Transformer air cooling fault diagnosis method and system based on Internet of Things technology
By analyzing power and temperature data of transformers and tap contacts using IoT technology, and utilizing DBSCAN clustering and weighted prediction algorithms, the lag problem in the existing technology of transformer air-cooled fault diagnosis has been solved, achieving more accurate temperature prediction and fault detection, and ensuring the safe and stable operation of transformers.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-26
- Publication Date
- 2026-04-07
AI Technical Summary
Existing methods for diagnosing transformer air-cooled faults rely on real-time monitoring and lack the ability to predict and anticipate temperature changes, resulting in insufficient or excessive cooling and an inability to detect transformer abnormalities in a timely manner, leading to inaccurate diagnostic results.
A transformer air-cooled fault diagnosis method based on Internet of Things technology is adopted. By acquiring power and temperature data sequences of the transformer and tap contacts, the DBSCAN clustering algorithm and weighted prediction algorithm are used to analyze power load fluctuation, temperature influence factor and heat distribution, predict transformer temperature changes and perform fault detection.
This improves the accuracy of transformer air-cooling fault detection, reduces temperature prediction errors, and ensures the safe and stable operation of transformers.
Smart Images
Figure CN121804697A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of measuring electrical variables, and in particular to a transformer air cooling fault diagnosis method and system based on Internet of Things technology. BACKGROUND
[0002] As an important device in the power system, the stable operation of the transformer is crucial for the safety and reliability of power transmission. With the change of power load, the load of the transformer also fluctuates, causing the internal temperature of the transformer to fluctuate with the load. In order to ensure the safety and efficiency of the transformer during operation, it is necessary to adjust its temperature through an effective cooling system to avoid equipment damage or performance degradation due to overheating. The existing transformer air cooling fault diagnosis methods mostly rely on real-time monitoring, lacking the ability to predict and predict the temperature changes of the transformer. Due to the hysteresis of the temperature response of the transformer, the traditional diagnostic system often fails to discover abnormal conditions of the transformer in time when facing sudden load changes, resulting in insufficient or excessive cooling, thus making the transformer air cooling fault diagnosis result inaccurate. SUMMARY
[0003] The present application provides a transformer air cooling fault diagnosis method and system based on Internet of Things technology to solve the existing problems: due to the hysteresis of the temperature response of the transformer, and the existing transformer air cooling fault diagnosis methods mostly rely on real-time monitoring, making the traditional diagnostic system often fails to discover abnormal conditions of the transformer in time when facing sudden load changes, resulting in insufficient or excessive cooling, thus making the transformer air cooling fault diagnosis result inaccurate.
[0004] The transformer air cooling fault diagnosis method and system based on Internet of Things technology of the present application adopts the following technical scheme: The present application proposes a transformer air cooling fault diagnosis method based on Internet of Things technology, which includes the following steps: Obtain the power data sequence and temperature data sequence of the transformer and each tap joint; According to the fluctuation difference between each time and the adjacent time in the power data sequence, obtain the power load fluctuation of the transformer and each tap joint at each time; according to the influence of the tap joint on the power load fluctuation between the transformer, obtain the tap load temperature influence factor at each time; according to the tap load temperature influence factor and the power load fluctuation, obtain the temperature local severe fluctuation in the transformer at each time; By comparing the power load fluctuations between the tapped contacts and the transformer, the heat contribution of each tapped contact is obtained; based on the heat contribution of different tapped contacts, the thermal disorder of the transformer's tapped contacts at each time point is obtained; based on the positional distribution of the tapped contacts and the differences in temperature data sequences, the local heat accumulation within the transformer at each time point is obtained; based on the local heat accumulation and the thermal disorder of the tapped contacts, the temperature non-uniformity within the transformer at each time point is obtained; based on the temperature non-uniformity and the severe local temperature fluctuations, the cooling concern at each time point is obtained. Predict future temperature changes of transformers based on cooling concerns and perform air-cooled fault detection.
[0005] Preferably, the specific method for obtaining the power load volatility of the transformer and each tap point at each moment based on the fluctuation difference between each moment and its neighboring moments in the power data sequence is as follows: Preset a neighborhood parameter In the power data sequence of the transformer, the first... The closest time before that moment The most recent moment and thereafter The time series range consisting of the nth moment is denoted as the nth moment. The temporal neighborhood range at each moment; The first The variance of the transformer power data at all times within the time-series neighborhood of time point n is compared with that of time point n. Within the temporal neighborhood of the i-th time step The absolute value of the difference between the variances of the transformer's power data at all times within the time-series neighborhood of time n is denoted as the nth time. The fluctuation difference value at time 1; the value of the fluctuation difference at time 2; The inversely proportional normalized value of the sum of fluctuation differences across all time intervals within the temporal neighborhood of time t is used as the t-th time interval. The power load fluctuation of the transformer at a given moment; Based on the method for obtaining the power load fluctuation of the transformer at each time point, the power load fluctuation of each tap contact at each time point is obtained.
[0006] Preferably, the specific method for obtaining the tap load temperature influence factor at each moment based on the influence of the tap contacts on the power load fluctuation between transformers is as follows: Using the DBSCAN clustering algorithm to cluster the first Clustering is performed on the power load fluctuations of all tap contacts on the transformer at a given time to obtain the first... The clustering results at a given time, wherein the clustering results include several clusters; The mean of the power load volatility of all taps in each cluster is recorded as the volatility mean of each cluster; the cluster with the largest volatility mean is obtained and recorded as the target cluster. The first The power load fluctuation of the transformer at time t and the first The absolute value of the difference between the mean power load volatility of all tapped nodes within the target cluster in the clustering results at time step i is denoted as the tap influence factor; the tap influence factor is then compared with the value at time step j. The inversely proportional normalized value of the product of the number of all taps within the target cluster in the clustering results at time step i is used as the i-th time step. The influence factor of tap load temperature at a given time.
[0007] Preferably, the specific method for obtaining the localized severe temperature fluctuations within the transformer at each moment based on the tap load temperature influence factor and power load fluctuation is as follows: The first The power load fluctuation of the transformer at time t and the first The product of the tap load temperature influence factors at time t is used as the product of the factors at time t. At any given moment, the temperature inside the transformer fluctuates drastically in a localized manner.
[0008] Preferably, the specific method for obtaining the heat generation contribution of each tap contact by comparing the power load fluctuation between the tap contact and the transformer is as follows: The first At the nth moment, the transformer's first The power load fluctuation of the tapped contact is related to the first The ratio between the power load fluctuation of the transformer at time t is used as the first... At the nth moment, the transformer's first The heat contribution of each tap connection.
[0009] Preferably, the specific method for obtaining the thermal disorder of the transformer taps at each moment based on the heat generation contribution of different taps is as follows: In the At any given moment, the tap contacts on the transformer can be arbitrarily combined in pairs to obtain several tap contact combinations; Obtain the absolute value of the difference in heat generation contribution between the two taps in each tap contact combination, and record it as the heat generation difference value of each tap contact combination. Each tap connection connects to several devices; Obtain the absolute value of the difference in the number of devices connected between two taps in each tap contact combination, and record it as the device number difference value for each tap contact combination. The product of the heat generation difference value of each tap connection combination and the equipment quantity difference value is recorded as the heat disturbance factor of each tap connection combination; the normalized value of the cumulative value of the heat disturbance factors of all tap connection combinations is used as the first... The thermal disorder of the transformer tap contacts at a given moment.
[0010] Preferably, the specific method for obtaining the local heat accumulation in the transformer at each moment based on the positional distribution of the tap contacts and the differences in temperature data sequences is as follows: Preset an adjacent parameter , obtain and the The tap contact is closest to the front Each tap point is considered as the first tap. Adjacent points of a tapped contact; For the At the nth moment, the transformer's first The nth tap contact, calculate the nth tap. The temperature data of the tap contact is related to the first tap contact. The absolute value of the difference between the temperature data of the nth adjacent junctions is denoted as the temperature difference value; the nth... The tap contact and the first The distance between the nth adjacent nodes is denoted as the positional difference value; the product of the temperature difference value and the positional difference value is denoted as the nth... The heat transfer influence factor of the nth adjacent node; The normalized value of the sum of the heat transfer influence factors of all adjacent points of the tapped contact is used as the value of the tapped contact. The degree of influence of heat transfer at each tap point; The first The ratio of the temperature data of the tap contact to the temperature data of the transformer is denoted as the ratio of the temperature data of the tap contact to the temperature data of the transformer. The first ratio of the tap contact; the first... The product of the first ratio of each tapped contact and the degree of heat transfer influence is denoted as the first ratio. The heat accumulation factor of the tapped contact; the normalized value of the mean of the heat accumulation factors of all tapped contacts is used as the first... Localized heat accumulation within the transformer at a given moment.
[0011] Preferably, the specific method for obtaining the temperature non-uniformity within the transformer at each moment based on local heat accumulation and tap contact heat disorder is as follows: The first The local heat accumulation inside the transformer at time point 1 is similar to that at time 2. The product of the thermal disorder of the transformer tap contacts at time n is used as the product of the thermal disorder of the transformer taps at time n. The temperature inside the transformer is chaotic and uneven at any given time.
[0012] Preferably, the specific method for obtaining the cooling concern at each moment based on the temperature non-uniformity and localized severe temperature fluctuations is as follows: The first The normalized value of the product of the temperature non-uniformity and the localized severe temperature fluctuations within the transformer at time t is used as the t-th time step. Cooling down attention at that moment.
[0013] The present invention also proposes a transformer air-cooled fault diagnosis system based on Internet of Things (IoT) technology, including a memory and a processor. The processor executes a computer program stored in the memory to implement the steps of the above-mentioned transformer air-cooled fault diagnosis method based on IoT technology.
[0014] The beneficial effects of the technical solution of this invention are as follows: This invention obtains the local drastic temperature fluctuations within the transformer at each moment based on the influence factor of tap load temperature and power load fluctuations; it obtains the thermal disorder of the transformer tap contacts at each moment based on the heat contribution of different tap contacts; it obtains the temperature non-uniformity within the transformer at each moment based on local heat accumulation and tap contact thermal disorder; it obtains the cooling concern at each moment based on the temperature non-uniformity and local drastic temperature fluctuations; it predicts future temperature changes of the transformer based on the cooling concern and performs air-cooling fault detection; by considering the cooling concern at the current moment and weighting the prediction of future temperature changes, the temperature trend of the transformer can be predicted more accurately, thereby reducing the temperature prediction error caused by lag and providing more reliable prediction results; by real-time monitoring and weighted prediction of temperature changes, it can not only improve the accuracy of transformer air-cooling fault detection, but also provide a guarantee for the safe and stable operation of the transformer. Attached Figure Description
[0015] 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.
[0016] Figure 1 This is a flowchart illustrating the steps of a transformer air-cooled fault diagnosis method based on Internet of Things (IoT) technology according to the present invention. Figure 2 This is a flowchart illustrating the characteristic relationships of a transformer air-cooled fault diagnosis method based on Internet of Things technology according to the present invention. Detailed Implementation
[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a transformer air-cooled fault diagnosis method and system based on Internet of Things technology proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.
[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.
[0019] The following description, in conjunction with the accompanying drawings, details the specific solution of the transformer air-cooled fault diagnosis method and system based on Internet of Things technology provided by this invention.
[0020] Please see Figure 1 The diagram illustrates a flowchart of a transformer air-cooled fault diagnosis method based on Internet of Things (IoT) technology, according to an embodiment of the present invention. The method includes the following steps: Step S001: Obtain the power data sequence and temperature data sequence of the transformer and each tap contact.
[0021] It should be noted that the output terminals of a transformer correspond to the tap contacts of the transformer; the high-voltage to low-voltage conversion of a transformer is achieved through different taps, that is, different taps represent different types of voltage conversion levels.
[0022] In a specific implementation of this invention, the method for obtaining the power data sequence and temperature data sequence of the transformer and each tap contact is as follows: Power sensors are installed at the input end of the transformer and at each tap point. Temperature sensors are installed inside the transformer and at each tap point. Power and temperature data of the transformer and each tap point are collected every second, for a total of 2 hours. The power and temperature data of the transformer at all times are used to construct two sequences, denoted as the transformer power data sequence and temperature data sequence. Similarly, the power and temperature data of each tap point at all times are used to construct two sequences, denoted as the power data sequence and temperature data sequence for each tap point.
[0023] Thus, the power data sequence and temperature data sequence of the transformer and each tap contact are obtained through the above method.
[0024] Step S002: Based on the fluctuation difference between each time step and the neighboring time step in the power data sequence, obtain the power load fluctuation of the transformer and each tap contact at each time step; based on the influence of the tap contact on the power load fluctuation between transformers, obtain the tap load temperature influence factor at each time step; based on the tap load temperature influence factor and the power load fluctuation, obtain the local severe temperature fluctuation within the transformer at each time step.
[0025] It should be noted that transformers typically have multiple taps, which are connected to different devices or loads. Each tap represents a different voltage value, and the load is connected to the transformer through these taps. The nature of each load affects changes in current and voltage, especially when the load changes. Changes in the load power at each tap affect the overall output power of the transformer. For example, when load devices start or stop, their changes in power demand cause load fluctuations, which in turn cause power load fluctuations in the transformer. When the power load fluctuation of the transformer is affected by the power load change of a single tap, it can severely lead to a sharp temperature change at that tap. In this case, the more precise the air-cooling effect, the greater the cooling attention required at that moment. Therefore, by analyzing the impact of taps on the power fluctuation of the transformer, we can obtain the localized severe temperature fluctuations within the transformer at each moment.
[0026] Preferably, in some implementations of the present invention, since the power data collected by the power sensor is in a time-series manner, when the power data at a certain moment is very close to the fluctuation of the power data at other moments in its time-series neighborhood, it indicates that the power data at that moment is in a stable fluctuation and there is no abnormal fluctuation. Therefore, the specific method for obtaining the power load fluctuation of the transformer and each tap contact at each moment by analyzing the difference in fluctuation between each moment and the neighboring moments in the power data sequence is as follows: Preset a neighborhood parameter In this embodiment, This example is used for illustration; no specific limitations are set in this embodiment. It depends on the specific implementation situation; In the power data sequence of the transformer, the first... The closest time before that moment The most recent moment and thereafter The time series range consisting of the nth moment is denoted as the nth moment. The temporal neighborhood range at each moment; The first The variance of the transformer power data at all times within the time-series neighborhood of time point n is compared with that of time point n. Within the temporal neighborhood of the i-th time step The absolute value of the difference between the variances of the transformer's power data at all times within the time-series neighborhood of time n is denoted as the nth time. The fluctuation difference value at time 1; the value of the fluctuation difference at time 2; The inversely proportional normalized value of the sum of fluctuation differences across all time intervals within the temporal neighborhood of time t is used as the t-th time interval. The power load fluctuation of the transformer at a given moment; The specific formula is as follows: In the formula, Indicates the first The power load fluctuation of the transformer at a given moment; Indicates the first The number of all times within the temporal neighborhood of a given time; Indicates the first The variance of the transformer power data at all times within the time-series neighborhood of a given time point; Indicates the first Within the temporal neighborhood of the i-th time step The variance of the transformer power data at all times within the time-series neighborhood of a given time point; Indicates taking the absolute value; This represents an exponential function with the natural constant as the base. The example uses... The model is used to represent the inverse proportional relationship and for normalization processing. As input to the model, implementers can choose between an inverse proportional function and a normalization function based on the actual situation.
[0027] Similarly, following the method described above, the power load fluctuation of each tap point at each time point is obtained.
[0028] Among them, for the first The power data of the transformer at the nth moment, when the nth moment The smaller the difference (i.e., the more similar) between the power data fluctuations at all times within the time-series neighborhood of a given moment and the power data fluctuations at other times within that time-series neighborhood, the more normal the power fluctuations of the transformer are, and the absence of abnormal fluctuations. Therefore, the power fluctuations at the given moment are considered normal. The smaller the power load fluctuation of the transformer at a given moment.
[0029] Preferably, in some implementations of the present invention, since power load fluctuations are accompanied by temperature changes, when the power load fluctuation of the transformer is affected by the power load change of a single tap connection, it will severely cause a sharp change in the temperature at that tap load connection. In this case, the air-cooling effect needs to be more precise, meaning that the cooling attention required at that moment is greater. Therefore, based on the influence of the tap connection on the power load fluctuations between transformers, the specific method for obtaining the tap load temperature influence factor at each moment is as follows: Using the DBSCAN clustering algorithm to cluster the first Clustering is performed on the power load fluctuations of all tap contacts on the transformer at a given time to obtain the first... The clustering results at a given time, wherein the clustering results include several clusters; The mean of the power load volatility of all taps in each cluster is recorded as the mean volatility of each cluster; the cluster with the largest mean volatility is recorded as the target cluster. The first The power load fluctuation of the transformer at time t and the first The absolute value of the difference between the mean power load volatility of all tapped nodes within the target cluster in the clustering results at time step i is denoted as the tap influence factor; the tap influence factor is then compared with the value at time step j. The inversely proportional normalized value of the product of the number of all taps within the target cluster in the clustering results at time step i is used as the i-th time step. Factors affecting tap load temperature at any given time; The specific formula is as follows: In the formula, Indicates the first Factors affecting tap load temperature at any given time; Indicates the first The number of all taps in the target cluster in the clustering results at each time point; Indicates the first The power load fluctuation of the transformer at any given time; Indicates the first The mean power load variability of all tapped points within the target cluster in the clustering results at a given time point; Indicates taking the absolute value; This represents an exponential function with the natural constant as the base. The example uses... The model is used to represent the inverse proportional relationship and for normalization processing. As input to the model, implementers can choose between an inverse proportional function and a normalization function based on the actual situation.
[0030] The DBSCAN clustering algorithm is an existing technology and will not be elaborated on in this embodiment. The target cluster indicates that the power load change of the taps in the cluster has a strong impact on the overall power fluctuation of the transformer. If the number of taps in the target cluster is smaller and the difference between the mean of its fluctuation and the power load fluctuation of the transformer is smaller, it means that the power load fluctuation of the transformer is affected by the power load change of a small number of taps, which will seriously cause the transformer to have a sharp temperature change at these taps.
[0031] Preferably, in some implementations of the present invention, when the tap load temperature influence factor is larger, the power load fluctuation of the transformer is greater, and it is easier to cause local drastic temperature fluctuations inside the transformer, increasing the possibility of transformer air-cooling failure. Therefore, the air-cooling effect needs to be more precise, meaning that greater attention needs to be paid to cooling at that moment. Thus, the specific method for obtaining the local drastic temperature fluctuations inside the transformer at each moment based on the tap load temperature influence factor and power load fluctuation is as follows: The first The power load fluctuation of the transformer at time t and the first The product of the tap load temperature influence factors at time t is used as the product of the factors at time t. The localized and drastic temperature fluctuations inside the transformer at a given moment; The specific formula is as follows: In the formula, Indicates the first The localized and drastic temperature fluctuations inside the transformer at a given moment; Indicates the first The power load fluctuation of the transformer at any given time; Indicates the first The influence factor of tap load temperature at a given time.
[0032] Thus, the localized and drastic temperature fluctuations within the transformer at each moment were obtained using the above method.
[0033] Step S003: By comparing the power load fluctuations between the tapped contacts and the transformer, obtain the heat generation contribution of each tapped contact; based on the heat generation contribution of different tapped contacts, obtain the thermal disorder of the transformer's tapped contacts at each time point; based on the positional distribution of the tapped contacts and the differences in temperature data sequences between adjacent tapped contacts, obtain the local heat accumulation within the transformer at each time point; based on the local heat accumulation and the thermal disorder of the tapped contacts, obtain the temperature non-uniformity within the transformer at each time point; based on the temperature non-uniformity and the severe local temperature fluctuations, obtain the cooling concern at each time point.
[0034] It should be noted that due to the differences in each tap contact and the different voltage types required by different devices (i.e., different high-voltage to low-voltage conversions) and the differences in device operation, the heat generated at each tap contact will vary. The more different the heat generated at different tap contacts of the transformer, and the less ideal the temperature transfer between adjacent tap contacts, the more chaotic the internal temperature of the transformer will be. In this case, the air-cooling effect needs to be more precise, and the greater the attention required for cooling at that moment.
[0035] Preferably, in some implementations of the present invention, power load fluctuation represents the degree of change in the power load of the tapped contact over time; the greater the power load fluctuation of the tapped contact, the more drastic the change in its power load, and the greater the heating effect will be; therefore, by comparing the power load fluctuation of the tapped contact with the power load fluctuation of the transformer, the contribution of the tapped contact to the overall heat generation of the transformer can be obtained; that is, if the fluctuation of a certain tapped contact is closer to the fluctuation of the transformer, then the heat generation contribution of the tapped contact is higher, and vice versa; the specific method for obtaining the heat generation contribution of each tapped contact by comparing the power load fluctuation between the tapped contact and the transformer is as follows: The first At the nth moment, the transformer's first The power load fluctuation of the tapped contact is related to the first The ratio between the power load fluctuation of the transformer at time t is used as the first... At the nth moment, the transformer's first The heat contribution of each tap connection; The specific formula is as follows: In the formula, Indicates the first At the nth moment, the transformer's first The heat contribution of each tap connection; Indicates the first The power load fluctuation of the transformer at any given time; Indicates the first At the nth moment, the transformer's first Power load fluctuation of each tap contact.
[0036] It should be noted that at the same time, the greater the difference in heat contribution from different taps, the more the load fluctuation of some equipment is much higher than that of other equipment. This difference will lead to uneven heat distribution inside the transformer, and some areas may overheat. The high-voltage to low-voltage conversion of the transformer is achieved through different taps, and different taps represent different voltage conversion levels. For example, some users can obtain 220V through one winding or tap, while other users can obtain 380V through another winding or tap; that is, one tap corresponds to one type of voltage conversion.
[0037] Preferably, in some implementations of the present invention, when the number of high-voltage to low-voltage conversion types of the transformer increases, that is, when the difference between the high-voltage to low-voltage conversion types and the number of devices is large, the heat distribution inside the transformer becomes more uneven, and it is easier to cause more chaotic temperature inside the transformer; therefore, the specific method for obtaining the degree of heat chaos of the transformer's tap contacts at each moment based on the heat contribution of different tap contacts is as follows: In the At any given moment, the tap contacts on the transformer can be arbitrarily combined in pairs to obtain several tap contact combinations; Obtain the absolute value of the difference in heat generation contribution between the two taps in each tap contact combination, and record it as the heat generation difference value of each tap contact combination. Since the number of devices connected to each tap may vary, the greater the difference in the number, the greater the temperature difference at the taps in the tap combination. Obtain the absolute value of the difference in the number of devices connected between two taps in each tap contact combination, and record it as the device number difference value for each tap contact combination. The product of the heat generation difference value of each tap connection combination and the equipment quantity difference value is recorded as the heat disturbance factor of each tap connection combination; the normalized value of the cumulative value of the heat disturbance factors of all tap connection combinations is used as the first... The thermal disorder of the transformer tap contacts at a given moment; The specific formula is as follows: In the formula, Indicates the first The thermal disorder of the transformer tap contacts at a given moment; Indicates the first The number of all tap contact combinations of the transformer at any given moment; Indicates the first At the nth moment, the transformer's first The difference in heat generation between individual tap contact combinations; Indicates the first At the nth moment, the transformer's first The difference in the number of devices for each tap contact combination; Linear normalization function.
[0038] It should be noted that due to the heat transfer effect, if the temperatures between adjacent taps of a transformer are all high, the heat generated at the taps will be difficult to dissipate, causing heat to accumulate in localized areas within the transformer; that is, if the temperature of the taps is high... The lower the temperature of the first tap contact and the greater the temperature difference between it and its adjacent contacts, the more effectively the generated heat can be transferred, and the less likely it is that heat will accumulate in a localized area; if the temperature of the first tap contact is lower, the temperature of the second tap contact is lower, the temperature of the third tap contact is lower, the temperature of the fourth tap contact is lower, the temperature of the fifth tap contact is lower, the temperature of the sixth tap contact is lower, the temperature of the seventh tap contact is lower, the temperature of the eighth tap contact is lower, the temperature of the ninth ... The higher the temperature of a tap contact and the smaller the temperature difference between it and its adjacent contacts, the less heat can be transferred, and the easier it is for heat to accumulate in local areas. Therefore, the more precise the air-cooling effect needs to be, and the greater the attention required for cooling at this time.
[0039] Preferably, in some implementations of the present invention, the specific method for obtaining the local heat accumulation in the transformer at each moment based on the positional distribution of the tap contacts and adjacent tap contacts and the differences in temperature data sequences is as follows: Preset an adjacent parameter In this embodiment, This example is used for illustration; no specific limitations are set in this embodiment. It depends on the specific implementation situation; For the transformer's first The tap contact, obtain the first... The distance between each tap contact and all other tap contacts; select the nearest one. Each tap point is considered as the first tap. Adjacent points of a tapped contact; The first At the nth moment, the transformer's first The temperature data of the tap contact is related to the first tap contact. At the nth moment, the transformer's first The first tap point The absolute value of the difference between the temperature data of the nth adjacent junctions is denoted as the nth... The temperature difference value of the first adjacent connection point; the transformer's first The tap contact and the first The distance between the nth adjacent nodes is denoted as the nth node. The positional difference value of the nth adjacent node; The product of the temperature difference and the positional difference of each of the nth adjacent junctions is denoted as the i-th The heat transfer influence factor of the nth adjacent node; At the nth moment, the transformer's first The normalized value of the sum of the heat transfer influence factors of all adjacent points of the tapped contact is used as the value of the tapped contact. At the nth moment, the transformer's first The degree of influence of heat transfer at each tap point; The specific formula is as follows: In the formula, Indicates the first At the nth moment, the transformer's first The degree of influence of heat transfer at each tap point; Indicates the first The number of all adjacent points of a tapped contact; Indicates the first At the nth moment, the transformer's first Temperature data of each tap contact; Indicates the first At the nth moment, the transformer's first The first tap point Temperature data of adjacent nodes; The transformer's first The tap contact and its first The distance between adjacent nodes; This represents an exponential function with the natural constant as the base. The example uses... The model is used to represent the inverse proportional relationship and for normalization processing. As input to the model, implementers can choose between an inverse proportional function and a normalization function based on the actual situation.
[0040] Among them, in the first At the nth moment, the transformer's first The higher the temperature of the tap contact and the more severe the impact of heat transfer, the more likely it is to cause localized heat accumulation within the transformer; based on the degree of heat transfer impact of the tap contact, the [missing information - likely a specific measurement or determination] can be obtained. The specific method for addressing localized heat accumulation within the transformer at a given time is as follows: The first At the nth moment, the transformer's first The temperature data of the tap contact is related to the first tap contact. The ratio of the transformer temperature data at time n is denoted as the nth time. The first ratio of the tap contact; the first... The product of the first ratio of each tapped contact and the degree of heat transfer influence is denoted as the first ratio. The heat accumulation factor of the tap contact; will the first tap contact's heat accumulation factor; The normalized value of the mean of the heat accumulation factor of all taps of the transformer at time n is used as the first time step. Localized heat accumulation within the transformer at a given moment; The specific formula is as follows: In the formula, Indicates the first Localized heat accumulation within the transformer at a given moment; This indicates the total number of all taps on the transformer. Indicates the first At the nth moment, the transformer's first Temperature data of each tap contact; Indicates the first Temperature data of the transformer at any given moment; Indicates the first At the nth moment, the transformer's first The degree of influence of heat transfer at each tap point; Linear normalization function.
[0041] Preferably, in some implementations of the embodiments of the present invention, if the first The higher the thermal disorder at the transformer's tap contacts and the greater the localized heat accumulation within the transformer at a given moment, the more uneven the overall temperature distribution within the transformer becomes. This necessitates a more precise air-cooling process for the transformer. Therefore, the specific method for determining the temperature unevenness within the transformer at each moment, based on localized heat accumulation and the thermal disorder at the tap contacts, is as follows: The first The local heat accumulation inside the transformer at time point 1 is similar to that at time 2. The product of the thermal disorder of the transformer tap contacts at time n is used as the product of the thermal disorder of the transformer taps at time n. The temperature inside the transformer is irregular and uneven at any given moment; The specific formula is as follows: In the formula, Indicates the first The temperature inside the transformer is irregular and uneven at any given moment; Indicates the first Localized heat accumulation within the transformer at a given moment; Indicates the first The thermal disorder of the transformer tap contacts at a given moment.
[0042] Preferably, in some implementations of the embodiments of the present invention, if the first The stronger the localized and drastic temperature fluctuations within the transformer at a given moment, and the greater the temperature heterogeneity, the more significant the potential safety hazard presents the transformer. In this case, the transformer's air-cooling system needs to be implemented more promptly and precisely. Therefore, the first... The greater the cooling concern at any given moment, the better. Therefore, based on the chaotic and uneven temperature distribution and the drastic local temperature fluctuations within the transformer, the specific method for obtaining the cooling concern at each moment is as follows: The first The normalized value of the product of the temperature non-uniformity and the localized severe temperature fluctuations within the transformer at time t is used as the t-th time step. Cooling down attention at a particular moment; The specific formula is as follows: In the formula, Indicates the first Cooling down attention at that moment; Indicates the first The temperature inside the transformer is irregular and uneven at any given moment; Indicates the first Localized and drastic temperature fluctuations inside the transformer at any given moment; Linear normalization function.
[0043] Thus, the cooling attention level at each moment is obtained through the above method.
[0044] Step S004: Predict future temperature changes of the transformer based on cooling concern and perform air-cooling fault detection.
[0045] It should be noted that a transformer air-cooling system is a device used to cool transformers. Its main function is to remove the heat generated by the transformer during operation by forcing airflow, thereby ensuring that the transformer operates within a safe and efficient temperature range. When the transformer is running, a temperature sensor continuously monitors its temperature. Once the temperature reaches a preset start-up threshold, the control unit activates the fan, which begins blowing air through the heat sink to remove heat from the transformer. When the temperature drops to a preset stop-loss threshold, the fan automatically stops to conserve energy. This automatic adjustment mechanism ensures that the transformer remains within a safe temperature range under various operating conditions.
[0046] Preferably, in some implementations of the embodiments of the present invention, the specific method for predicting future temperature changes of the transformer based on cooling concern and performing air-cooling fault detection is as follows: Preset a threshold parameter and a time parameter In this embodiment, , This example is used for illustration; no specific limitations are set in this embodiment. , It depends on the specific implementation situation; The current cooling concern level is used as a weighted input into a weighted prediction algorithm to predict the transformer's temperature data and obtain future... A sequence of transformer temperature data for minutes; if the mean of all temperature data in the transformer temperature data sequence is greater than or equal to a threshold parameter. If this is the case, it indicates a fault in the transformer's air cooling system, requiring immediate inspection and repair by staff.
[0047] The weighted prediction algorithm is existing technology and will not be described in detail here.
[0048] Please see Figure 2 It shows a flowchart of the characteristic relationship of a transformer air-cooled fault diagnosis method based on Internet of Things technology; Through the above steps, a method for diagnosing transformer air-cooled faults based on Internet of Things (IoT) technology is completed.
[0049] Another embodiment of the present invention provides a transformer air-cooled fault diagnosis system based on Internet of Things technology. The system includes a memory and a processor. When the processor executes the computer program stored in the memory, it performs the above method steps S001 to S004.
[0050] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A method for diagnosing transformer air-cooled faults based on Internet of Things (IoT) technology, characterized in that, The method includes the following steps: Acquire the power data sequence and temperature data sequence of the transformer and each tap contact; Based on the fluctuation differences between each time step and its neighboring time steps in the power data sequence, the power load fluctuation of the transformer and each tap contact at each time step is obtained; based on the influence of the tap contact on the power load fluctuation between transformers, the tap load temperature influence factor at each time step is obtained; based on the tap load temperature influence factor and the power load fluctuation, the local severe temperature fluctuation within the transformer at each time step is obtained. By comparing the power load fluctuations between the tapped contacts and the transformer, the heat contribution of each tapped contact is obtained; based on the heat contribution of different tapped contacts, the thermal disorder of the transformer's tapped contacts at each time point is obtained; based on the positional distribution of the tapped contacts and the differences in temperature data sequences, the local heat accumulation within the transformer at each time point is obtained; based on the local heat accumulation and the thermal disorder of the tapped contacts, the temperature non-uniformity within the transformer at each time point is obtained; based on the temperature non-uniformity and the severe local temperature fluctuations, the cooling concern at each time point is obtained. Predict future temperature changes of transformers based on cooling concerns and perform air-cooled fault detection.
2. The transformer air-cooled fault diagnosis method based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the power load fluctuation of the transformer and each tap point at each moment based on the fluctuation difference between each moment and its neighboring moments in the power data sequence is as follows: Preset a neighborhood parameter In the power data sequence of the transformer, the first... The closest time before that moment The most recent moment and thereafter The time series range consisting of the nth moment is denoted as the nth moment. The temporal neighborhood range at each moment; The first The variance of the transformer power data at all times within the time-series neighborhood of time point n is compared with that of time point n. Within the temporal neighborhood of time step n, the first... The absolute value of the difference between the variances of the transformer's power data at all times within the time-series neighborhood of time n is denoted as the nth time. The fluctuation difference value at time t; will the t moment The inversely proportional normalized value of the sum of fluctuation differences across all time intervals within the temporal neighborhood of time t is used as the t-th time interval. The power load fluctuation of the transformer at a given moment; Based on the method for obtaining the power load fluctuation of the transformer at each time point, the power load fluctuation of each tap contact at each time point is obtained.
3. The method for diagnosing transformer air-cooled faults based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the tap load temperature influence factor at each moment based on the impact of tap contacts on the power load fluctuation between transformers is as follows: Using the DBSCAN clustering algorithm to cluster the first Clustering is performed on the power load fluctuations of all tap contacts on the transformer at a given time to obtain the first... The clustering results at a given time, wherein the clustering results include several clusters; The mean of the power load volatility of all taps in each cluster is recorded as the volatility mean of each cluster; the cluster with the largest volatility mean is obtained and recorded as the target cluster. The first The power load fluctuation of the transformer at time t and the first The absolute value of the difference between the mean power load volatility of all tapped nodes within the target cluster in the clustering results at time step i is denoted as the tap influence factor; the tap influence factor is then compared with the value at time step j. The inversely proportional normalized value of the product of the number of all taps within the target cluster in the clustering results at time step i is used as the i-th time step. The influence factor of tap load temperature at a given time.
4. The method for diagnosing transformer air-cooled faults based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the localized severe temperature fluctuations within the transformer at each moment based on the tap load temperature influence factor and power load volatility is as follows: The first The power load fluctuation of the transformer at time t and the first The product of the tap load temperature influence factors at time t is used as the product of the factors at time t. At any given moment, the temperature inside the transformer fluctuates drastically in a localized manner.
5. The transformer air-cooled fault diagnosis method based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the heat contribution of each tap point by comparing the power load fluctuation between the tapped contact and the transformer is as follows: The first At the nth moment, the transformer's first The power load fluctuation of the tapped contact is related to the first The ratio between the power load fluctuation of the transformer at time t is used as the first... At the nth moment, the transformer's first The heat contribution of each tap connection.
6. The transformer air-cooled fault diagnosis method based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the thermal disorder of the transformer taps at each moment based on the heat generation contribution of different taps is as follows: In the At any given moment, the tap contacts on the transformer can be arbitrarily combined in pairs to obtain several tap contact combinations; Obtain the absolute value of the difference in heat generation contribution between the two taps in each tap contact combination, and record it as the heat generation difference value of each tap contact combination. Each tap connection connects to several devices; Obtain the absolute value of the difference in the number of devices connected between two taps in each tap contact combination, and record it as the device number difference value for each tap contact combination. The product of the heat generation difference value of each tap connection combination and the equipment quantity difference value is recorded as the heat disorder factor of each tap connection combination; the normalized value of the cumulative value of the heat disorder factors of all tap connection combinations is used as the first... The thermal disorder of the transformer tap contacts at a given moment.
7. The method for diagnosing transformer air-cooled faults based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the local heat accumulation in the transformer at each moment based on the positional distribution of the tap contacts and adjacent tap contacts and the differences in temperature data sequences is as follows: Preset an adjacent parameter , obtain and the The tap contact is closest to the front Each tap point is considered as the first tap. Adjacent points of a tapped contact; For the At the nth moment, the transformer's first For the nth tap contact, calculate the nth tap. The temperature data of the tap contact is related to the first tap contact. The absolute value of the difference between the temperature data of the nth adjacent junctions is denoted as the temperature difference value; the nth The tap contact and the first The distance between the nth adjacent nodes is denoted as the positional difference value; the product of the temperature difference value and the positional difference value is denoted as the nth... The heat transfer influence factor of the nth adjacent node; The normalized value of the sum of the heat transfer influence factors of all adjacent points of the tapped contact is used as the value of the tapped contact. The degree of influence of heat transfer at each tap point; The first The ratio of the temperature data of the tap contact to the temperature data of the transformer is denoted as the ratio of the temperature data of the tap contact to the temperature data of the transformer. The first ratio of the tap contact; the first... The product of the first ratio of each tapped contact and the degree of heat transfer influence is denoted as the first ratio. The heat accumulation factor of the tapped contact; the normalized value of the mean of the heat accumulation factors of all tapped contacts is used as the first... Localized heat accumulation within the transformer at a given moment.
8. The method for diagnosing transformer air-cooled faults based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the temperature non-uniformity inside the transformer at each moment based on local heat concentration and tap contact heat disorder is as follows: The first The local heat accumulation inside the transformer at time point 1 is similar to that at time 2. The product of the thermal disorder of the transformer tap contacts at time n is used as the product of the thermal disorder of the transformer taps at time n. The temperature inside the transformer is chaotic and uneven at any given time.
9. The method for diagnosing transformer air-cooled faults based on Internet of Things technology according to claim 1, characterized in that, The specific method for obtaining the cooling attention level at each moment based on the chaotic and uneven temperature and the drastic local temperature fluctuations is as follows: The first The normalized value of the product of the temperature non-uniformity and the localized severe temperature fluctuations within the transformer at time t is used as the t-th time step. Cooling down attention at that moment.
10. A transformer air-cooled fault diagnosis system based on Internet of Things (IoT) technology, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of the transformer air-cooled fault diagnosis method based on Internet of Things technology as described in any one of claims 1-9.