Power system fault monitoring method and system

Abstract

The invention relates to the technical field of fault monitoring, and discloses a power system fault monitoring method and system.The system comprises a data acquisition unit, a data preprocessing unit, a feature extraction unit, an abnormity judgment unit and a fault judgment unit.Temperature sensors, magnetic field sensors and vibration sensors are evenly distributed at power equipment and line acquisition points; multiple types of data are collected in real time in an omnibearing manner, the working condition of the system is accurately reflected, and one-sidedness and misjudgment of single-parameter monitoring are avoided; filtering is carried out according to specific rules for different data types, noise interference is removed, and data reliability is ensured; the change rate, the average value, the standard deviation and other characteristics of the temperature, the magnetic field and the vibration frequency are obtained according to a rigorous algorithm in the observation time period, and the operation state is deeply observed; characteristics are compared with a threshold value, each parameter is independently judged, an alarm is given in case of abnormity, and risks are captured in time; adjacent collection point data deep analysis, association and consistency index assistance are combined, and the misjudgment risk is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of fault monitoring, and in particular to a method and system for monitoring faults in an electric power system. Background Art

[0002] With the continuous increase in electricity demand and the continuous expansion of the power grid, the structure of the power system has become increasingly complex and the operating environment has become increasingly diverse.

[0003] Power equipment is in a high-load operation state for a long time. For example, transformers in substations have to bear huge power conversion tasks during peak hours of electricity consumption. Transmission lines need to transmit a large amount of electricity over long distances. The heating phenomenon of equipment and lines is becoming more frequent and serious. Overheating can easily cause insulation aging, short circuit and other faults. Real-time and accurate temperature monitoring has become an urgent need to ensure the safe operation of the power system. There are many complex electromagnetic interference sources in the surrounding environment of the power system, such as communication base stations, industrial electromagnetic equipment, etc. These external interferences will make the magnetic field environment of power equipment and lines unstable. Abnormal fluctuations in magnetic field strength often indicate that there may be hidden faults such as loose core and winding short circuit inside the equipment. The traditional method of relying only on manual inspection and observation of the appearance is difficult to detect such subtle but dangerous magnetic field changes. There is an urgent need for a technical means to monitor the magnetic field strength in real time and continuously. Furthermore, the vibration of power equipment during operation should not be ignored. For rotating power equipment such as generators and motors, the vibration amplitude and frequency may exceed the normal range due to mechanical wear, unbalanced force and other factors. Once the vibration intensifies, it will not only accelerate the damage of the equipment itself, but also cause the connection components to loosen, affecting the stability of the entire power system. The previous mode of detecting vibration problems through regular shutdown and maintenance cannot meet the requirements of uninterrupted power supply in modern power systems, and there is an urgent need to monitor the changes in vibration frequency online in real time.

[0004] In summary, in order to overcome the shortcomings of existing power system monitoring technology, meet the increasingly complex power system operation needs, and ensure the stability and reliability of power supply, there is an urgent need for a new method that can integrate multiple parameters and accurately monitor power system faults. Summary of the invention

[0005] The purpose of the present invention is to provide a method and system for monitoring faults in an electric power system, which solves the technical problems raised in the background technology.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] A method for monitoring faults in a power system comprises the following steps:

[0008] Step 1: Data collection:

[0009] The temperature data, magnetic field strength data and vibration frequency data are collected in real time by evenly distributing the temperature sensors, magnetic field sensors and vibration sensors installed at the corresponding collection points of the power equipment and lines;

[0010] Step 2: Data preprocessing:

[0011] The median filter algorithm is used to filter the collected temperature data, magnetic field intensity data and vibration frequency data;

[0012] Step 3: Feature extraction:

[0013] Feature extraction is performed on the filtered temperature values, total magnetic field intensity and vibration frequency values ​​at all time nodes within the specified observation period;

[0014] Step 4: Abnormal judgment:

[0015] Compare the feature extraction result with the corresponding preset feature threshold, and determine whether there is an abnormality at the relevant collection point based on the comparison result, and generate a relevant abnormal alarm signal based on the existing abnormality;

[0016] Step 5: Comprehensive fault determination:

[0017] Perform fault analysis based on the relevant abnormal alarm signals generated at the relevant collection points, and determine whether a fault signal is generated based on the fault analysis results.

[0018] As a further solution of the present invention: wherein the magnetic field strength data includes horizontal magnetic field strength, vertical magnetic field strength and total magnetic field strength.

[0019] As a further solution of the present invention: the filtering processing method is as follows:

[0020] For temperature data, among the temperature data obtained at multiple time nodes, a time node corresponding to the temperature data is selected, and then the temperature data of n time nodes before and after the time node are selected. Then, the data points are sorted in descending order, and then the middle value is taken from the sorting result as the filtered temperature value of the time node;

[0021] For the magnetic field intensity data, the horizontal magnetic field intensity and the vertical magnetic field intensity are filtered in the same way as the temperature data, and the filtered horizontal magnetic field intensity and vertical magnetic field intensity are obtained, which are then marked as Ha and Hb respectively;

[0022] Then pass: Calculate the total magnetic field strength H after filtering.

[0023] The filtering process of the vibration frequency data is the same as that of the temperature data, and the filtered vibration frequency value is obtained.

[0024] As a further solution of the present invention: the temperature value feature extraction method is as follows:

[0025] During the specified observation period, extract the filtered temperature values ​​of all time nodes and mark them as T j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all temperature values ​​in the observation period;

[0026] pass:

[0027] Calculate the rate of change of temperature during the observation period TB;

[0028] Among them, t0 is the time interval between adjacent time nodes;

[0029] Also through:

[0030] Calculate the average temperature TP during the observation period.

[0031] As a further solution of the present invention: the total magnetic field intensity feature extraction method is as follows:

[0032] In the specified observation period, the total magnetic field intensity after filtering at all time nodes is extracted and marked as H j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all total magnetic field intensities during the observation period;

[0033] pass:

[0034] Calculate the rate of change HB of the total magnetic field intensity during the observation period;

[0035] Among them, t0 is the time interval between adjacent time nodes;

[0036] Also through:

[0037] Calculate the standard deviation HC of the total magnetic field intensity during the observation period;

[0038] Where HP is the average value of the total magnetic field intensity during the observation period.

[0039] As a further solution of the present invention: the feature extraction method in the vibration frequency value is as follows:

[0040] During the specified observation period, the filtered vibration frequency values ​​of all time nodes are extracted and marked as G j, j = 1, 2, ... m, m represents the number of time nodes corresponding to all total vibration frequencies in the observation period;

[0041] pass:

[0042] Calculate the rate of change GB of the vibration frequency during the observation period;

[0043] Among them, t0 is the time interval between adjacent time nodes;

[0044] Also through:

[0045] Calculate the standard deviation GC of the vibration frequency during the observation period;

[0046] Where GP is the average value of the vibration frequency during the observation period.

[0047] As a further solution of the present invention: the comparison method in abnormality judgment is as follows:

[0048] The average temperature TP and the temperature change rate TB during the observation period are compared with the preset temperature threshold TPy and the temperature change rate threshold TBy respectively:

[0049] When TP≥TPy and TB≥TBy are both true, it is determined that there is a temperature anomaly at the collection point where the sensor is located, and a temperature anomaly alarm signal is generated;

[0050] The standard deviation HC of the total magnetic field intensity and the rate of change HB of the total magnetic field intensity during the observation period are compared with the preset total magnetic field intensity standard deviation threshold HPy and total magnetic field intensity change rate threshold HBy respectively:

[0051] When HP≥HPy and HB≥HBy are both true, it is determined that there is a magnetic field anomaly at the collection point where the sensor is located, and a magnetic field anomaly alarm signal is generated;

[0052] The standard deviation GC of the vibration frequency and the rate of change GB of the vibration frequency during the observation period are compared with the preset vibration frequency standard deviation threshold GPy and vibration frequency change rate threshold GBy respectively:

[0053] When GP≥GPy and GB≥GBy are both established, it is determined that there is an abnormal vibration frequency at the collection point where the sensor is located, and a vibration frequency abnormality alarm signal is generated.

[0054] As a further solution of the present invention: the fault analysis method is as follows:

[0055] During the specified observation period, when the temperature abnormality alarm signal, magnetic field abnormality alarm signal, and vibration frequency abnormality alarm signal are all generated at the same collection point, it is determined that the collection point has a fault, and a fault signal is generated accordingly;

[0056] When at least one of the temperature abnormality alarm signal, magnetic field abnormality alarm signal, and vibration frequency abnormality alarm signal is generated at the same collection point, and not all of them are generated, then the generated related abnormal alarm signal is obtained, and the corresponding data value is extracted based on the generated related abnormal alarm signal, and the data value at the collection point adjacent to the collection point is obtained at the same time, and then a secondary fault analysis is performed based on it, and based on the result of the secondary fault analysis, it is determined whether a fault signal is generated.

[0057] As a further solution of the present invention: the secondary fault analysis method is as follows:

[0058] Select the abnormal temperature alarm signal. When only the abnormal temperature alarm signal is generated at the same collection point, obtain the temperature values ​​of the collection point and its adjacent collection points during the observation period and mark them as T1 respectively. j and T2 j ;

[0059] Then, T1 is calculated according to the feature extraction steps. j The average value and rate of change are marked as TP1 and TB1, and T2 j The average value and rate of change are marked as TP2 and TB2 respectively;

[0060] Then pass:

[0061] Calculate the correlation index QU of the temperature values ​​of the collection point and its adjacent collection points during the observation period;

[0062] Also through:

[0063] Calculate the consistency index QS of the temperature change of the collection point and its adjacent collection points during the observation period;

[0064] The correlation index QU of the temperature value of the acquisition point and its adjacent acquisition points during the observation period and the consistency index QS of the temperature change are compared with the pre-set correlation index threshold QUy and consistency index threshold QSy respectively:

[0065] When QU≥QUy and QS≥QSy are established at the same time, it is determined that there is a fault in the power equipment and line where the collection point and its adjacent collection points are located, and a fault signal is generated; otherwise, no fault signal is generated.

[0066] A power system fault monitoring system, the system is used to implement a power system fault monitoring method, the system comprising:

[0067] A data acquisition unit is used to collect temperature data, magnetic field strength data and vibration frequency data in real time through temperature sensors, magnetic field sensors and vibration sensors evenly distributed and installed on corresponding acquisition points of power equipment and lines;

[0068] A data preprocessing unit, used to filter the collected temperature data, magnetic field intensity data and vibration frequency data using a median filtering algorithm;

[0069] A feature extraction unit is used to extract features from the filtered temperature values, total magnetic field strength and vibration frequency values ​​at all time nodes within a specified observation period;

[0070] The abnormality judgment unit is used to compare the feature extraction result with the corresponding preset feature threshold, and determine whether there is an abnormality at the relevant collection point based on the comparison result, and generate a relevant abnormality alarm signal based on the existing abnormality;

[0071] The fault determination unit is used to perform fault analysis based on the relevant abnormal alarm signals generated at the relevant collection points, and to determine whether a fault signal is generated based on the fault analysis results.

[0072] Beneficial effects of the present invention:

[0073] Multi-parameter monitoring with high accuracy: By evenly distributing and installing temperature sensors, magnetic field sensors and vibration sensors on the corresponding collection points of power equipment and lines, and collecting temperature data, magnetic field strength data and vibration frequency data at the same time, multi-dimensional real-time monitoring of the power system operation status is achieved. Compared with single parameter monitoring, it can more comprehensively and accurately reflect the actual working conditions of equipment and lines, and effectively avoid missed detection or misdiagnosis of faults caused by misjudgment of a single parameter.

[0074] Effective noise reduction, reliable data: The median filter algorithm is used to pre-process the collected data. Whether it is temperature, magnetic field strength or vibration frequency data, noise interference can be removed according to specific rules. For example, when filtering temperature data, multiple data before and after the time node are comprehensively considered, and the middle value is selected as the filtered result to ensure that the acquired data is true and reliable, laying a solid foundation for subsequent accurate fault judgment and reducing false alarms caused by data fluctuations.

[0075] Scientific feature extraction, deep insight into operating status: During the specified observation period, rigorous and scientific feature extraction methods are designed for different types of data. For temperature, the calculation of the rate of change and the average value can clearly provide insight into the dynamic trend and overall level of temperature; for magnetic field strength, the extraction of standard deviation and rate of change can accurately reflect the stability and changes of the magnetic field; the rate of change and standard deviation of vibration frequency can also fully display its operating characteristics. These characteristic parameters provide rich and valuable information for fault judgment, making the identification of potential problems in the power system more in-depth and accurate.

[0076] Layered abnormality judgment and timely warning: A detailed abnormality judgment system has been built to compare the results of various feature extractions with the corresponding preset thresholds. Based on the respective threshold standards of temperature, magnetic field, and vibration frequency, it is independently judged whether there is an abnormality. Once an abnormality is found, a corresponding alarm signal is immediately generated. This layered judgment mechanism can capture abnormal fluctuations of a single parameter in a timely manner, allowing operation and maintenance personnel to know potential risk points at the first time, and buy precious time for rapid response.

[0077] Comprehensive fault judgment reduces the risk of misjudgment: In the comprehensive fault judgment link, the combination of multiple alarm signals at the same collection point is fully considered. When all abnormal alarm signals are generated, it is decisively determined that there is a fault; when only some signals are generated, further in-depth analysis is performed in combination with the data of adjacent collection points, and the correlation index and consistency index are calculated and compared with the threshold. Through this comprehensive and rigorous comprehensive judgment method, the probability of fault misjudgment is greatly reduced, the accuracy and reliability of fault judgment are ensured, and the safe and stable operation of the power system is guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] The present invention will be further described below in conjunction with the accompanying drawings.

[0079] Figure 1 The present invention is a system block diagram of a power system fault monitoring system.

[0080] Figure 2 It is a flow chart of a method for monitoring faults in a power system according to the present invention. DETAILED DESCRIPTION

[0081] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0082] Embodiment 1

[0083] See also Figure 1and Figure 2 As shown, the present invention is a method for monitoring faults in a power system, which performs fault monitoring based on temperature data, and the method comprises the following steps:

[0084] Step 1: Data collection:

[0085] Real-time temperature data collection is achieved through temperature sensors evenly distributed and installed on the corresponding collection points of power equipment and lines;

[0086] Step 2: Data preprocessing:

[0087] The median filter algorithm is used to filter the collected temperature data to remove noise interference. The method is as follows:

[0088] Among the temperature data obtained at multiple time nodes, a time node corresponding to the temperature data is selected, and then the temperature data of n time nodes before and after the time node are selected, and then the data points are sorted in descending order, and then the middle value is taken from the sorting result as the filtered temperature value of the time node;

[0089] Step 3: Feature extraction:

[0090] During the specified observation period, extract the filtered temperature values ​​of all time nodes and mark them as T j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all temperature values ​​in the observation period;

[0091] pass:

[0092] Calculate the rate of change of temperature during the observation period TB;

[0093] Among them, t0 is the time interval between adjacent time nodes;

[0094] Also through:

[0095] Calculate the average temperature TP during the observation period;

[0096] Step 4: Abnormal judgment:

[0097] The average temperature TP and the temperature change rate TB during the observation period are compared with the preset temperature threshold TPy and the temperature change rate threshold TBy respectively:

[0098] When TP≥TPy and TB≥TBy are both true, it is determined that there is a temperature anomaly at the collection point where the sensor is located, and a temperature anomaly alarm signal is generated;

[0099] This embodiment collects temperature data in real time by evenly installing temperature sensors at the corresponding collection points of power equipment and lines, which can more intuitively reflect the temperature conditions of the equipment during operation. The collected data is preprocessed using the median filter algorithm to effectively remove noise interference, making the subsequent feature extraction and abnormality judgment based on the processed data more accurate and reliable, and can detect temperature anomalies in a timely manner and generate alarm signals, which helps to detect potential power system failures caused by temperature factors as early as possible and ensure the safe and stable operation of the power system.

[0100] Embodiment 2

[0101] See also Figure 1 and Figure 2 As shown, as the second embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the technical solution of the present embodiment is different from that of the first embodiment only in that the present embodiment performs fault monitoring based on magnetic field data, and the method includes the following steps:

[0102] Step 1: Data collection:

[0103] The magnetic field intensity data is collected in real time by evenly distributing magnetic field sensors installed at the corresponding collection points of power equipment and lines. The magnetic field intensity data includes horizontal magnetic field intensity, vertical magnetic field intensity and total magnetic field intensity.

[0104] Step 2: Data preprocessing:

[0105] According to the filtering processing method of the temperature data, the horizontal magnetic field intensity and the vertical magnetic field intensity in the magnetic field intensity data are filtered, and the filtered horizontal magnetic field intensity and vertical magnetic field intensity are obtained, which are then marked as Ha and Hb respectively;

[0106] Then pass:

[0107] Calculate the total magnetic field strength H after filtering.

[0108] Step 3: Feature extraction:

[0109] In the specified observation period, the total magnetic field intensity after filtering at all time nodes is extracted and marked as H j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all total magnetic field intensities during the observation period;

[0110] pass:

[0111] Calculate the rate of change HB of the total magnetic field intensity during the observation period;

[0112] Among them, t0 is the time interval between adjacent time nodes;

[0113] Also through:

[0114] Calculate the standard deviation HC of the total magnetic field intensity during the observation period;

[0115] Where HP is the average value of the total magnetic field intensity during the observation period;

[0116] Step 4: Abnormal judgment:

[0117] The standard deviation HC of the total magnetic field intensity and the rate of change HB of the total magnetic field intensity during the observation period are compared with the preset total magnetic field intensity standard deviation threshold HPy and total magnetic field intensity change rate threshold HBy respectively:

[0118] When HP≥HPy and HB≥HBy are both true, it is determined that there is a magnetic field anomaly at the collection point where the sensor is located, and a magnetic field anomaly alarm signal is generated;

[0119] This embodiment uses evenly distributed magnetic field sensors to collect magnetic field strength data in real time, broadening the monitoring dimension. The relevant magnetic field strength data is processed according to the filtering processing method of temperature data, and through reasonable feature extraction and abnormality judgment steps, the magnetic field abnormality can be accurately judged and a magnetic field abnormality alarm signal can be generated. This can play an effective early warning role for power equipment failures that may be caused by magnetic field changes, and further improve the comprehensiveness of power system fault monitoring.

[0120] Embodiment 3

[0121] See also Figure 1 and Figure 2 As shown, as the third embodiment of the present invention, when the present application is implemented, compared with the first and second embodiments, the technical solution of this embodiment is to combine the solutions of the first and second embodiments, and the difference between the technical solution of this embodiment and the first and second embodiments is that: in this embodiment, fault monitoring is performed based on vibration frequency, and the method includes the following steps:

[0122] Step 1: Data collection:

[0123] Vibration frequency data can be collected in real time by evenly distributing vibration sensors installed on corresponding collection points of power equipment and lines;

[0124] Step 2: Data preprocessing:

[0125] According to the filtering processing method of temperature data, the vibration frequency data is filtered and the filtered vibration frequency value is obtained;

[0126] Step 3: Feature extraction:

[0127] During the specified observation period, the filtered vibration frequency values ​​of all time nodes are extracted and marked as G j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all total vibration frequencies in the observation period;

[0128] pass:

[0129] Calculate the rate of change GB of the vibration frequency during the observation period;

[0130] Among them, t0 is the time interval between adjacent time nodes;

[0131] Also through:

[0132] Calculate the standard deviation GC of the vibration frequency during the observation period;

[0133] Where GP is the average value of the vibration frequency during the observation period;

[0134] Step 4: Abnormal judgment:

[0135] The standard deviation GC of the vibration frequency and the rate of change GB of the vibration frequency during the observation period are compared with the preset vibration frequency standard deviation threshold GPy and vibration frequency change rate threshold GBy respectively:

[0136] When GP≥GPy and GB≥GBy are both true, it is determined that there is an abnormal vibration frequency at the collection point where the sensor is located, and a vibration frequency abnormality alarm signal is generated;

[0137] This embodiment collects vibration frequency data in real time by installing vibration sensors at the collection points, and then performs filtering, feature extraction, and abnormality judgment operations. It can timely detect abnormal conditions in vibration frequency and generate alarm signals. In view of the fact that during the operation of power equipment, abnormal vibration frequency often indicates problems in mechanical structure and other aspects, this embodiment can effectively monitor power system faults caused by vibration-related factors, supplement the content of fault monitoring from another perspective, and enhance the effectiveness of overall monitoring.

[0138] Embodiment 4

[0139] See also Figure 1 and Figure 2 As shown, as the fourth embodiment of the present invention, when the present application is implemented, compared with the first embodiment, the second embodiment and the third embodiment, the difference between the present embodiment and the first embodiment, the second embodiment and the third embodiment is that the present embodiment also performs comprehensive fault determination on the basis of the first embodiment to the third embodiment, and the method is as follows:

[0140] During the specified observation period, when the temperature abnormality alarm signal, magnetic field abnormality alarm signal, and vibration frequency abnormality alarm signal are all generated at the same collection point, it is determined that the collection point has a fault, and a fault signal is generated accordingly;

[0141] When at least one of the abnormal temperature alarm signal, abnormal magnetic field alarm signal, and abnormal vibration frequency alarm signal is generated at the same collection point, but not all of them are generated;

[0142] When the generated related abnormal alarm signal is obtained, the corresponding data value is extracted according to the generated related abnormal alarm signal, and the data characteristic value at the collection point adjacent to the collection point is obtained at the same time;

[0143] Taking the temperature abnormality alarm signal as an example, when only the temperature abnormality alarm signal is generated at the same collection point, the temperature values ​​of the collection point and its adjacent collection points during the observation period are obtained and marked as T1 respectively. j and T2 j ;

[0144] Then, T1 is calculated according to the feature extraction steps. j The average value and rate of change are marked as TP1 and TB1, and T2 j The average value and rate of change are marked as TP2 and TB2 respectively;

[0145] Then pass:

[0146] Calculate the correlation index QU of the temperature values ​​of the collection point and its adjacent collection points during the observation period;

[0147] Also through:

[0148] Calculate the consistency index QS of the temperature change of the collection point and its adjacent collection points during the observation period;

[0149] The correlation index QU of the temperature value of the acquisition point and its adjacent acquisition points during the observation period and the consistency index QS of the temperature change are compared with the pre-set correlation index threshold QUy and consistency index threshold QSy respectively:

[0150] When QU≥QUy and QS≥QSy are both true, it is determined that there is a fault in the power equipment and line where the collection point and its adjacent collection points are located, and a fault signal is generated;

[0151] This embodiment implements comprehensive fault determination on the basis of fault monitoring based on temperature, magnetic field, and vibration frequency and generating corresponding abnormal alarm signals in the previous embodiment. Not only can it directly determine that there is a fault when all abnormal alarm signals of temperature, magnetic field, and vibration frequency are generated at the same acquisition point, but also, for the case where only some abnormal alarm signals are generated, by obtaining the corresponding data values ​​and the data feature values ​​of adjacent acquisition points, further calculating the correlation index and consistency index, and comparing them with the threshold to determine the fault, this multi-level and comprehensive determination method makes fault monitoring more accurate and comprehensive, greatly reduces the possibility of misjudgment and missed judgment, and is of great significance to the troubleshooting and prevention of power system faults.

[0152] Embodiment 5

[0153] See also Figure 1 and Figure 2 As shown, as the fifth embodiment of the present invention, when the present application is implemented specifically, compared with the first, second, third and fourth embodiments, the technical solution of this embodiment is to combine the solutions of the above-mentioned first, second, third and fourth embodiments for implementation.

[0154] This embodiment combines and implements the solutions of the previous embodiments, fully integrating the advantages of various methods such as fault monitoring based on temperature, magnetic field, vibration frequency and comprehensive fault judgment, forming a more complete, comprehensive and efficient power system fault monitoring system, maximizing the advantages of various monitoring methods, comprehensively ensuring the safe and reliable operation of the power system, and improving the accuracy and reliability of fault monitoring of the entire power system.

[0155] See also Figure 1 and Figure 2 As shown, the present invention also provides a power system fault monitoring system, which is used to implement a power system fault monitoring method, and the system includes:

[0156] A data acquisition unit is used to collect temperature data, magnetic field strength data and vibration frequency data in real time through temperature sensors, magnetic field sensors and vibration sensors evenly distributed and installed on corresponding acquisition points of power equipment and lines;

[0157] A data preprocessing unit, used to filter the collected temperature data, magnetic field intensity data and vibration frequency data using a median filtering algorithm;

[0158] A feature extraction unit is used to extract features from the filtered temperature values, total magnetic field strength and vibration frequency values ​​at all time nodes within a specified observation period;

[0159] The abnormality judgment unit is used to compare the feature extraction result with the corresponding preset feature threshold, and determine whether there is an abnormality at the relevant collection point based on the comparison result, and generate a relevant abnormality alarm signal based on the existing abnormality;

[0160] The fault determination unit is used to perform fault analysis based on the relevant abnormal alarm signals generated at the relevant collection points, and to determine whether a fault signal is generated based on the fault analysis results.

[0161] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters and thresholds in the formula are set by technicians in this field according to actual conditions.

[0162] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for monitoring faults in a power system, characterized in that: The following steps are involved: Step 1: Data collection: The temperature data, magnetic field strength data and vibration frequency data are collected in real time by evenly distributing the temperature sensors, magnetic field sensors and vibration sensors installed at the corresponding collection points of the power equipment and lines; Step 2: Data preprocessing: The median filter algorithm is used to filter the collected temperature data, magnetic field intensity data and vibration frequency data; Step 3: Feature extraction: Extract features from the filtered temperature values, total magnetic field intensity and vibration frequency values ​​at all time nodes within the specified observation period; Step 4: Abnormal judgment: Compare the feature extraction result with the corresponding preset feature threshold, and determine whether there is an abnormality at the relevant collection point based on the comparison result, and generate a relevant abnormal alarm signal based on the existing abnormality; Step 5: Comprehensive fault determination: Perform fault analysis based on the relevant abnormal alarm signals generated at the relevant collection points, and determine whether a fault signal is generated based on the fault analysis results.

2. A method for monitoring power system faults according to claim 1, characterized in that: The filtering process is as follows: For temperature data, among the temperature data obtained at multiple time nodes, a time node corresponding to the temperature data is selected, and then the temperature data of n time nodes before and after the time node are selected. Then, the data points are sorted in descending order, and then the middle value is taken from the sorting result as the filtered temperature value of the time node; For magnetic field intensity data, the magnetic field intensity data includes horizontal magnetic field intensity, vertical magnetic field intensity and total magnetic field intensity, which is to filter the horizontal magnetic field intensity and vertical magnetic field intensity, and the filtering method is the same as that of temperature data, and the filtered horizontal magnetic field intensity and vertical magnetic field intensity are obtained, which are then marked as Ha and Hb respectively; Then pass: Calculate the total magnetic field strength H after filtering; The filtering process of the vibration frequency data is the same as that of the temperature data, and the filtered vibration frequency value is obtained.

3. A method for monitoring power system faults according to claim 2, characterized in that: The temperature value feature extraction method is as follows: During the specified observation period, extract the filtered temperature values ​​of all time nodes and mark them as T j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all temperature values ​​in the observation period; pass: Calculate the rate of change of temperature during the observation period TB; Among them, t0 is the time interval between adjacent time nodes; Also through: Calculate the average temperature TP during the observation period.

4. A method for monitoring power system faults according to claim 3, characterized in that: The total magnetic field strength feature extraction method is as follows: In the specified observation period, the total magnetic field intensity after filtering at all time nodes is extracted and marked as H j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all total magnetic field intensities during the observation period; pass: Calculate the rate of change HB of the total magnetic field intensity during the observation period; Among them, t0 is the time interval between adjacent time nodes; Also through: Calculate the standard deviation HC of the total magnetic field intensity during the observation period; Where HP is the average value of the total magnetic field intensity during the observation period.

5. A method for monitoring power system faults according to claim 4, characterized in that: The vibration frequency feature extraction method is as follows: During the specified observation period, the filtered vibration frequency values ​​of all time nodes are extracted and marked as G j , j = 1, 2, ... m, m represents the number of time nodes corresponding to all total vibration frequencies in the observation period; pass: Calculate the rate of change GB of the vibration frequency during the observation period; Among them, t0 is the time interval between adjacent time nodes; Also through: Calculate the standard deviation GC of the vibration frequency during the observation period; Where GP is the average value of the vibration frequency during the observation period.

6. A method for monitoring power system faults according to claim 5, characterized in that: The comparison method in abnormal judgment is as follows: The average temperature TP and the temperature change rate TB during the observation period are compared with the preset temperature threshold TPy and the temperature change rate threshold TBy respectively: When TP≥TPy and TB≥TBy are both true, it is determined that there is a temperature anomaly at the collection point where the sensor is located, and a temperature anomaly alarm signal is generated; The standard deviation HC of the total magnetic field intensity and the rate of change HB of the total magnetic field intensity during the observation period are compared with the preset total magnetic field intensity standard deviation threshold HPy and total magnetic field intensity change rate threshold HBy respectively: When HP≥HPy and HB≥HBy are both true, it is determined that there is a magnetic field anomaly at the collection point where the sensor is located, and a magnetic field anomaly alarm signal is generated; The standard deviation GC of the vibration frequency and the rate of change GB of the vibration frequency during the observation period are compared with the preset vibration frequency standard deviation threshold GPy and vibration frequency change rate threshold GBy respectively: When GP≥GPy and GB≥GBy are both established, it is determined that there is an abnormal vibration frequency at the collection point where the sensor is located, and a vibration frequency abnormality alarm signal is generated.

7. A method for monitoring faults in a power system according to claim 6, characterized in that: The fault analysis method is as follows: within the specified observation period, when the temperature abnormality alarm signal, magnetic field abnormality alarm signal, and vibration frequency abnormality alarm signal are all generated at the same collection point, it is determined that the collection point has a fault, and a fault signal is generated accordingly; When at least one of the temperature abnormality alarm signal, magnetic field abnormality alarm signal, and vibration frequency abnormality alarm signal is generated at the same collection point, and not all of them are generated, then the generated related abnormal alarm signal is obtained, and the corresponding data value is extracted based on the generated related abnormal alarm signal, and the data value at the collection point adjacent to the collection point is obtained at the same time, and then a secondary fault analysis is performed based on it, and based on the result of the secondary fault analysis, it is determined whether a fault signal is generated.

8. A method for monitoring power system faults according to claim 7, characterized in that: The secondary fault analysis method is as follows: Select the abnormal temperature alarm signal. When only the abnormal temperature alarm signal is generated at the same collection point, obtain the temperature values ​​of the collection point and its adjacent collection points during the observation period and mark them as T1 respectively. j and T2 j ; Then, T1 is calculated according to the feature extraction steps. j The average value and rate of change are marked as TP1 and TB1, and T2 j The average value and rate of change are marked as TP2 and TB2 respectively; Then pass: Calculate the correlation index QU of the temperature values ​​of the collection point and its adjacent collection points during the observation period; Also through: Calculate the consistency index QS of the temperature change of the collection point and its adjacent collection points during the observation period; The correlation index QU of the temperature value and the consistency index QS of the temperature change of the acquisition point and its adjacent acquisition points during the observation period are compared with the preset correlation index threshold QUy and consistency index threshold QSy respectively, and based on the comparison result, it is determined whether to generate a fault signal.

9. A method for monitoring power system faults according to claim 8, characterized in that: When QU≥QUy and QS≥QSy are established at the same time, it is determined that there is a fault in the power equipment and line where the collection point and its adjacent collection points are located, and a fault signal is generated; otherwise, no fault signal is generated.

10. A power system fault monitoring system, the system being used to implement a power system fault monitoring method according to any one of claims 1 to 9, characterized in that: include: A data acquisition unit is used to collect temperature data, magnetic field strength data and vibration frequency data in real time through temperature sensors, magnetic field sensors and vibration sensors evenly distributed and installed on corresponding acquisition points of power equipment and lines; A data preprocessing unit, used to filter the collected temperature data, magnetic field intensity data and vibration frequency data using a median filtering algorithm; A feature extraction unit is used to extract features from the filtered temperature values, total magnetic field strength and vibration frequency values ​​at all time nodes within a specified observation period; The abnormality judgment unit is used to compare the feature extraction result with the corresponding preset feature threshold, and determine whether there is an abnormality at the relevant collection point based on the comparison result, and generate a relevant abnormality alarm signal based on the existing abnormality; The fault determination unit is used to perform fault analysis based on the relevant abnormal alarm signals generated at the relevant collection points, and to determine whether a fault signal is generated based on the fault analysis results.

Images