state of the transformer under test is determined to be a fault state.

5. The transformer winding condition assessment method according to claim 1, characterized in that, Obtain the health baseline of the transformer under test, including: When the transformer under test is in a healthy state, acquire multiple sets of healthy vibration signals of the transformer under test; Based on each set of healthy vibration signals, determine the health LZ complexity corresponding to each set of healthy vibration signals; The average of all healthy LZ complexities is used as the health baseline.

6. The transformer winding condition assessment method according to claim 1, characterized in that, Before determining the winding state of the transformer under test based on the relative deviation rate and the preset range, the method further includes: Determine the sample entropy based on the current vibration signal; The correction value is determined based on the sample entropy and the LZ complexity; The relative deviation rate is corrected based on the correction value to obtain the corrected relative deviation rate.

7. The transformer winding condition assessment method according to claim 1, characterized in that, Before determining the winding state of the transformer under test based on the relative deviation rate and the preset range, the method further includes: The LZ complexity is input into a preset complexity state probability prediction model to obtain the complexity state probability of the LZ complexity. Based on the complexity state probability, the preset range is optimized and adjusted to obtain the optimized preset range.

8. A transformer winding condition assessment device, characterized in that, The device includes: The acquisition module is used to acquire the current vibration signal of the transformer under test; The first determining module is used to determine the LZ complexity based on the current vibration signal; The second determining module is used to determine the relative deviation rate based on the LZ complexity and the health baseline; The third determining module is used to determine the winding state of the transformer under test based on the relative deviation rate and the preset range.

9. A computer-readable storage medium, characterized in that, The system contains a computer program that, when executed by a processor, causes the processor to perform the transformer winding condition assessment method as described in any one of claims 1 to 7.

10. A computer device, characterized in that, It includes a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the transformer winding condition assessment method as described in any one of claims 1 to 7.

Images