A method and system for judging the oil level status of an oil storage cabinet transformer

Abstract

The present invention relates to the field of oil level monitoring technology, specifically to a method and system for judging the oil level status of an oil storage cabinet transformer, comprising the following steps: collecting oil level signals through sensors, filtering the collected oil level signals in real time, adjusting the filter coefficients according to the environmental noise level, and recording and outputting the adjusted oil level signals. In the present invention, by adjusting the filter coefficients to adapt to changes in environmental noise, a high degree of adaptability in oil level signal processing is achieved, and frequency domain analysis enables the system to identify specific frequency components related to normal operation and potential fault states of the equipment, greatly enhancing the meticulousness of monitoring and early warning capabilities. By dynamically monitoring changes in frequency components, anomalies can be detected at the initial stage, effectively preventing equipment failures, not only enhancing real-time monitoring capabilities, but also improving the response speed to complex changes, providing comprehensive safety protection for transformer oil level monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of oil level monitoring, and in particular to a method and system for judging the oil level status of an oil storage cabinet transformer. Background Art

[0002] Oil level monitoring technology involves the use of various sensors and measuring devices to accurately detect and monitor the height of liquids (especially oil) in storage containers or systems. Oil level monitoring can help prevent liquid leaks, ensure safe operation of equipment, and optimize the use of resources. It usually includes sensors for real-time data acquisition, and software that analyzes data to predict and solve problems. As technology develops, modern oil level monitoring systems increasingly use wireless sensors and cloud-based data analysis to enhance their flexibility and scalability.

[0003] Among them, the oil level status judgment method of the oil conservator transformer refers to a method for judging and monitoring the oil level status in the transformer oil conservator. The core is to ensure that the oil level in the transformer cooling system is in a suitable state, thereby preventing the transformer from overheating or being damaged due to insufficient oil level. It usually includes using an oil level sensor to detect the oil level height in the oil tank, and analyzing the data through an algorithm or control system to ensure that the oil level is within a safe and effective range, which is crucial to ensuring the stable operation of the power system and the long-term reliability of its equipment.

[0004] Existing technologies show their limitations when processing rapidly changing or complex data, especially when they fail to fully utilize real-time environmental information and lack efficient processing mechanisms. In traditional methods, the collection and analysis process of oil level data fails to achieve efficient real-time response, especially in terms of adjustments to environmental noise, which is usually passive, resulting in increased delays and errors in the data processing process. In addition, the analysis methods of existing technologies usually do not involve in-depth frequency domain analysis, which limits their ability to diagnose potential equipment problems at an early stage. For example, when encountering rapid changes in oil levels, the monitoring strategy cannot be adjusted in time, resulting in delays or omissions of key alarms, which not only affects the normal operation of the equipment, but also poses a threat to the long-term stability of the system. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a method for judging the oil level status of an oil storage cabinet transformer.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: a method for judging the oil level status of an oil storage cabinet transformer, comprising the following steps:

[0007] S1: collect oil level signals through sensors, filter the collected oil level signals in real time, adjust the filter coefficient according to the environmental noise level, and record and output the adjusted oil level signals;

[0008] S2: using the adjusted oil level signal, converting the oil level signal from the time domain to the frequency domain through digital signal processing, analyzing all frequency components in the spectrum, calibrating each frequency component according to the analysis result, identifying the target frequency components associated with the normal operation and potential fault state of the equipment, and obtaining the spectrum analysis result;

[0009] S3: Based on the spectrum analysis results, by dynamically monitoring the changes in the target frequency components, recording the frequency disturbances in the abnormal variation range, and recording and analyzing the abnormal frequency components in real time according to the recording results, revealing the real-time changes in the oil level signal, and generating abnormal frequency records;

[0010] S4: According to the abnormal frequency record, the alarm triggering condition is adjusted and updated to reflect the real-time oil level change of the oil storage cabinet transformer, dynamically update the oil level abnormal status monitoring, and generate a real-time alarm update record.

[0011] As a further solution of the present invention, the step of obtaining the adjusted oil level signal is:

[0012] S111: Install and configure the oil level sensor, collect oil level data at the set detection point, output an electrical signal through the sensor to indicate the oil level height, and obtain the original oil level signal;

[0013] S112: Based on the original oil level signal, perform real-time filtering processing, using the formula:

[0014]

[0015] After calculation, the oil level signal value after filtering is generated. ,in, Represents the original signal at the current moment, and Represent the historical lowest and highest original signal values ​​respectively, is the regulating factor, It is the filtering result of the previous cycle;

[0016] S113: Dynamically adjust the adjustment factor according to the oil level signal after filtering and the current environmental noise data , re-optimize the oil level signal accuracy, record and output the adjusted oil level signal.

[0017] As a further solution of the present invention, the frequency component analysis step is:

[0018] S211: Based on the adjusted oil level signal, start signal preprocessing, convert the oil level signal in the time domain into a frequency domain signal by fast Fourier transform, and obtain a preliminary frequency domain converted oil level signal;

[0019] S212: Analyze the preliminary frequency domain converted oil level signal, evaluate the energy distribution of the frequency components in the signal, and generate power spectrum characteristic data;

[0020] S213: Based on the power spectrum characteristic data, key frequency components are screened and the impact on device performance is evaluated. The formula is used:

[0021]

[0022] Calculate the energy value of key frequency components , and get the frequency component analysis results, where Indicates frequency The power at Representative frequency The frequency domain signal at is the total number of samples.

[0023] As a further solution of the present invention, the step of obtaining the spectrum analysis result is:

[0024] S221: Based on the frequency component analysis result, frequency point screening is performed, frequency components with signal energy higher than a set threshold are selected, marked as key indicators for equipment performance monitoring, and a preliminary screening frequency data set is generated;

[0025] S222: using the preliminary screening frequency data set, combined with the historical operation records and fault logs of the equipment, performing data matching analysis, determining the correlation between the frequency components and the equipment status, and obtaining the equipment status associated frequency data;

[0026] S223: Statistically analyze the device status association frequency data, extract key variables from the data, and use the formula:

[0027]

[0028] Calculate principal component analysis values , and get the spectrum analysis results, where represents the eigenvectors of the covariance matrix, Represents the frequency value in the frequency data set obtained by Fourier transform, represents the mean of the data, is a single frequency data point.

[0029] As a further solution of the present invention, the step of recording the frequency disturbance within the abnormal variation range is:

[0030] S311: extracting key frequency components based on the spectrum analysis results, monitoring changes in amplitude and phase in real time, and generating a target frequency component change monitoring record;

[0031] S312: Analyze the target frequency component change monitoring record, identify abnormal frequency disturbances through a preset threshold, and generate abnormal frequency change range records;

[0032] S313: Systematically store the abnormal frequency variation range record, mark the time stamp to track the actual time when the abnormality occurs, and generate a frequency disturbance record.

[0033] As a further solution of the present invention, the step of obtaining the abnormal frequency record is:

[0034] S321: Based on the frequency disturbance record, the frequency components in the record are tracked in real time, spectrum dynamic analysis is performed, changes at each frequency point are monitored, each significant frequency disturbance is marked, the time series and amplitude of the frequency components are recorded, and a real-time frequency analysis record is generated;

[0035] S322: Analyze the change of the oil level signal by using the real-time frequency analysis record, distinguish normal fluctuations from frequency disturbances caused by oil level changes, and generate an oil level signal change record;

[0036] S323: The oil level signal change records are summarized and uploaded to a central database, where the records include the frequency disturbance, time tag and oil level reading corresponding to each oil level change, and an abnormal frequency record is generated.

[0037] As a further solution of the present invention, the step of obtaining the real-time alarm update record is:

[0038] S411: Based on the abnormal frequency record, analyze the relationship between the current oil level data and the historical alarm record, locate the false alarm oil level threshold, and obtain the initial alarm triggering condition;

[0039] S412: Combining the initial alarm triggering conditions with the real-time monitoring oil level data, using the formula:

[0040]

[0041] Calculate oil level alarm update threshold , output new oil level abnormal state monitoring parameters, where, Represents the average value of the oil level in the past. Indicates the minimum value of the oil level in the past. Represents the difference between the current real-time oil level and the historical average oil level;

[0042] S413: Dynamically monitor the real-time oil level data in combination with the oil level abnormal state monitoring parameters, and if it is detected that the oil level exceeds the oil level alarm update threshold, record and generate a real-time alarm update record.

[0043] An oil level status judgment system for an oil storage tank transformer, comprising:

[0044] The oil level signal adjustment module is installed and configured with an oil level sensor, collects oil level data at set detection points, performs real-time filtering, dynamically adjusts the adjustment factor in combination with current environmental noise data, optimizes the accuracy of the oil level signal, and records and outputs the adjusted oil level signal;

[0045] The frequency domain conversion evaluation module converts the time domain oil level signal into a frequency domain signal based on the adjusted oil level signal, evaluates the energy distribution of the frequency components in the signal, calculates the energy value of the key frequency components, evaluates the impact on the equipment performance, and obtains the frequency component analysis result;

[0046] The spectrum analysis and identification module performs frequency point screening based on the frequency component analysis results, combines the historical operation records and fault logs of the equipment, determines the correlation between the frequency components and the equipment status, extracts key variables, and obtains the spectrum analysis results;

[0047] The frequency disturbance monitoring module monitors the changes of amplitude and phase in real time based on the spectrum analysis results, identifies abnormal frequency disturbances through preset thresholds, marks timestamps to track the actual time when the abnormality occurs, and generates frequency disturbance records;

[0048] Based on the frequency disturbance record, the abnormal frequency tracking module tracks the frequency components in the record in real time, monitors the changes of each frequency point, analyzes the changes of the oil level signal, distinguishes normal fluctuations from frequency disturbances caused by oil level changes, summarizes and uploads them to the central database, and generates abnormal frequency records;

[0049] The alarm update and recording module analyzes the relationship between the current oil level data and the historical alarm records based on the abnormal frequency records, calculates the oil level alarm update threshold in combination with the real-time monitoring oil level data, and dynamically monitors the real-time oil level data. If it is detected that the oil level exceeds the oil level alarm update threshold, the real-time alarm update record is recorded and generated.

[0050] Compared with the prior art, the advantages and positive effects of the present invention are:

[0051] In the present invention, by adjusting the filter coefficient to adapt to the change of environmental noise, the high adaptability of oil level signal processing is achieved. Frequency domain analysis enables the system to identify specific frequency components related to the normal operation and potential fault status of the equipment, greatly enhancing the detailed monitoring and early warning capabilities. By dynamically monitoring the changes in frequency components, anomalies can be detected at the initial stage, effectively preventing equipment failures. This not only enhances the real-time monitoring capability, but also improves the response speed to complex changes, providing comprehensive safety protection for transformer oil level monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a main step flow chart of the present invention;

[0053] Figure 2 The flowchart of obtaining the oil level signal after adjustment of the present invention;

[0054] Figure 3 is a flow chart of the analysis of frequency components of the present invention;

[0055] Figure 4 A flowchart of obtaining spectrum analysis results of the present invention;

[0056] Figure 5 A flow chart for recording frequency disturbances within the abnormal variation range of the present invention;

[0057] Figure 6 The flowchart of obtaining abnormal frequency records of the present invention;

[0058] Figure 7 This is a flow chart for obtaining real-time alarm update records of the present invention. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0060] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0061] See also Figure 1 , a method for judging the oil level status of an oil storage cabinet transformer, comprising the following steps:

[0062] S1: collect oil level signals through sensors, filter the collected oil level signals in real time, adjust the filter coefficient according to the environmental noise level, and record and output the adjusted oil level signals;

[0063] S2: Using the adjusted oil level signal, convert the oil level signal from the time domain to the frequency domain through digital signal processing, analyze all frequency components in the spectrum, calibrate each frequency component according to the analysis results, identify the target frequency components associated with the normal operation and potential fault status of the equipment, and obtain the spectrum analysis results;

[0064] S3: Based on the spectrum analysis results, the target frequency components are dynamically monitored to record the frequency disturbances in the abnormal variation range. The abnormal frequency components are recorded and analyzed in real time according to the recording results to reveal the real-time changes of the oil level signal and generate abnormal frequency records.

[0065] S4: According to the abnormal frequency record, the alarm triggering conditions are adjusted and updated to reflect the real-time oil level changes of the oil storage cabinet transformer, dynamically update the abnormal oil level status monitoring, and generate real-time alarm update records.

[0066] The adjusted oil level signal includes signal amplitude, signal phase and filter coefficient. The spectrum analysis results include target frequency components, frequency amplitude and frequency phase. The abnormal frequency records include abnormal frequency records and variation range. The real-time alarm update records include alarm trigger conditions, alarm status and updated timestamp.

[0067] See also Figure 2 , the steps to obtain the adjusted oil level signal are:

[0068] S111: Install and configure the oil level sensor, collect oil level data at the set detection point, output an electrical signal through the sensor to indicate the oil level height, and obtain the original oil level signal;

[0069] During the installation and configuration of the oil level sensor, first select an appropriate installation location in the oil storage cabinet transformer. Under the premise of ensuring that there is no external interference, install a special liquid level detection sensor. The sensor can monitor the changes in the oil level in real time and convert the physical height into an electrical signal output. In order to ensure the accuracy of the data, the selection and setting of the sensor are very critical, including the sensitivity and accuracy of the sensor. The parameters directly affect the subsequent data processing and analysis effects. By arranging sensors at different detection points, the comprehensive information of the oil level can be captured, so that the oil level height is represented by the electrical signal output. The process needs to be adjusted according to the design parameters and environmental factors of the oil tank to adapt to different working conditions. The original oil level signal obtained will provide the basis for subsequent data processing.

[0070] S112: Based on the original oil level signal, perform real-time filtering processing using the formula:

[0071]

[0072] After calculation, the oil level signal value after filtering is generated. ,in, Represents the original signal at the current moment, and Represent the historical lowest and highest original signal values ​​respectively, is the regulating factor, It is the filtering result of the previous cycle;

[0073] Represents the value of the original oil level signal at the current moment, which is recorded as 50. and are the minimum and maximum values ​​recorded during the monitoring period, which are 20 and 80 respectively. It is the filter coefficient adjusted according to the ambient noise level. The current adjustment value is 0.6. is the filtering result of the previous moment, which is 45. The calculation process is as follows:

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081] The results show that considering the current noise environment and historical data, the filtered oil level signal calculated this time is 18.3, which is used to further optimize the response sensitivity and accuracy of the oil level monitoring system and reduce the impact of noise interference.

[0082] S113: Dynamically adjust the adjustment factor based on the oil level signal after filtering and the current environmental noise data , re-optimize the oil level signal accuracy, record and output the adjusted oil level signal;

[0083] The process of adjusting the filter coefficient according to the current environmental noise data involves continuous monitoring and data analysis of environmental noise. Sensors are set up to collect sound data from the surrounding environment, and the sound intensity is used to determine the adjustment value of the filter coefficient. The collection of sound data requires the use of highly sensitive sound sensors. The sensors can identify different levels of sound intensity and convert the sound into electrical signals. The noise level is determined by analyzing the signal intensity. This data is compared with a preset noise threshold and the filter coefficient is adjusted based on the comparison result. The adjustment process is dynamic to adapt to different working environments and ensure the adaptability and flexibility of data processing. Finally, the adjusted oil level signal is output, indicating an improvement in adaptability and accuracy, which is a direct improvement in the responsiveness of the oil level monitoring system.

[0084] See also Figure 3 , the analysis steps of frequency components are:

[0085] S211: Based on the adjusted oil level signal, start signal preprocessing, convert the oil level signal in the time domain into a frequency domain signal by fast Fourier transform, and obtain a preliminary frequency domain converted oil level signal;

[0086] Start signal preprocessing. First, adjust the original oil level signal data. The main purpose is to optimize the signal quality and improve the accuracy of subsequent analysis. The spectrum leakage effect is reduced by using the Hanning window technology. This technology can effectively suppress the jumps at both ends of the frequency domain signal, thereby reducing the analysis error. Next, use the fast Fourier transform (FFT) technology to convert the adjusted time domain oil level signal into a frequency domain signal. FFT is an efficient frequency domain conversion method that can convert complex time domain signals into a combination of different frequency components. The conversion process is not only fast but also can retain all the information in the signal. Through two-step processing, the main frequency components of the oil level signal are enhanced, and unnecessary noise and interference are removed. The final frequency domain signal, that is, the preliminary frequency domain conversion oil level signal, provides accurate basic data for the next step of energy spectrum analysis and frequency component evaluation.

[0087] S212: analyzing the preliminary frequency domain converted oil level signal, evaluating the energy distribution of the frequency components in the signal, and generating power spectrum characteristic data;

[0088] After obtaining the preliminary frequency domain conversion oil level signal, a detailed frequency domain analysis is performed, mainly using the power spectrum density estimation method to evaluate the energy distribution of each frequency component in the signal. Power spectrum density is an important tool for describing the frequency domain characteristics of a signal. By calculating the power spectrum density of each frequency point, not only can the main energy peaks in the signal be identified, but also the energy distribution of the signal in the frequency domain can be understood in detail. By quantifying the energy of each frequency component, the frequencies that are most critical to the performance of the oil level monitoring system can be identified. During this analysis process, the power spectrum density of each frequency component is calculated, and then the important energy peaks are calibrated. This information is crucial for subsequent fault diagnosis, system maintenance, and performance evaluation. It provides a method based on energy distribution to analyze and interpret the frequency domain characteristics of the oil level signal, ensuring the efficient operation of the entire monitoring system.

[0089] S213: Based on the power spectrum characteristic data, filter the key frequency components and evaluate the impact on the equipment performance. Use the formula:

[0090]

[0091] Calculate the energy value of key frequency components , and get the frequency component analysis results, where Indicates frequency The power at Representative frequency The frequency domain signal at is the total number of samples;

[0092] N is 1024, The value of is 50 (the amplitude of a certain frequency component after FFT processing), and the calculation process is as follows:

[0093]

[0094] The results show that at the frequency At , the energy of the signal is 2.441 units, which shows the importance of this frequency component in the entire signal. It plays a key role in identifying the operating status or potential faults of the equipment. The numerical results can be used for further equipment maintenance decisions or fault diagnosis.

[0095] See also Figure 4 , the steps to obtain the spectrum analysis results are:

[0096] S221: Based on the frequency component analysis results, frequency point screening is performed to select frequency components whose signal energy is higher than a set threshold, mark them as key indicators for equipment performance monitoring, and generate a preliminary screening frequency data set;

[0097] Based on the screening work of power spectrum characteristic data, the frequency components exceeding a specific threshold are selected as key monitoring indicators. This threshold is obtained by analyzing the relationship between the historical operation data of the equipment and the frequency distribution. The purpose is to improve the pertinence and accuracy of monitoring. In the step, statistical analysis technology is used to determine the energy threshold for screening. The preliminary screening of frequency points is carried out by an automated script. The script automatically marks the frequency points exceeding the energy threshold according to the set algorithm and compiles the data points into the preliminary screening frequency data set. The data set will be used for more in-depth analysis in the future to more accurately identify abnormal frequency components that affect equipment performance.

[0098] S222: using the preliminary screening frequency data set, combined with the historical operation records and fault logs of the equipment, performing data matching analysis, determining the correlation between the frequency components and the equipment status, and obtaining the equipment status associated frequency data;

[0099] Using the filtered frequency data set, combined with the historical operation records and fault logs of the equipment, an in-depth data matching analysis is performed. In this step, a linear regression model is used to determine the correlation between frequency components and equipment status, as well as their relationship with equipment failure and performance degradation. This approach not only enhances the accuracy of fault prediction, but also provides a scientific basis for the formulation of equipment maintenance strategies. During the analysis process, each frequency point undergoes detailed statistical tests to ensure the accuracy of its correlation calculation. In addition, the analysis results are presented through visualization technology, which can intuitively see which frequencies have a strong correlation with equipment failures. The final equipment status-associated frequency data lists in detail the direct correlation between each key frequency point and the equipment status, which is crucial for subsequent maintenance decisions and fault response plan formulation.

[0100] S223: Statistical analysis is performed on the equipment status correlation frequency data, key variables in the data are extracted, and the formula is used.

[0101]

[0102] Calculate principal component analysis values , and get the spectrum analysis results, where represents the eigenvectors of the covariance matrix, Represents the frequency value in the frequency data set obtained by Fourier transform, represents the mean of the data, is a single frequency data point;

[0103] There is a set of frequency data , indicating the frequency value detected by the device at different time points.

[0104] Calculate the average :

[0105]

[0106] Calculate the difference between each frequency point and the mean :

[0107]

[0108] Calculate the sum of the squares of the differences and take the square root to get the denominator:

[0109]

[0110]

[0111] Eigenvector ,but The calculation is as follows:

[0112]

[0113]

[0114]

[0115] Substitute the formula to calculate PCA(F):

[0116]

[0117] The results show that, using the current eigenvector and data set, the principal component analysis value after PCA processing is 0, indicating that under the selected eigenvector, the data set has no change in the principal component direction, or the change is balanced by the eigenvector, which helps to understand the main change trend of the oil level signal data and provides a scientific basis for the monitoring and prediction of the oil level status.

[0118] See also Figure 5 , the steps for recording the frequency disturbance of the abnormal variation range are,

[0119] S311: Based on the spectrum analysis results, extract key frequency components, monitor the changes in amplitude and phase in real time, and generate target frequency component change monitoring records;

[0120] Based on the results of spectrum analysis, key frequency components are extracted, and the changes in amplitude and phase of these components are monitored. Each frequency fluctuation is continuously recorded by the sensor, including the specific value of the frequency and its corresponding timestamp, to ensure that each data point can accurately correspond to the observation time. By performing real-time analysis on these monitoring data, the amplitude or phase changes that deviate from the normal range can be quickly identified. If the change exceeds the preset threshold, the system will automatically record these abnormal events. The recorded data not only helps us track the trend of frequency changes, but also provides fault warnings when necessary. This method can effectively prevent potential equipment failures or operating errors caused by frequency anomalies, thereby ensuring the stable operation of the system. The monitoring and recording process of the target frequency components is fully automatic, which greatly improves the monitoring efficiency and response speed. The monitoring records finally generated are the basis for subsequent analysis.

[0121] S312: Analyze the target frequency component change monitoring record, identify the abnormal frequency disturbance through the preset threshold, and generate the abnormal frequency change range record;

[0122] After monitoring and recording the changes in the target frequency components, further analysis is performed, especially focusing on abnormal frequency changes, including comparing the real-time monitored data with the set threshold. The threshold is set based on historical data analysis, and the standard deviation and mean are used to define the preset threshold. When the current data exceeds this defined range, it is considered abnormal. Whenever the monitored frequency data exceeds the normal range, the system will generate an abnormal frequency change record, which lists in detail the occurrence time and specific frequency change values ​​of all abnormal events. Detailed data is very important, not only helping engineers to quickly locate problems, but also can be used for trend analysis and prediction of possible future failures. Through in-depth data analysis, we can better understand the operating status of the equipment and take preventive measures to avoid larger failures. Records of abnormal frequency changes provide strong data support for equipment maintenance and fault diagnosis.

[0123] S313: Systematically store the abnormal frequency variation range records, mark the timestamp to track the actual time when the abnormality occurs, and generate frequency disturbance records;

[0124] Determine the records of abnormal frequency changes and store the data in a systematic manner for future query and analysis. This storage process uses database management, which not only ensures data security, but also improves the efficiency of data retrieval. Each record of abnormal frequency is assigned a unique timestamp, which is automatically added based on the actual time when the record is generated, allowing accurate tracking of the specific time when each abnormality occurs, which is essential for analyzing the pattern of frequency changes and identifying potential causes of problems. The introduction of timestamps also facilitates the comparison and trend analysis of historical data. In addition, the systematic storage method allows data to be quickly traced back and called, greatly facilitating daily operation and maintenance and fault response work. Through detailed records and accurate time stamps, the maintenance team can monitor and maintain the equipment more effectively to ensure the reliability and safety of equipment operation.

[0125] See also Figure 6 ,The steps to obtain abnormal frequency records are,

[0126] S321: Based on the frequency disturbance record, the frequency components in the record are tracked in real time, the spectrum dynamic analysis is performed, the changes of each frequency point are monitored, each significant frequency disturbance is marked, the time series and amplitude of the frequency components are recorded, and a real-time frequency analysis record is generated;

[0127] Based on frequency disturbance records, the detailed changes of each frequency point are monitored and recorded in real time, especially the frequencies beyond the normal fluctuation range. Through precise sensors and high-frequency data acquisition technology, each frequency fluctuation is dynamically captured, and the records are updated in real time using high-speed data processing algorithms. Accurate timestamps and amplitude values ​​are marked for each significant frequency disturbance. The data is automatically processed by the real-time monitoring system to ensure that all important frequency changes are recorded and analyzed in a timely manner. Key information can be quickly identified from complex data streams and converted into actionable data points for further fault analysis and system optimization. In addition, the real-time recorded data also supports continuous monitoring of the equipment's operating status, providing a data basis for equipment fault warning and performance evaluation.

[0128] S322: Analyze the change of the oil level signal by using the real-time frequency analysis record, distinguish normal fluctuations from frequency disturbances caused by the oil level change, and generate an oil level signal change record;

[0129] Using the data obtained from the real-time frequency analysis records, we deeply analyze the changes in the oil level signal. We can accurately distinguish the specific frequency disturbances caused by oil level changes and the noise caused by other external or internal factors of the equipment. First, we eliminate the irrelevant frequency components, and then conduct in-depth analysis of the key frequency points to evaluate the correlation between their change patterns and oil level changes. This process not only includes conventional frequency analysis, but also introduces machine learning technology to enhance recognition accuracy. It can extract frequency components directly related to oil level changes from continuous data streams and generate oil level signal change records, which provides accurate data support for oil level monitoring and effectively guides oil depot management and operational decisions.

[0130] S323: Summarize and upload the oil level signal change records to the central database, including the frequency disturbance, time tag and oil level reading corresponding to each oil level change, and generate abnormal frequency records;

[0131] The oil level signal change records obtained through analysis are summarized and uploaded to the central database for systematic management. During the process, each record includes the specific value of the frequency disturbance, the corresponding time point and the oil level reading. The integrated storage of data not only ensures the integrity and traceability of the information, but also improves the security and accessibility of the data. The design of the central database supports efficient data retrieval and complex data analysis, allowing operators to easily access historical and real-time data, and then conduct trend analysis and prediction of oil level changes. The systematic data management method provides a solid data foundation for daily maintenance and fault response of equipment. Through continuous data monitoring and intelligent analysis, the accuracy and response speed of oil level management can be significantly improved, bringing unprecedented operational convenience and efficiency improvement to oil depot management.

[0132] See also Figure 7,The steps to obtain real-time alarm update records are,

[0133] S411: Based on the abnormal frequency record, analyze the relationship between the current oil level data and the historical alarm record, locate the false alarm oil level threshold, and obtain the initial alarm triggering condition;

[0134] Through detailed comparison between historical alarm data and current oil level monitoring results, key oil level fluctuation points are identified. The process includes collecting oil level data and identifying abnormal frequency records therein. For example, alarm logs are analyzed to determine which oil level records frequently trigger alarms but are actually false alarms after verification. Past data is deeply mined and analyzed, and the data is classified and organized to determine the critical points of oil level fluctuations and the conditions for frequent false alarms. The data is used to adjust the alarm triggering conditions to reduce the false alarm rate, ensure the stability and reliability of the system, and obtain preliminary alarm triggering conditions, so as to reset the sensitive threshold of oil level monitoring and achieve more accurate oil level monitoring and alarm triggering.

[0135] S412: Combine the initial alarm triggering conditions with the real-time monitoring oil level data and use the formula:

[0136]

[0137] Calculate oil level alarm update threshold , output new oil level abnormal state monitoring parameters, where, Represents the average value of the oil level in the past. Indicates the minimum value of the oil level in the past. Represents the difference between the current real-time oil level and the historical average oil level;

[0138] Represents the average value of the oil level in the past, which is obtained by monitoring the oil level data in the past 30 days and calculating the average, which is 150cm. The minimum oil level in the past is 120 cm, which is also based on the data of the past 30 days. Represents the difference between the current real-time oil level and the historical average oil level. If the current oil level is 155cm, then . Substitute the values ​​into the formula to calculate:

[0139]

[0140]

[0141] The results show that, based on the comparison between the current real-time oil level and past data, the new alarm trigger threshold should be adjusted to trigger the alarm when the oil level reaches above 5.44 cm. The value reflects the sensitive adjustment of the alarm system to abnormal oil level, which can effectively monitor the real-time changes of the oil level and trigger the alarm when necessary.

[0142] S413: dynamically monitor the real-time oil level data in combination with the oil level abnormal state monitoring parameters, and if it is detected that the oil level exceeds the oil level alarm update threshold, record and generate a real-time alarm update record;

[0143] Based on the new oil level monitoring parameters, the system will monitor the oil level changes in the oil storage tank in real time. When the oil level exceeds the newly set alarm threshold, the system will immediately generate an alarm and record it in the system log. The process includes capturing and analyzing real-time data. By continuously monitoring the oil level data, once the oil level exceeds the threshold, the system will trigger an alarm and record detailed information of the event, such as time, oil level reading and alarm status. The data will be stored in the system log for subsequent data analysis and system maintenance, ensuring that the monitoring system can play a role at critical moments and promptly notify maintenance personnel to take corresponding measures, thereby ensuring the normal operation of the system and oil level safety.

[0144] An oil level status judgment system for an oil storage tank transformer, comprising:

[0145] The oil level signal adjustment module is installed and configured with an oil level sensor, collects oil level data at set detection points, performs real-time filtering, dynamically adjusts the adjustment factor in combination with current environmental noise data, optimizes the accuracy of the oil level signal, and records and outputs the adjusted oil level signal;

[0146] The frequency domain conversion evaluation module converts the oil level signal in the time domain into a frequency domain signal based on the adjusted oil level signal, evaluates the energy distribution of the frequency components in the signal, calculates the energy value of the key frequency components, evaluates the impact on the equipment performance, and obtains the frequency component analysis results;

[0147] The spectrum analysis and identification module performs frequency point screening based on the frequency component analysis results, combines the equipment's historical operation records and fault logs, determines the correlation between the frequency components and the equipment status, extracts key variables, and obtains the spectrum analysis results;

[0148] The frequency disturbance monitoring module monitors the changes in amplitude and phase in real time based on the spectrum analysis results, identifies abnormal frequency disturbances through preset thresholds, marks timestamps to track the actual time when the abnormality occurs, and generates frequency disturbance records;

[0149] The abnormal frequency tracking module tracks the frequency components in the record in real time based on the frequency disturbance record, monitors the changes of each frequency point, analyzes the changes of the oil level signal, distinguishes between normal fluctuations and frequency disturbances caused by oil level changes, summarizes and uploads them to the central database, and generates abnormal frequency records;

[0150] The alarm update and recording module analyzes the relationship between the current oil level data and the historical alarm records based on the abnormal frequency records, calculates the oil level alarm update threshold based on the real-time monitoring oil level data, and dynamically monitors the real-time oil level data. If it is detected that the oil level exceeds the oil level alarm update threshold, it records and generates a real-time alarm update record.

[0151] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A method for judging the oil level status of an oil storage tank transformer, characterized in that: The following steps are involved: The oil level signal is collected through the sensor, the collected oil level signal is filtered in real time, the filter coefficient is adjusted according to the environmental noise level, and the adjusted oil level signal is recorded and output; The oil level signal data after adjustment is used to convert the oil level signal from the time domain to the frequency domain through digital signal processing, and all frequency components in the spectrum are analyzed. According to the analysis results, each frequency component is calibrated to identify the target frequency components associated with the normal operation and potential fault state of the equipment, and the spectrum analysis results are obtained; Based on the spectrum analysis results, by dynamically monitoring the changes in the target frequency components, recording the frequency disturbances in the abnormal variation range, and recording and analyzing the abnormal frequency components in real time according to the recording results, the real-time changes in the oil level signal are revealed, and abnormal frequency records are generated; According to the abnormal frequency record, the alarm triggering condition is adjusted and updated to reflect the real-time oil level change of the oil storage cabinet transformer, dynamically update the abnormal oil level status monitoring, and generate real-time alarm update records; The steps for obtaining the real-time alarm update record are: Based on the abnormal frequency record, the relationship between the current oil level data and the historical alarm record is analyzed, the false alarm oil level threshold is located, and the initial alarm triggering condition is obtained; Combining the initial alarm triggering conditions with the real-time monitoring oil level data, the formula is adopted: Calculate oil level alarm update threshold , output new oil level abnormal state monitoring parameters, where, Represents the average value of the oil level in the past. Indicates the minimum value of the oil level in the past. Represents the difference between the current real-time oil level and the historical average oil level; In combination with the abnormal oil level status monitoring parameters, the real-time oil level data is dynamically monitored. If it is detected that the oil level exceeds the oil level alarm update threshold, a real-time alarm update record is recorded and generated.

2. The method for judging the oil level status of an oil storage tank transformer according to claim 1, characterized in that: The steps for obtaining the adjusted oil level signal are: Install and configure the oil level sensor, collect oil level data at the set detection point, output electrical signals through the sensor to indicate the oil level height, and obtain the original oil level signal; Based on the original oil level signal, real-time filtering is performed using the formula: After calculation, the oil level signal value after filtering is generated. ,in, Represents the original signal at the current moment, and Represent the historical lowest and highest original signal values ​​respectively, is the regulating factor, It is the filtering result of the previous cycle; According to the oil level signal after filtering, combined with the current environmental noise data, the adjustment factor is dynamically adjusted. , re-optimize the oil level signal accuracy, record and output the adjusted oil level signal.

3. The method for judging the oil level status of an oil storage tank transformer according to claim 2, characterized in that: The analysis steps of the frequency components are: Based on the adjusted oil level signal, signal preprocessing is started to convert the oil level signal in the time domain into a frequency domain signal by fast Fourier transform to obtain a preliminary frequency domain converted oil level signal; Analyzing the preliminary frequency domain converted oil level signal, evaluating the energy distribution of the frequency components in the signal, and generating power spectrum characteristic data; Based on the power spectrum characteristic data, the key frequency components are screened and the impact on the equipment performance is evaluated. The formula is used. Calculate the energy value of key frequency components , and get the frequency component analysis results, where Indicates frequency The power at Representative frequency The frequency domain signal at is the total number of samples.

4. The method for judging the oil level status of an oil storage tank transformer according to claim 3 is characterized in that: The steps for obtaining the spectrum analysis results are: Based on the frequency component analysis results, frequency point screening is performed to select frequency components whose signal energy is higher than a set threshold, mark them as key indicators for equipment performance monitoring, and generate a preliminary screening frequency data set; Using the preliminary screening frequency data set, combined with the historical operation records and fault logs of the equipment, performing data matching analysis, determining the correlation between the frequency components and the equipment status, and obtaining the equipment status associated frequency data; Perform statistical analysis on the device status association frequency data, extract key variables from the data, and use the formula: Calculate principal component analysis values , and get the spectrum analysis results, where represents the eigenvectors of the covariance matrix, Represents the frequency value in the frequency data set obtained by Fourier transform, represents the mean of the data, is a single frequency data point.

5. The method for judging the oil level status of the oil storage tank transformer according to claim 4 is characterized in that: The steps for recording the frequency disturbance within the abnormal variation range are as follows: Based on the spectrum analysis results, extract key frequency components, monitor changes in amplitude and phase in real time, and generate target frequency component change monitoring records; Analyze the target frequency component change monitoring record, identify abnormal frequency disturbances through a preset threshold, and generate abnormal frequency change range records; According to the abnormal frequency variation range record, it is systematically stored, and the actual time when the abnormality occurs is tracked with a timestamp to generate a frequency disturbance record.

6. The method for judging the oil level status of an oil storage tank transformer according to claim 5, characterized in that: The steps for obtaining the abnormal frequency record are: Based on the frequency disturbance record, the frequency components in the record are tracked in real time, spectrum dynamic analysis is performed, changes at each frequency point are monitored, each significant frequency disturbance is marked, the time series and amplitude of the frequency components are recorded, and a real-time frequency analysis record is generated; Analyze the change of the oil level signal by using the real-time frequency analysis record, distinguish normal fluctuation from the frequency disturbance caused by the oil level change, and generate the oil level signal change record; The oil level signal change records are summarized and uploaded to the central database, including the frequency disturbance, time tag and oil level reading corresponding to each oil level change, to generate abnormal frequency records.

7. An oil level status judgment system for an oil storage tank transformer, characterized in that: The system is used to execute the method for determining the oil level status of the oil conservator transformer according to any one of claims 1 to 6, comprising: The oil level signal adjustment module is installed and configured with an oil level sensor, collects oil level data at set detection points, performs real-time filtering, dynamically adjusts the adjustment factor in combination with current environmental noise data, optimizes the accuracy of the oil level signal, and records and outputs the adjusted oil level signal; The frequency domain conversion evaluation module converts the time domain oil level signal into a frequency domain signal based on the adjusted oil level signal, evaluates the energy distribution of the frequency components in the signal, calculates the energy value of the key frequency components, evaluates the impact on the equipment performance, and obtains the frequency component analysis result; The spectrum analysis and identification module performs frequency point screening based on the frequency component analysis results, combines the historical operation records and fault logs of the equipment, determines the correlation between the frequency components and the equipment status, extracts key variables, and obtains the spectrum analysis results; The frequency disturbance monitoring module monitors the changes of amplitude and phase in real time based on the spectrum analysis results, identifies abnormal frequency disturbances through preset thresholds, marks timestamps to track the actual time when the abnormality occurs, and generates frequency disturbance records; Based on the frequency disturbance record, the abnormal frequency tracking module tracks the frequency components in the record in real time, monitors the changes of each frequency point, analyzes the changes of the oil level signal, distinguishes normal fluctuations from frequency disturbances caused by oil level changes, summarizes and uploads them to the central database, and generates abnormal frequency records; The alarm update and recording module analyzes the relationship between the current oil level data and the historical alarm records based on the abnormal frequency records, calculates the oil level alarm update threshold in combination with the real-time monitoring oil level data, and dynamically monitors the real-time oil level data. If it is detected that the oil level exceeds the oil level alarm update threshold, the real-time alarm update record is recorded and generated.

Images