Pointer type oil level gauge for monitoring oil level of oil conservator of transformer

Abstract

The invention relates to the technical field of oil level gauges, and discloses a pointer type oil level gauge for monitoring the oil level of a transformer oil conservator, which comprises an oil level gauge assembly and a control module, the control module is electrically connected with the oil level gauge assembly, and the control module is used for controlling the oil level gauge assembly; the oil level gauge assembly comprises a dial plate, a fixing flange, two limiting rods, an angle sensor, a limiting block, a rotating rod and a floater, the dial plate is fixedly connected to the upper end of the fixing flange, the angle sensor is electrically connected into the dial plate, the limiting rods are fixedly connected to the lower end of the fixing flange, and the limiting block is fixedly connected to the end, away from the fixing flange, of each limiting rod. A rotating rod is arranged between the two limiting rods, the two ends of the rotating rod are rotationally connected with the dial plate and the limiting block respectively, threads are axially arranged on the surface of the rotating rod, the rotating rod is sleeved with the floater, and the two sides of the floater are slidably connected with the two limiting rods. Deformation or fatigue damage caused by long-term stress is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of oil level gauges, and more particularly, to a pointer type oil level gauge for monitoring the oil level of a transformer conservator tank. Background Art

[0002] A transformer is a core device in the power system. Its internal insulating oil not only undertakes the functions of insulation and heat dissipation, but also can reflect the operating state of the device through volume changes. The conservator tank (also known as the oil pillow), as an important part of the transformer, compensates for the expansion or contraction of the insulating oil caused by temperature changes to maintain the pressure stability in the oil tank. Therefore, real-time monitoring of the oil in the oil pillow, especially the oil level monitoring, is of great significance for preventing transformer failures and ensuring the safe operation of the power system. If the oil level is too low, air or moisture may invade, resulting in a decrease in insulation performance, partial discharge or even short circuit. If the oil level is too high, leakage or explosion may occur.

[0003] However, after long-term use of the existing oil pillow oil level gauges, various problems such as the floating ball getting stuck and the telescopic rod breaking may occur, which may lead to the generation of the "false oil level" phenomenon. At this time, the actual oil level may have dropped below the safety threshold, but the oil level alarm cannot be triggered, which may lead to insulation exposure and flashover accidents, or cause local overheating inside the transformer, accelerating the deterioration of the insulating oil. Moreover, the false oil level will cover up the true trend of oil volume changes, causing maintenance personnel to miss the best opportunity for refueling.

[0004] Therefore, it is necessary to design a pointer type oil level gauge for monitoring the oil level of a transformer conservator tank to solve the problems existing in the current technology. Summary of the Invention

[0005] In view of this, the present invention proposes a pointer type oil level gauge for monitoring the oil level of a transformer conservator tank, aiming to solve the problem of low detection accuracy of the current oil level gauge.

[0006] The present invention proposes a pointer type oil level gauge for monitoring the oil level of a transformer conservator tank, comprising:

[0007] An oil level gauge assembly and a control module, the control module is electrically connected to the oil level gauge assembly, and the control module is used to control the oil level gauge assembly; the oil level gauge assembly includes:

[0008] A dial, a fixed flange, a limit rod, an angle sensor, a limit block, a rotating rod, and a float. The upper end of the fixed flange is fixedly connected to the dial. The angle sensor is electrically connected inside the dial. The lower end of the fixed flange is fixedly connected to the limit rod. One end of the limit rod away from the fixed flange is fixedly connected to the limit block. There are two limit rods, and a rotating rod is arranged between the two limit rods. The two ends of the rotating rod are respectively rotatably connected to the dial and the limit block. A thread is axially arranged on the surface of the rotating rod. The float is sleeved on the rotating rod, and the two sides of the float are slidably connected to the two limit rods.

[0009] Further, the control module includes:

[0010] An acquisition unit configured to acquire sensor data, store the sensor data at a fixed time interval, and perform denoising. The acquisition unit is further configured to align the timestamps of the sensor data and use KNN interpolation to process missing values;

[0011] A judgment unit configured to extract the sensor data and input the sensor data into a pre-trained state model for comparison. The judgment unit is further configured to determine the current operating state of the oil level gauge assembly based on the comparison result of the state model;

[0012] An early warning unit configured to, when the current operating state of the oil level gauge assembly is abnormal, determine the abnormal area based on the abnormal state and send corresponding early warning information based on each abnormal area.

[0013] Further, when inputting the sensor data into a pre-trained state model for comparison, it includes:

[0014] Obtain two time series of the sensor data. One time series includes the time series waveform of the oil level value and the time series waveform of the oil temperature data. The other time series includes the effective value of the load current, the external environmental temperature and humidity, and the vibration intensity of the oil conservator;

[0015] In the underlying feature learning stage, splice the time series waveforms of the two time series into an oil level - oil temperature joint feature and an auxiliary feature, and form an enhanced time series feature based on the oil level - oil temperature joint feature and the auxiliary feature;

[0016] In the high-level feature learning stage, use a dense network with a two-layer structure to extract high-order non-linear features layer by layer based on the enhanced time series feature to form the state model;

[0017] Input the sensor data into the state model, perform waveform reconstruction and obtain the oil level prediction value;

[0018] Determine the reconstruction error value and prediction deviation value of the state model based on the predicted oil level value.

[0019] Further, when determining the reconstruction error value and prediction deviation value of the state model based on the predicted oil level value, it includes:

[0020] Determine the reconstruction error value based on the time series waveform of the oil level value, and when the external environmental temperature fluctuates, perform temperature coefficient compensation on the reconstruction error value;

[0021] Determine the reconstruction error threshold based on the historical normal oil level value;

[0022] Obtain the actual data of the oil level value, and determine the prediction deviation value based on the predicted oil level value;

[0023] Arrange the prediction deviation values in ascending order, and take the 95th percentile as the prediction deviation threshold.

[0024] Further, when determining the current operating state of the oil level gauge assembly based on the comparison result of the state model, it includes:

[0025] Determine the anomaly score based on the prediction deviation threshold and the reconstruction error threshold:

[0026] ;

[0027] Wherein, is the anomaly score, and are the weight coefficients, and + = 1, is the reconstruction error value, is the threshold of the reconstruction error, is the prediction deviation value, is the threshold of the prediction deviation value.

[0028] Further, when the current operating state of the oil level gauge assembly is abnormal, it includes:

[0029] When the anomaly score is less than or equal to 1, determine that the oil level gauge assembly is normal;

[0030] When the anomaly score is greater than 1, determine that the oil level gauge assembly is abnormal and trigger fault feature analysis;

[0031] The fault feature analysis is divided into false oil level, oil leakage and sensor failure;

[0032] When is greater than 0.7, the reconstruction error value is greater than twice the reconstruction error threshold, and the prediction deviation value is less than the prediction deviation threshold, determine it as a false oil level fault;

[0033] When is greater than 0.7, the reconstruction error value is less than twice the reconstruction error threshold, and the prediction deviation value is greater than twice the prediction deviation threshold, it is determined as a sensor failure;

[0034] When is greater than 0.7, the reconstruction error value is less than twice the reconstruction error threshold, and the prediction deviation value is greater than twice the prediction deviation threshold, it is determined as an oil leakage failure.

[0035] Furthermore, when the current operating state of the oil level gauge assembly is abnormal, it further includes:

[0036] When the fault feature analysis determines it as a false oil level fault, but the oil level value suddenly changes, it is preferentially determined as a sensor failure;

[0037] When the fault feature analysis determines it as a false oil level fault, but the effective value of the load current suddenly increases and the vibration intensity of the conservator changes synchronously, it is preferentially determined as a vibration interference.

[0038] Furthermore, when determining the abnormal area based on the abnormal state and sending corresponding warning information based on each abnormal area, it includes:

[0039] When the warning unit determines it as a false oil level, a first-level warning information is generated;

[0040] When the warning unit determines it as an oil leakage, a second-level warning information is generated;

[0041] When the warning unit determines it as a sensor failure, a third-level warning information is generated.

[0042] Furthermore, when the warning unit determines that the current operating state is abnormal, the abnormal information is stored in the historical database and the fault type is marked, and the state model is trained at a fixed time interval and the threshold is updated.

[0043] Furthermore, it further includes: a remote monitoring unit, the remote monitoring unit is electrically connected to the warning unit, and the remote monitoring unit is configured to monitor the operating state of the warning unit and send the operating state monitoring information to the Internet and the cloud data platform.

[0044] Compared with the prior art, the beneficial effects of the present invention are as follows: Through the rigid connection between the fixed flange and the limiting rod, the mechanical strength of the overall structure is enhanced, which can disperse the impact of external vibration or pressure on key components, reduce deformation or fatigue damage caused by long-term stress. The sliding setting of the float along the limiting rod replaces the free-floating mode of the traditional floating ball. By physically guiding and restricting the movement trajectory of the float, the jamming problem caused by mechanical looseness or deviation is avoided. Moreover, the double-sided support of the limiting rod further improves the stability of the float movement, ensuring its linear response even when the oil density changes or the temperature fluctuates, thus improving the reliability of long-term operation. In addition, the cooperation between the rotating rod and the limiting rod improves the force transmission path, reducing the wear risk caused by local stress concentration. The direct linkage between the angle sensor and the rotating rod eliminates the clearance error in traditional indirect transmissions (such as gears and chains), and can capture the minute changes in the oil level in real time. The linear movement of the threaded rotating rod converts the vertical displacement of the float into a rotation angle signal, avoiding the non-linear error caused by gear meshing or lever swing. This not only improves the sensitivity of oil level detection but also shortens the response time of signal transmission, enabling the control module to quickly reflect the dynamic changes in the oil level. At the same time, the axial threads on the surface of the rotating rod increase the contact area between the float and the rod body, reducing the movement resistance caused by oil viscosity or impurity accumulation, and further improving the measurement consistency. The physical constraint of the limiting rod on the float movement suppresses the misoperation caused by violent oil fluctuations or mechanical vibrations, filtering out high-frequency interference and ensuring that the float only generates displacement when the oil level truly changes. The electrical connection between the control module and the angle sensor realizes the real-time acquisition and digital conversion of oil level data, providing a basis for remote monitoring and automated analysis. By integrating the control module, the trend characteristics of oil level changes can be identified, potential faults (such as leakage and jamming) can be warned in advance, and the dependence on manual inspection can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] By reading the following detailed description of the preferred embodiments, various other advantages and benefits will become clear to those of ordinary skill in the art. The drawings are only for the purpose of showing the preferred embodiments and are not considered to be a limitation of the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components. In the drawings:

[0046] Figure 1 FIG. is a schematic structural diagram of the pointer type oil level gauge for monitoring the oil level of the transformer conservator provided by the embodiment of the present invention.

[0047] Figure 2 FIG. is a functional block diagram of the pointer type oil level gauge for monitoring the oil level of the transformer conservator provided by the embodiment of the present invention.

[0048] Among them, 1. Oil level gauge assembly; 101. Dial; 102. Fixed flange; 103. Limit rod; 104. Angle sensor; 105. Rotating rod; 106. Float; 107. Limit block; 108. Thread. Detailed implementation mode

[0049] Hereinafter, exemplary embodiments of the present disclosure will be described in more detail with reference to the accompanying drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be completely conveyed to those skilled in the art. It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. Hereinafter, the present invention will be described in detail with reference to the drawings and in combination with the embodiments.

[0050] In some embodiments of the present application, refer to Figure 1-2 As shown, a pointer-type oil level gauge for monitoring the oil level of a transformer conservator includes:

[0051] An oil level gauge assembly 1 and a control module, the control module is electrically connected to the oil level gauge assembly, and the control module is used to control the oil level gauge assembly. The oil level gauge assembly includes:

[0052] A dial 101, a fixed flange 102, a limit rod 103, an angle sensor 104, a limit block 107, a rotating rod 105 and a float 106. The upper end of the fixed flange 102 is fixedly connected to the dial 101, the angle sensor 104 is electrically connected inside the dial 101, the lower end of the fixed flange 102 is fixedly connected to the limit rod 103, a limit block 107 is fixedly connected to the end of the limit rod 103 away from the fixed flange 102, there are two limit rods 103, a rotating rod 105 is arranged between the two limit rods 103, and the two ends of the rotating rod 105 are respectively rotatably connected to the dial 101 and the limit block 107. A thread 108 is axially arranged on the surface of the rotating rod 105, the float 106 is sleeved on the rotating rod 105, and the two sides of the float 106 are slidably connected to the two limit rods 103.

[0053] Specifically, the dial 101 is arranged above the fixed flange 102. There are two limit rods 103 arranged below the fixed flange 102. A rotating rod 105 is arranged between the two limit rods 103, and the rotating rod 105 is rotatably connected to the dial 101. The dial 101 is electrically connected to an angle sensor 104. When the oil level gauge assembly 1 operates, the float 106 is located on the upper surface of the cooling oil. Through the floating of the float 106, the rotating rod 105 is driven to rotate. Furthermore, the dial 101 can display the oil level data at the current moment. At the same time, the angle sensor 104 can obtain the oil level data more accurately, providing a data basis for the control module. The control module predicts the future oil level change trend in advance by obtaining data such as the oil temperature and oil level of the oil level gauge. When it is predicted that the change exceeds the threshold, an early warning message can be generated, facilitating the staff to determine the abnormal area or the cause of the abnormal event, and coping with the abnormal event more conveniently.

[0054] It can be understood that through the rigid connection setting of the fixed flange 102 and the limiting rod 103, the stability of the structure of the oil level gauge assembly 1 is enhanced, the stress generated by external mechanical vibration or oil fluid fluctuation can be dispersed, and the deformation or wear of components caused by long-term uneven stress is reduced. The guiding setting of the float 106 sliding along the limiting rod 103 replaces the traditional free-floating structure, avoiding the jamming problem of the float 106 caused by mechanical looseness or deviation, thereby improving the reliability of long-term operation. The cooperative cooperation between the rotating rod 105 and the limiting rod 103 further optimizes the movement trajectory of the float 106, ensuring that it can still maintain a linear response when the oil fluid density changes or the temperature fluctuates, and reducing the measurement deviation caused by environmental factors. The direct linkage between the angle sensor 104 and the rotating rod 105 eliminates the clearance error in traditional indirect transmissions (such as gears, chains), and can capture the minute changes in the oil level in real time. The linear motion mechanism of the threaded 108 rotating rod 105 converts the vertical displacement of the float 106 into an accurate rotation angle signal, avoiding the non-linear error caused by gear meshing or lever swing. It not only improves the sensitivity of oil level detection, but also shortens the response time of signal transmission, enabling the control module to quickly reflect the dynamic changes in the oil level. In addition, the contact between the float 106 and the axial thread 108 of the rotating rod 105 also increases the moving contact area, reducing the moving resistance caused by oil fluid viscosity or impurity accumulation, and further improving the measurement consistency. The physical constraint of the limiting rod 103 on the movement of the float 106 suppresses the misoperation caused by violent oil fluid fluctuation or mechanical vibration. In scenarios with frequent external vibrations, this rigid guiding structure can filter out high-frequency interference, ensuring that the float 106 only generates displacement when the oil level truly changes. The dial 101 is a sealed dial, which can isolate external dust, oil stains and moisture, reducing the pollution risk of the angle sensor 104 and extending the service life of key components. The control module can dynamically analyze the trend characteristics of the oil level change by collecting the oil level data of the angle sensor 104 in real time and combining parameters such as oil temperature. Based on the prediction model, potential abnormalities (such as leakage, jamming) can be identified in advance, and warning information can be generated, shortening the fault response time, reducing the dependence on manual inspection, and at the same time reducing the work complexity of the operation and maintenance personnel by positioning the cause of the abnormality (such as distinguishing mechanical failure from sensor failure).

[0055] In some embodiments of the present application, the control module includes:

[0056] The control module includes:

[0057] An acquisition unit, configured to acquire sensor data, store the sensor data at fixed time intervals, and perform denoising. The acquisition unit is also configured to align the timestamps of the sensor data and use KNN interpolation to process missing values.

[0058] A judgment unit, configured to extract sensor data, input the sensor data into a pre-trained state model for comparison, and the judgment unit is further configured to determine the current operating state of the oil level gauge assembly based on the comparison result of the state model.

[0059] An early warning unit, configured to determine an abnormal area based on the abnormal state and send corresponding early warning information based on each abnormal area when the current operating state of the oil level gauge assembly is abnormal.

[0060] Specifically, the acquisition unit acquires sensor data, which includes but is not limited to oil level value data, oil temperature data, effective value data of the transformer load current, external environment temperature and humidity data, vibration intensity of the oil conservator (oil storage tank), etc. The state model is an LSTM-Autoencoder model, which can predict the short-term change trend of the oil level. At the same time, it is sensitive to hidden faults such as slow leakage and false oil level, and can identify these hidden faults. Through the prediction of the state model, the complex time series relationship between the oil level and the oil temperature can be captured, adapting to non-linear changes. Furthermore, by using the dual indicators of reconstruction error and prediction deviation, the detection reliability can be improved and the false alarm rate can be reduced. In the actual use process, sudden changes in the oil level may occur. Therefore, the KNN interpolation is used to process the missing values of the sensor signals. At the same time, the Kalman filter is combined with wavelet denoising, and the filtering parameters can be adjusted to suppress high-frequency noise and low-frequency drift. The low-frequency parameters (such as external environment temperature data) are upsampled and aligned with the high-frequency oil level data. And the sensor data is stored in the historical database, and the parameters of the model are fine-tuned regularly with new historical data to make it adapt to the change of data distribution. The judgment unit can identify abnormal situations by comparing the reconstructed oil level data. Since all the data used in the training of the state model are data operating under normal working conditions, when abnormal situations occur, the data cannot match the state model, resulting in a reconstruction error. Thus, the future change trend can be predicted, and abnormal situations can also be identified. Combining with the actual measured data, such as the actual oil level data, actual oil temperature data, etc., the difference between the predicted situation and the actual situation can be seen. Combining with the output prediction value of the state model, the abnormal area or the cause of the abnormal situation can be determined more accurately. Then, the early warning unit sends out early warning information, which is convenient for the staff to quickly find the abnormal situation and solve it.

[0061] It is understandable that by integrating multi-source sensor data acquisition and an intelligent preprocessing mechanism, the data integrity and reliability are improved. For the dynamic change characteristics of parameters such as oil level, oil temperature, and load current, time series alignment and interpolation filling techniques are adopted to solve the problem of data loss caused by sensor communication delay or temporary failure. Combined with an adaptive filtering algorithm, the interference of high-frequency noise and low-frequency drift on the original signal is suppressed, and the risk of measurement distortion caused by environmental factors is reduced. In addition, through the upsampling process of low-frequency parameters, synchronous analysis of multi-frequency data is achieved, enhancing the accuracy of cross-parameter correlation analysis and providing a consistent input basis for subsequent model inference. Based on deep time series modeling technology, the characterization ability of the dynamic change of oil level and the coupling relationship of multi-parameters is constructed. By capturing the complex non-linear correlations between parameters such as oil level, oil temperature, and load current, the model can identify hidden fault modes (such as slow leakage and float jamming) that are difficult to detect by traditional threshold methods. The dual criterion design of reconstruction error and prediction deviation reduces the misjudgment probability caused by a single index, and at the same time enhances the ability to distinguish sudden anomalies and progressive faults. During the model training process, focusing on normal operating condition data makes it easier to capture the pattern deviation under abnormal conditions, thus enhancing the sensitivity to early faults. The continuous update mechanism of the historical database and the model parameter fine-tuning strategy ensure the adaptability of the system to long-term drifts such as equipment aging and environmental changes. By regularly integrating new data to re-optimize the model, the problem of performance degradation caused by data distribution shift is avoided. The threshold adjustment function further balances the false alarm rate and the miss rate, enabling the system to maintain stable detection performance in different seasons or operating stages and enhancing the sustainable service ability in complex industrial scenarios. The combination of anomaly detection results and multi-dimensional parameter correlation analysis provides support for fault root cause location. By comparing the deviation characteristics between the predicted trend and the actual measured value, different anomaly types such as mechanical faults, sensor failures, or external interferences can be distinguished, and targeted warning information can be generated. This directional diagnosis ability shortens the fault troubleshooting path of maintenance personnel and reduces the secondary risks that may be brought by blind maintenance. At the same time, the structured storage of historical data provides a data basis for fault mode backtracking and experience accumulation, helping to form a preventive maintenance strategy. For challenges such as electromagnetic interference and mechanical vibration commonly found in industrial sites, the signal processing link adopts a multi-stage filtering and redundancy check mechanism to reduce the impact of instantaneous interference on the detection results. The introduction of time series dependence modeling in the model design can distinguish the real oil level change from the short-term fluctuations caused by noise, avoiding invalid alarms triggered by occasional anomalies.

[0062] In some embodiments of the present application, when inputting sensor data into a pre-trained state model for comparison, it includes:

[0063] Obtain two sets of time series of sensor data. One time series includes the time series waveform of the oil level value and the time series waveform of the oil temperature data, and the other time series includes the effective value of the load current, the ambient temperature and humidity, and the vibration intensity of the conservator.

[0064] In the underlying feature learning stage, splice the time series waveforms of the two sets of time series into the oil level-oil temperature joint feature and the auxiliary feature, and form the enhanced time series feature based on the oil level-oil temperature joint feature and the auxiliary feature.

[0065] In the high-level feature learning stage, use a dense network with a two-layer structure to extract high-order non-linear features layer by layer based on the enhanced time series feature to form a state model.

[0066] Input the sensor data into the state model for waveform reconstruction and obtain the oil level prediction value.

[0067] Determine the reconstruction error value and the prediction deviation value of the state model based on the oil level prediction value.

[0068] Specifically, by obtaining the oil level data, the oil temperature data and the auxiliary time series data, the auxiliary time series data includes but is not limited to the effective value of the load current: reflecting the load condition, used to analyze the operating state and load change of the equipment, the ambient temperature: monitoring the ambient temperature around the conservator, used to evaluate the impact of temperature on the equipment operation, the ambient humidity: monitoring the ambient humidity around the conservator, used to evaluate the impact of humidity on the insulation performance of the equipment, the power factor: reflecting the power utilization efficiency of the equipment, used to analyze the operating efficiency and energy consumption of the equipment, the active power and reactive power: used to evaluate the actual power output and energy consumption of the equipment, the equipment temperature: monitoring the temperature of the oil temperature, used to evaluate the operating state and overheating risk of the equipment, the vibration data: monitoring the vibration condition of the equipment, used to judge the mechanical state and potential faults of the equipment, etc. The state model is the LSTM-Autoencoder model. By inputting the multi-variable time series data (such as oil level, oil temperature, load current, etc.) within the time window, the time series features are extracted through the encoder, the latent space representation is compressed, and then the oil level value at the future time step is predicted through the decoder, and the reconstruction error is calculated. In the underlying feature learning stage, by inputting the original waveform, a low-dimensional time series feature vector is obtained, while the auxiliary data encoding is processed by inputting data such as load current, temperature, humidity, etc. through a fully connected encoder. Finally, the oil level-oil temperature feature and the auxiliary feature are spliced to obtain the enhanced time series feature. Furthermore, in the high-level feature learning, training is performed through a dense network. Finally, by inputting new data that has not participated in the training, the reconstruction error value and the prediction deviation value can be obtained, that is, the reconstruction error value:

[0069] , where, is the reconstruction error value, is the actual oil level value at time t (unit: %), The oil level value for model reconstruction (unit: %), T is the length of the time series window, and the prediction deviation value: , where is the prediction deviation value, is the number of prediction time steps, is the actual oil level value at the k-th future time point, is the oil level value predicted by the model at the k-th future time point.

[0070] In some embodiments of the present application, when determining the reconstruction error value and the prediction deviation value of the state model based on the oil level prediction value, it includes:

[0071] Determining the reconstruction error value based on the time series waveform of the oil level value, and when the external environmental temperature fluctuates, compensating the reconstruction error value with a temperature coefficient.

[0072] Determining the reconstruction error threshold based on the historical normal oil level value.

[0073] Obtaining the actual data of the oil level value, and determining the prediction deviation value based on the oil level prediction value.

[0074] Arranging the prediction deviation values in ascending order, and taking the 95th percentile as the prediction deviation threshold.

[0075] Specifically, the temperature coefficient compensation: , where is the environmental change temperature. If the environmental temperature fluctuates (such as the temperature change is greater than 15°C within 24 hours), then multiply the reconstruction error by the temperature compensation coefficient , and the reconstruction error threshold is determined by using the data for training the state model: , where is the percentile of the dataset used for training the state model, is the set of reconstruction errors of the dataset used for training the state model, is the reconstruction error threshold. For all the prediction errors Arrange them in ascending order, and take the 95th percentile as the prediction deviation threshold.

[0076] In some embodiments of the present application, when determining the current operating state of the oil level gauge assembly based on the comparison result of the state model, it includes:

[0077] Determining the anomaly score based on the prediction deviation threshold and the reconstruction error threshold:

[0078] .

[0079] Where is the anomaly score, and are the weight coefficients, and + = 1, is the reconstruction error value, is the threshold of the reconstruction error, is the prediction deviation value, is the threshold of the prediction deviation value.

[0080] In some embodiments of the present application, when the current operating state of the oil level gauge assembly is abnormal, it includes:

[0081] When the abnormality score is less than or equal to 1, it is determined that the oil level gauge assembly is normal.

[0082] When the abnormality score is greater than 1, it is determined that the oil level gauge assembly is abnormal and the fault feature analysis is triggered.

[0083] The fault feature analysis is divided into false oil level, oil leakage and sensor failure.

[0084] When is greater than 0.7, the reconstruction error value is greater than twice the reconstruction error threshold, and the prediction deviation value is less than the prediction deviation threshold, it is determined as a false oil level fault.

[0085] When is greater than 0.7, the reconstruction error value is less than twice the reconstruction error threshold, and the prediction deviation value is greater than twice the prediction deviation threshold, it is determined as a sensor failure.

[0086] When is greater than 0.7, the reconstruction error value is less than twice the reconstruction error threshold, and the prediction deviation value is greater than twice the prediction deviation threshold, it is determined as an oil leakage fault.

[0087] It is understandable that by integrating the dual criteria of reconstruction error and prediction deviation, abnormal signals in oil level monitoring can be captured more comprehensively. The reconstruction error focuses on the sensitivity to the deviation of the current data pattern, while the prediction deviation strengthens the early warning ability for abnormal future trends. The weighted combination of the two reduces the risk of misjudgment that may be caused by a single indicator. For example, relying solely on the reconstruction error may ignore the early signs of slow leakage, while relying solely on the prediction deviation may over-respond to instantaneous noise. This synergy improves the coverage and reliability of anomaly detection, ensuring the accurate capture of complex fault modes (such as intermittent jamming or gradual oil leakage). The decision-making logic based on the proportion of anomaly score components and the threshold combination improves the diagnostic accuracy of the root cause of the fault. By setting different thresholds for the weight coefficients (such as emphasizing the reconstruction error for false oil level and emphasizing the prediction deviation for oil leakage), mechanical failures, sensor failures, and real oil level anomalies can be distinguished. For example, in the determination of false oil level, the over-limit of the reconstruction error and the relative stability of the prediction deviation can exclude the interference of sensor noise or environmental interference. In the determination of oil leakage, the continuous exceeding of the prediction deviation and the controllable range of the reconstruction error can reflect the real risk of the continuous decline of the oil level. By refined classification, the possibility of fault confusion is reduced, providing a clear basis for targeted maintenance. Adopting multi-level threshold determination (such as twice the error threshold) can adapt to different severity levels of fault scenarios. For the significant over-limit of the reconstruction error or prediction deviation (such as exceeding twice the threshold), high-risk faults (such as sudden sensor failure or serious oil leakage) can be quickly identified, shortening the response time for critical faults. For the case of mild over-limit, it is further verified through a comprehensive scoring mechanism combined with other parameters to avoid over-responding to occasional fluctuations, thereby balancing the detection sensitivity and specificity and improving the stability under complex working conditions. The direct mapping relationship between the anomaly score and the fault type simplifies the diagnostic process for maintenance personnel. By clearly distinguishing false oil level, oil leakage, and sensor faults, targeted alarm information can be generated to guide maintenance personnel to prioritize the investigation of high-probability fault points. For example, a false oil level alarm can directly prompt an inspection of the float mechanism or transmission components, while a sensor fault alarm preferably recommends calibrating the sensor circuit or communication link. The targeted guidance reduces the time loss of blind investigation and reduces unnecessary equipment downtime caused by misjudgment.

[0088] In some embodiments of the present application, when the current operating state of the oil level gauge assembly is abnormal, it further includes:

[0089] When the fault feature analysis determines a false oil level fault, but the oil level value suddenly changes, it is preferably determined as a sensor fault.

[0090] When the fault feature analysis determines a false oil level fault, but the effective value of the load current suddenly increases and the vibration intensity of the conservator changes synchronously, it is preferably determined as vibration interference.

[0091] In some embodiments of the present application, when determining the abnormal area based on the abnormal state and sending the corresponding warning information based on each abnormal area, it includes:

[0092] When the warning unit determines a false oil level, a first-level warning information is generated.

[0093] When the warning unit determines an oil leak, a second-level warning information is generated.

[0094] When the warning unit determines a sensor failure, a third-level warning information is generated.

[0095] It can be understood that by introducing a dynamic priority determination mechanism, it is possible to more accurately distinguish similar fault scenarios and avoid diagnostic deviations caused by misjudgment of a single feature. For example, when a false oil level fault and an oil level mutation occur simultaneously, giving priority to determining a sensor failure combines the instantaneous nature of the mutation signal and the persistence feature of mechanical jamming, thereby excluding the interference of mechanical faults. This multi-parameter cross-verification mechanism enhances the reliability of fault location, reduces the dependence on a single data source, and thus reduces the risk of misjudgment caused by local anomalies. In a scenario where the load current suddenly increases and the vibration changes synchronously, it is preferentially determined as mechanical vibration interference rather than a false oil level, improving the ability to distinguish complex anomalies. By correlating the co-variation of electrical parameters (load current) and mechanical parameters (vibration intensity), it is possible to identify the indirect impact of the external environment (such as transformer overload) on the oil level data and avoid misattributing the oil level fluctuation caused by mechanical vibration to a stuck float or sensor failure. The robustness in a complex industrial environment is improved through multi-modal data analysis. Through a hierarchical warning mechanism (from the first level to the third level), it is possible to push warning information differently according to the fault type and the degree of urgency. For example, the first-level warning (false oil level) prompts an inspection of mechanical components, the second-level warning (oil leak) triggers an emergency shutdown protection, and the third-level warning (sensor failure) recommends a circuit overhaul. When the warning unit determines vibration interference, the current data is re-compared, and a new judgment is made based on the comparison result of the new state model. If it is still determined as vibration interference, no alarm is given. If it is determined as other anomalies, an alarm is given based on other anomaly information. This hierarchical strategy helps operation and maintenance personnel quickly identify the priority, shorten the response time for critical faults (such as oil leaks), and at the same time avoid overreacting to low-risk events, thereby optimizing the allocation of human resources and maintenance costs. The priority determination rule suppresses the limitations of the traditional static threshold method. For example, in a scenario of oil level mutation, by dynamically switching the determination logic, the possibility of a transient sensor failure is preferentially excluded instead of directly binding it to a mechanical anomaly, reducing the over-reliance on preset thresholds, being able to adapt to complex scenarios of sudden interference or transient signal anomalies, and reducing false alarms caused by accidental events.

[0096] In some embodiments of the present application, when the warning unit determines that the current operating state is abnormal, it stores the abnormal information in the historical database and marks the fault type, and trains the state model at a fixed time interval and updates the threshold value.

[0097] It can be understood that by storing the abnormal information in the historical database and marking the fault type, multi-dimensional operation data and fault cases can be continuously accumulated. The historical data is no longer an isolated event record, but is transformed into a traceable and analyzable knowledge base. By marking the fault type, the data has clear semantic information, providing a structured basis for subsequent root cause analysis and pattern mining, improving the reuse value of fault cases, and reducing the risk of analysis discontinuity caused by experience dependence or personnel flow. Training the state model at regular fixed intervals ensures that the model can be dynamically updated following changes in the device state or the emergence of new fault patterns. By incorporating the latest abnormal data, the model continuously learns new feature distributions and fault laws, avoiding performance degradation caused by equipment aging, environmental changes, or working condition adjustments. It improves the generalization ability of the model to new scenarios and reduces the misjudgment probability of traditional static models due to data distribution shift. At the same time, the incremental training strategy reduces the full dependence on historical data and optimizes the utilization efficiency of computing resources. Based on the updated model and historical abnormal data, the decision thresholds for reconstruction error and prediction deviation can be dynamically adjusted, overcoming the problems of insufficient sensitivity or over-response that may occur in traditional fixed thresholds during long-term operation. For example, after the device enters the stable aging stage, the threshold can be appropriately relaxed to avoid excessive alarms. In high-precision monitoring scenarios, the threshold can be tightened to capture early abnormal signs. This flexibility improves the adaptability to different life cycle stages or external environmental changes, reduces the workload of manual frequent calibration, and updates the threshold according to new data: , where is the forgetting factor (usually taken as 0.9), is the threshold calculated from the latest data, is the old threshold, is the updated threshold.

[0098] In some embodiments of the present application, it further includes: a remote monitoring unit, which is electrically connected to the warning unit and is configured to monitor the operating state of the warning unit and send the operating state monitoring information to the Internet and the cloud data platform.

[0099] It is understandable that through the connection between the remote monitoring unit and the early warning unit, the real-time tracking and cloud synchronization of the oil level monitoring status are achieved. This enables operation and maintenance personnel to obtain the device operation status without geographical restrictions and promptly grasp the occurrence of abnormal situations. The instant transmission of monitoring information ensures the timeliness of fault response, shortens the time window from the occurrence of an anomaly to the intervention of handling, and reduces the risk of fault expansion caused by information delay. At the same time, the access to the cloud platform facilitates multi-terminal access, supports multi-channel monitoring such as mobile terminals and PC terminals, and enhances the flexibility of emergency handling. The introduction of the cloud data platform enables the centralized storage and management of scattered monitoring data. By means of the centralized processing mode, the spatial limitation of traditional local storage is broken, providing a basis for the long-term preservation and rapid retrieval of a large amount of historical data. Through the data aggregation in the cloud, the horizontal comparative analysis across devices and regions can be realized to identify potential regional risks or common fault patterns.

[0100] In summary, the beneficial effects of the present invention are as follows: Through the rigid connection between the fixed flange and the limiting rod, the mechanical strength of the overall structure is enhanced, which can disperse the impact of external vibration or pressure on key components, reduce deformation or fatigue damage caused by long-term stress, and the sliding setting of the float along the limiting rod replaces the free floating mode of the traditional floating ball. By physically guiding and restricting the movement trajectory of the float, the jamming problem caused by mechanical loosening or offset is avoided, and the bilateral support of the limiting rod further improves the stability of the float movement, ensuring its linear response even when the oil density changes or the temperature fluctuates, thereby improving the reliability of long-term operation. In addition, the cooperation between the rotating rod and the limiting rod improves the force transmission path, reduces the wear risk caused by local stress concentration, and the direct linkage between the angle sensor and the rotating rod eliminates the clearance error in traditional indirect transmission (such as gears and chains), enabling the real-time capture of minute changes in the oil level. The linear movement of the threaded rotating rod converts the vertical displacement of the float into a rotation angle signal, avoiding the non-linear error caused by gear meshing or lever swinging, not only improving the sensitivity of oil level detection but also shortening the response time of signal transmission, enabling the control module to quickly reflect the dynamic changes in the oil level. At the same time, the axial thread on the surface of the rotating rod increases the contact area between the float and the rod body, reducing the movement resistance caused by oil viscosity or impurity accumulation, and further improving the measurement consistency. The physical constraint of the limiting rod on the float movement suppresses the false operation caused by violent oil fluctuations or mechanical vibrations, can filter high-frequency interference, and ensures that the float only generates displacement when the oil level truly changes. The electrical connection between the control module and the angle sensor realizes the real-time acquisition and digital conversion of oil level data, providing a basis for remote monitoring and automated analysis. Through the integrated control module, the trend characteristics of oil level changes can be identified, potential faults (such as leakage and jamming) can be warned in advance, and the dependence on manual inspections can be reduced.

[0101] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0102] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one or more of the processes and / or blocks Figure 1 one or more of the blocks.

[0103] These computer program instructions can also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable storage medium generate a manufactured product including an instruction device, and the instruction device implements the functions specified in Figure 1 one or more of the processes and / or blocks Figure 1 one or more of the blocks.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Therefore, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in Figure 1 one or more of the processes and / or blocks Figure 1 one or more of the blocks.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: still can modify the specific implementation manners of the present invention or make equivalent replacements, and any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the protection scope of the claims of the present invention.

Claims

1. A pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank, characterized in that: include: An oil level gauge assembly and a control module, wherein the control module is electrically connected to the oil level gauge assembly, and the control module is used to control the oil level gauge assembly; The oil level gauge assembly includes: A dial, a fixed flange, a limit rod, an angle sensor, a limit block, a rotating rod and a float, wherein the upper end of the fixed flange is fixedly connected to the dial, the angle sensor is electrically connected inside the dial, the lower end of the fixed flange is fixedly connected to the limit rod, the end of the limit rod away from the fixed flange is fixedly connected to the limit block, two limit rods are provided, the rotating rod is provided between the two limit rods, and the two ends of the rotating rod are respectively rotatably connected to the dial and the limit block, the surface of the rotating rod is axially provided with a thread, the float is sleeved on the rotating rod, and the two sides of the float are slidably connected to the two limit rods.

2. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 1 is characterized in that: The control module comprises: A collection unit is configured to collect sensor data, store the sensor data at fixed time intervals, and perform denoising, and the collection unit is also configured to align timestamps of the sensor data and use KNN interpolation to process missing values; a judgment unit configured to extract the sensor data and input the sensor data into a pre-trained state model for comparison, and the judgment unit is further configured to determine the current operating state of the oil level gauge assembly based on the comparison result of the state model; The early warning unit is configured to determine an abnormal area based on the abnormal state and send corresponding early warning information based on each abnormal area when the current operating state of the oil level gauge component is abnormal.

3. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 2 is characterized in that: When the sensor data is input into the pre-trained state model for comparison, it includes: Acquire two sets of time series of the sensor data, wherein one of the time series includes a time series waveform of the oil level value and a time series waveform of the oil temperature data, and the other time series includes an effective value of the load current, external environment temperature and humidity, and vibration intensity of the oil storage cabinet; In the underlying feature learning stage, the time series waveforms of the two groups of time series are spliced ​​into oil level-oil temperature joint features and auxiliary features, and enhanced time series features are formed based on the oil level-oil temperature joint features and the auxiliary features; In the high-level feature learning stage, a dense network with a two-layer structure is used to extract high-order nonlinear features layer by layer based on the enhanced temporal features to form the state model; Inputting the sensor data into the state model, reconstructing the waveform and obtaining the predicted value of the oil level; A reconstruction error value and a prediction deviation value of the state model are determined based on the oil level prediction value.

4. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 3 is characterized in that: When determining the reconstruction error value and the prediction deviation value of the state model based on the oil level prediction value, the method includes: determining the reconstruction error value based on the time series waveform of the oil level value, and performing temperature coefficient compensation on the reconstruction error value when the external ambient temperature fluctuates; determining a reconstruction error threshold based on historical normal oil level values; Acquiring actual data of the oil level value, and determining the prediction deviation value based on the oil level prediction value; The prediction deviation values ​​are arranged in ascending order, and the 95% quantile is taken as the prediction deviation threshold.

5. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 4, characterized in that: Determining the current operating state of the oil level gauge assembly based on the comparison result of the state model includes: Determine anomaly score based on prediction deviation threshold and reconstruction error threshold: ； in, For the abnormality score, and is the weight coefficient, and + =1, is the reconstruction error value, is the reconstruction error threshold, is the prediction deviation value, is the threshold of the prediction deviation value.

6. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 5, characterized in that: When the current operating state of the oil level gauge component is abnormal, it includes: When the abnormality score is less than or equal to 1, it is determined that the oil level gauge assembly is normal; When the abnormality score is greater than 1, the oil level gauge component is determined to be abnormal, and a fault feature analysis is triggered; The fault feature analysis is divided into false oil level, oil leakage and sensor failure; when When the value is greater than 0.7, the reconstruction error value is greater than twice the reconstruction error threshold, and the prediction deviation value is less than the prediction deviation threshold, it is determined to be a false oil level fault; when When the value is greater than 0.7, the reconstruction error value is less than twice the reconstruction error threshold, and the prediction deviation value is greater than twice the prediction deviation threshold, it is determined to be a sensor failure; when When the value of the reconstruction error is greater than 0.7, the reconstruction error value is less than twice the reconstruction error threshold, and the prediction deviation value is greater than twice the prediction deviation threshold, it is determined to be an oil leakage fault.

7. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 6, characterized in that: When the current operating state of the oil level gauge assembly is abnormal, the method further includes: When the fault feature analysis determines that it is a false oil level fault, but the oil level value changes suddenly, it is preferentially determined to be a sensor fault; When the fault characteristic analysis determines that it is a false oil level fault, but the effective value of the load current increases suddenly and the vibration intensity of the oil storage cabinet changes synchronously, it is preferentially determined to be a vibration interference.

8. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 7, characterized in that: When determining abnormal areas based on abnormal states and sending corresponding warning information based on each abnormal area, it includes: When the early warning unit determines that the oil level is false, a first-level early warning message is generated; When the early warning unit determines that there is an oil leak, a second-level early warning message is generated; When the warning unit determines that the sensor is faulty, a third-level warning message is generated.

9. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 8, characterized in that: When the early warning unit determines that the current operating state is abnormal, the abnormal information is stored in the history database and the fault type is marked, and the state model is trained and the threshold is updated at fixed time intervals.

10. The pointer type oil level gauge for monitoring the oil level of a transformer oil storage tank according to claim 8, characterized in that: Also includes: A remote monitoring unit, wherein the remote monitoring unit is electrically connected to the early warning unit, and the remote monitoring unit is configured to monitor the operating status of the early warning unit and send operating status monitoring information to the Internet and a cloud data platform.

Images