A method and system for real-time dynamic monitoring of power equipment systems based on the Internet of Things

Abstract

This application relates to the field of electrical variable measurement technology, specifically to a method and system for real-time dynamic monitoring of power equipment systems based on the Internet of Things (IoT). The method includes: acquiring different data types; constructing vibration sequences from the vibration signals and obtaining fitting curve functions for all vibration sequences; determining a trend sequence based on the coefficient of variation of the vibration sequences; combining the trend sequence and the vibration sequences with the fitting curve function to determine the vibration trend coefficient; clustering the periods based on the vibration trend coefficient, obtaining local and overall change trends based on the cluster data and all data respectively, and obtaining the equipment disturbance coefficient based on both; determining a prediction function based on the equipment disturbance coefficient, and obtaining a corrected vibration signal by combining the difference in vibration trend coefficients between adjacent periods, thus completing dynamic monitoring. This application enhances the accuracy of monitoring power equipment faults.

Description

Technical Field

[0001] This application relates to the field of electrical variable measurement technology, specifically to a method and system for real-time dynamic monitoring of power equipment systems based on the Internet of Things. Background Technology

[0002] With the rapid development of smart grids and the energy internet, power equipment, as a critical infrastructure of the power system, directly impacts the safety, stability, and reliability of the power grid and power supply. In recent years, the Internet of Things (IoT) technology has deeply empowered the field of power equipment monitoring. By deploying various types of sensors to collect operating parameters such as temperature, vibration, and current in real time, and combining edge computing and cloud-based intelligent analysis, it has driven the transformation and upgrading of power equipment monitoring from "periodic maintenance" to "real-time dynamic prediction." Analyzing sensor monitoring data helps determine whether there are any anomalies in power equipment; therefore, the accuracy of sensor data analysis and mining is crucial to ensuring the quality of dynamic monitoring of power equipment.

[0003] However, existing methods primarily utilize sensor data by setting thresholds or performing simple trend analysis. This involves fault detection based on the relationship between sensor data and set thresholds, or fault prediction for power equipment based on changes in historical sensor data. These methods are highly dependent on the accuracy of the sensor data. In actual sensor monitoring, factors such as environmental changes, aging, or power fluctuations can cause significant sensor drift. This inherent error in the analysis results leads to inherent inaccuracies in the monitoring of power equipment. Therefore, a more accurate IoT-based dynamic monitoring method for power equipment is urgently needed. Summary of the Invention

[0004] To address the technical problem of low monitoring accuracy, this application provides a real-time dynamic monitoring method and system for power equipment systems based on the Internet of Things (IoT). The specific technical solution adopted is as follows:

[0005] Firstly, this application proposes a real-time dynamic monitoring method for power equipment systems based on the Internet of Things (IoT), which includes the following steps:

[0006] Acquire vibration signals, voltage data, current data, and equipment temperature over a time interval for each monitoring point in each cycle;

[0007] Vibration signals within a cycle are used to form a vibration sequence. The vibration sequence is then converted into a spectrum and fitted to obtain a fitting curve function. For each monitoring point, the coefficients of variation of the vibration sequences in all previous cycles are sorted to obtain the trend sequence of the cycle. The correlation between vibration sequences and the correlation between trend sequences among monitoring points in the same cycle are recorded as the horizontal correlation measure and the vertical correlation measure, respectively. The amplitude change trend is determined by the fitted curve function. The vibration trend coefficient is determined by the horizontal and vertical correlation measures among all monitoring points in each cycle, as well as the amplitude, fluctuation, and amplitude change trend of the vibration sequences of all monitoring points in each cycle.

[0008] The period is clustered based on the vibration trend coefficient; voltage data, current data, and equipment temperature are denoted as data types; for each data type, the changes in data within the cluster and the changes in all data are denoted as local change trend and overall change trend, respectively; the equipment disturbance coefficient for each period is obtained based on the local change trend and overall change trend.

[0009] The equipment disturbance coefficient is fitted to determine the fluctuation prediction function. Based on the fluctuation prediction function and the difference in vibration trend coefficient between adjacent periods, the vibration signal is corrected using an optimization algorithm to obtain the corrected vibration signal, and dynamic monitoring is completed.

[0010] In the aforementioned scheme, this application first calculates a vibration trend coefficient based on changes in vibration signals during power equipment monitoring. This index quantifies the trend of vibration data changes from both horizontal and vertical perspectives, enhancing the accuracy of measuring power equipment performance changes. Then, combining the electrical parameters and temperature of the power equipment, it calculates an equipment disturbance coefficient. This index quantifies the global parameter changes of the power equipment, improving the distinction between fault-related and non-fault-related factors affecting the monitoring parameters and the measurement of disturbance levels, thus contributing to enhanced accuracy in detecting power equipment faults. By considering the horizontal and vertical changes in vibration data and quantifying the impact of fault-related factors based on global parameter changes, this method enhances the removal of the influence of non-fault-related factors, correcting the vibration data. The corrected vibration data more accurately reflects the fault status of the power equipment, thereby realizing a more accurate IoT-based dynamic monitoring method for power equipment and enhancing the accuracy of power equipment fault monitoring.

[0011] In one embodiment, the method for obtaining the fitting curve function by converting the vibration sequence into a spectrum is as follows:

[0012] The vibration sequence of each monitoring point in each cycle is used as input, and the corresponding spectrum is obtained by Fourier transform. In the spectrum, the frequency is used as the horizontal axis and the corresponding amplitude is used as the vertical axis. Polynomial fitting is performed to output the fitted curve function.

[0013] In one embodiment, the vibration trend coefficient is positively correlated with both the horizontal and vertical correlation measures, and negatively correlated with the amplitude, fluctuation, and amplitude change trend of the vibration sequence.

[0014] In one embodiment, the amplitude of the vibration sequence is the mean of all peaks in the vibration sequence, the fluctuation of the vibration sequence is the variance of all amplitudes in the vibration sequence, and the amplitude change trend of the vibration sequence is the proportion of data points with positive derivatives in the fitted curve function.

[0015] In one embodiment, the device temperature for each cycle is the device temperature within the time interval of that cycle.

[0016] In one embodiment, the method of recording changes in data within a cluster and changes in all data as local and overall trends, respectively, is as follows:

[0017] The periods of each data type in each cluster are sorted in chronological order. The data values ​​are used as the x-axis and the corresponding time series are used as the y-axis. Polynomial fitting is performed to obtain the polynomial function of each data type, which is denoted as the first polynomial function.

[0018] Sort all periods of each data type in chronological order, use the data values ​​as the x-axis and the corresponding time series as the y-axis, and perform polynomial fitting to obtain the polynomial function of each data type, which is denoted as the second polynomial function.

[0019] Integrating the first and second polynomial functions respectively yields the overall trend and the local trend.

[0020] In one embodiment, the method of integrating the first polynomial function and the second polynomial function respectively to obtain the overall trend and the local trend is as follows:

[0021] The overall trend is the area of ​​the integral of the second polynomial function;

[0022] The expression for the local trend of change is:

[0023] , This represents the sum of the vibration trend coefficients in the j-th cluster. This represents the sum of all vibration trend coefficients. Let the integral area of ​​the first polynomial function in the j-th cluster be the area of ​​the m-th data type. This represents the local change trend of the m-th data type in the i-th period.

[0024] In one embodiment, the device disturbance coefficient is positively correlated with the difference between the overall change trend and the local change trend.

[0025] In one embodiment, the objective function of the optimization algorithm is:

[0026] , Let v be the vibration trend coefficient of the x-th cycle. This represents the vibration trend coefficient for the (x-1)th period. This represents the predicted fluctuation value for the x-th period. This represents the target value in the x-th period and the v-th iteration.

[0027] Secondly, embodiments of this application also provide a real-time dynamic monitoring system for power equipment systems based on the Internet of Things (IoT), including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any of the above-described methods for real-time dynamic monitoring of power equipment systems based on the IoT.

[0028] The beneficial effects of this application are as follows:

[0029] This application first calculates a vibration trend coefficient based on changes in vibration signals during power equipment monitoring. This index quantifies the trend of vibration data changes from both horizontal and vertical perspectives, enhancing the accuracy of measuring performance changes in power equipment. Then, combining the electrical parameters and temperature of the power equipment, it calculates an equipment disturbance coefficient. This index quantifies the global parameter changes of the power equipment, improving the distinction between fault-related and non-fault-related factors affecting the monitoring parameters and the measurement of disturbance levels, thus contributing to enhanced accuracy in detecting power equipment faults. By considering the horizontal and vertical changes in vibration data and quantifying the impact of fault-related factors based on global parameter changes, this method enhances the removal of the influence of non-fault-related factors, correcting the vibration data. The corrected vibration data more accurately reflects the fault status of the power equipment, thereby realizing a more accurate IoT-based dynamic monitoring method for power equipment and enhancing the accuracy of power equipment fault monitoring. Attached Figure Description

[0030] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0031] Figure 1 This is a flowchart illustrating a real-time dynamic monitoring method for an IoT-based power equipment system, provided as an embodiment of this application. Detailed Implementation

[0032] To further illustrate the technical means and effects adopted by this application to achieve the intended inventive purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a real-time dynamic monitoring method and system for power equipment systems based on the Internet of Things (IoT) proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0033] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0034] A method and system embodiment for real-time dynamic monitoring of power equipment systems based on the Internet of Things:

[0035] The following description, in conjunction with the accompanying drawings, details a specific scheme for a real-time dynamic monitoring method and system for power equipment systems based on the Internet of Things (IoT) provided in this application.

[0036] Please see Figure 1 The diagram illustrates a flowchart of a real-time dynamic monitoring method for an IoT-based power equipment system according to an embodiment of this application. The method includes the following steps:

[0037] Step S001: Collect vibration signals, voltage data, current data, and equipment temperature.

[0038] Install on the front, bottom, and outer surface of the high-voltage side of the power equipment At each monitoring point, a piezoelectric accelerometer is used to collect vibration signals from the power equipment. In this embodiment, the vibration signal acquisition frequency is 20kHz. Then, voltage and current data are collected at the input and output terminals of the power equipment using voltage and current transformers, respectively. In this embodiment, the acquisition frequency is 10kHz. The acquisition period for the vibration signal, voltage data, and current data is 1 second. Thermocouple sensors are used at key components of the power equipment to collect the equipment temperature. The equipment temperature is collected every preset time interval, and the time interval between two adjacent temperature acquisitions is obtained. The average of the two acquired temperatures is taken as the equipment temperature for the time interval. In this embodiment, the time interval is 30 seconds.

[0039] In the IoT sensing layer, monitoring data of power equipment is collected through various sensors in the manner described above. Then, in the network layer, the collected data is transmitted to the application layer through the power IoT communication network. The application layer analyzes this monitoring data to achieve dynamic monitoring of power equipment.

[0040] At this point, vibration signals, voltage data, current data, and equipment temperature over a time interval were collected for each monitoring point for each cycle.

[0041] Step S002: Construct vibration sequences from the vibration signals and obtain fitting curve functions for all vibration sequences; determine trend sequences based on the coefficient of variation of the vibration sequences; combine the trend sequences and vibration sequences with the fitting curve functions to determine the vibration trend coefficient.

[0042] Various types of power equipment play a crucial role in ensuring the stable operation and monitoring and protection of the power system. Therefore, monitoring power equipment is of paramount importance to the power system. The development of Internet of Things (IoT) technology has enhanced the real-time monitoring capabilities of power equipment, enabling comprehensive monitoring of its operating status through multiple sensors at the sensing layer. In analyzing this sensor monitoring data, current technologies largely rely on the accuracy of the sensor data. However, in practical applications, the complexity of the environment in which power equipment operates inevitably affects the detection accuracy of sensors, leading to misjudgments or missed diagnoses of power equipment faults. Therefore, it is necessary to enhance the ability to analyze and utilize sensor data, further eliminate errors, and improve the accuracy of power equipment monitoring.

[0043] As highly sensitive mechanical devices, electrical equipment is susceptible to changes in its operating conditions or performance. These changes can alter the equipment's response speed, motion patterns, or electromagnetic field distribution, all of which affect its vibration. Under normal circumstances, due to the relatively good performance of electrical equipment, its vibration amplitude and fluctuations are small, with larger amplitudes near the fundamental frequency and smaller or negligible amplitudes for high-frequency components. Furthermore, the vibration changes are primarily influenced by operating conditions, which affect the overall fluctuations across different monitoring points on the equipment. Therefore, there should be a high correlation between the vibration patterns and changes at each monitoring point. When the performance of electrical equipment deteriorates, it affects the equipment's motion, leading to increased vibration fluctuations during operation. This performance degradation also increases the energy of high-frequency components in the vibration signal. Moreover, since performance degradation is often due to the deterioration of certain components, the vibration signals near these components are significantly affected. The farther away from these components, the smaller the impact. The more affected the monitoring point, the greater the difference in vibration signal changes compared to other points.

[0044] For each monitoring point, the vibration signal of each cycle is used to form a vibration sequence. The vibration sequence of each monitoring point is used as input, and a Fourier transform is employed to obtain the corresponding spectrum. In the spectrum, frequency is plotted on the x-axis, and the corresponding amplitude is plotted on the y-axis. A polynomial fitting is then performed, and the fitted curve function is output. Fourier transform and polynomial fitting are well-known techniques and will not be elaborated further.

[0045] Calculate the coefficient of variation of the vibration sequence at each monitoring point for each cycle; for each monitoring point, sort the coefficients of variation of all previous cycles in chronological order to obtain the trend sequence of the monitoring point in each cycle.

[0046] When the power equipment is performing well, its vibration is relatively stable, resulting in small amplitude and fluctuations. The low-frequency component has a high amplitude, while the high-frequency component is less, meaning the amplitude of the high-frequency component is basically insignificant. Therefore, the amplitude change from low frequency to high frequency shows a clear decreasing trend. Secondly, the vibration conditions between different monitoring points on the power equipment are highly similar, and the longitudinal time series changes of vibration data between different monitoring points are also highly similar, resulting in a relatively large vibration trend coefficient.

[0047] The correlation between vibration sequences of the same period monitoring points is used as the horizontal correlation measure, and the correlation between trend sequences of the same period monitoring points is used as the vertical correlation measure. In this embodiment, the correlation is calculated using the Pearson correlation coefficient, but implementers may use other correlation calculation methods.

[0048] The vibration trend coefficient is determined by measuring the horizontal and vertical correlation between all monitoring points in each cycle, as well as the amplitude, fluctuation, and amplitude change trend of the vibration sequence of all monitoring points in each cycle.

[0049] The vibration trend coefficient is positively correlated with both the horizontal and vertical correlation measures; and negatively correlated with the amplitude, fluctuation, and amplitude change trend of the vibration sequence.

[0050] It should be noted that positive correlation means that when one variable increases, the other variable also increases, and the two variables change in the same direction. When one variable changes from large to small or from small to large, the other variable also changes from large to small or from small to large. The specific relationship is determined by the actual application, and this application does not impose any special restrictions.

[0051] It should be noted that negative correlation means that when one variable increases, the other variable decreases accordingly, and the two variables change in opposite directions. When one variable changes from large to small or from small to large, the other variable also changes from small to large or from large to small. The specific relationship is determined by practical application, and this application does not impose any special restrictions.

[0052] Preferably, the expression for the vibration trend coefficient is:

[0053] , This represents the mean of the horizontal correlation measure among all monitoring points in the i-th period. This represents the mean of the longitudinal correlation measure among all monitoring points in the i-th period; Let represent the mean of all peak values ​​in the vibration sequence at the nth monitoring point during the i-th period. This represents the variance of the amplitude in the vibration sequence at the nth monitoring point during the i-th period. This represents the percentage of points in the fitted curve function corresponding to the nth monitoring point in the i-th period where the derivative is positive. Indicates the number of monitoring points. This represents the vibration trend coefficient for the i-th period.

[0054] The amplitude of the vibration sequence is the mean of all peaks in the vibration sequence; the larger the mean, the larger the amplitude. The fluctuation of the vibration sequence is the variance of all amplitudes in the vibration sequence. A positive derivative in the fitted curve function indicates that the point is an increasing point. The proportion of positive derivatives can be regarded as the number of increasing points, i.e. the amplitude change trend. The more increasing points there are, the lower the decreasing trend, the more unstable the vibration, and the smaller the vibration trend coefficient.

[0055] Thus, the vibration trend coefficient for each cycle has been obtained.

[0056] Step S003: Cluster the period based on the vibration trend coefficient, obtain the local change trend and the overall change trend based on the cluster data and all data respectively, and obtain the equipment disturbance coefficient based on the two.

[0057] The vibration trend coefficient measures the performance changes of power equipment from the perspective of vibration signal variation. However, this performance measurement can only measure the trend of performance change from the perspective of vibration signal. Since vibration signals at different monitoring points vary during power equipment operation, and these different trends result in varying degrees of disturbance impact, for example, when the quality of the power input to the equipment fluctuates, the fault point may be more significantly affected by these fluctuations. Therefore, vibration signal analysis based solely on the vibration trend coefficient cannot determine which monitoring point's vibration signal is similar to the disturbance experienced by the overall power equipment monitoring data. This makes it impossible to correct the power equipment monitoring data and thus affects subsequent power equipment fault detection. Therefore, it is necessary to further combine this with global parameter changes such as electrical parameters and equipment temperature to conduct disturbance analysis of the power equipment monitoring data, thereby determining the degree of disturbance experienced by the power equipment.

[0058] For the oscillation trend coefficient of each cycle, the oscillation trend coefficients of all previous cycles are used as input. A K-means clustering algorithm is then used to classify all cycles, where the K value in the clustering algorithm is determined using the elbow method. K-means clustering and the elbow method are well-known techniques and will not be elaborated further.

[0059] Record voltage data, current data, and device temperature as data types;

[0060] For equipment temperature, the equipment temperature within a time interval is taken as the equipment temperature belonging to the period within that time interval.

[0061] The periods of each data type in each cluster are sorted in chronological order. The data values ​​are used as the x-axis and the corresponding time series are used as the y-axis. Polynomial fitting is performed to obtain the polynomial function for each data type, which is denoted as the first polynomial function.

[0062] For each data type, all periods are sorted in chronological order. The data values ​​are used as the x-axis, and the corresponding time series are used as the y-axis. Polynomial fitting is then performed to obtain the polynomial function for each data type, denoted as the second polynomial function. The x-axis lengths of the first and second polynomial functions are the same.

[0063] Since global data types such as equipment temperature, voltage, and current are caused by non-faulty reasons, clustering the periods using a clustering algorithm results in relatively uniform characteristics for each period. Therefore, the equipment disturbance coefficient can be determined by comparing the clusters with the overall data. Thus, the initial equipment disturbance coefficient is adjusted based on the difference between the overall and local trends to determine the equipment disturbance coefficient for each period. The overall trend is represented by the area integral of the second polynomial function, while the trend of the locally integrated data types is the weighted sum of the areas integral of all clusters.

[0064] The disturbance coefficient of the equipment is positively correlated with the difference between the overall change trend and the local change trend.

[0065] First, the first polynomial function is weighted using the ratio of the sum of vibration trend coefficients in each cluster to the sum of all vibration trends to obtain the local change trend.

[0066] The preferred expression for the local change trend is:

[0067] , This represents the sum of the vibration trend coefficients in the j-th cluster. This represents the sum of all vibration trend coefficients. Let the integral area of ​​the first polynomial function in the j-th cluster be the area of ​​the m-th data type. This represents the local change trend of the m-th data type in the i-th period.

[0068] Preferably, in this embodiment, the expression for the device disturbance coefficient is:

[0069] , This represents the overall trend of change of the m-th data type in the i-th period. This represents the local change trend of the m-th data type in the i-th period. Indicates the number of data types. This represents the initial value of the equipment disturbance coefficient; in this embodiment, the initial value is 1. This represents the equipment disturbance coefficient in the i-th cycle.

[0070] It is understandable that the more similar the vibration changes, the more similar the performance of the corresponding power equipment. This means that the differences in global parameters such as temperature, voltage, and current are mainly due to non-fault causes. Therefore, by performing curve fitting on each cluster, the changing trend of power equipment parameters under non-fault causes can be determined. The changing trends of the actual obtained equipment temperature, voltage, and current data are the coupling result of fault and non-fault causes. Therefore, the impact of fault causes can be measured by the changing trends under non-fault causes. Secondly, since the data changing trends in each cluster may be different, when the impact of fault causes is greater, it is difficult to completely eliminate the influence of fault causes in the data changes. Therefore, the results of different clusters are integrated by weighted averaging. Thus, the obtained equipment disturbance coefficient represents the data fluctuation caused by fault causes.

[0071] Thus, the device disturbance coefficient for the i-th cycle has been obtained.

[0072] Step S004: Determine the prediction function based on the equipment disturbance coefficient, and obtain the corrected vibration signal by combining the difference in vibration trend coefficients between adjacent periods, thus completing dynamic monitoring.

[0073] When monitoring power equipment, the equipment disturbance coefficient for each cycle is obtained. The equipment disturbance coefficient is used as the abscissa, and the difference in vibration trend coefficient between two adjacent cycles is used as the ordinate. Polynomial fitting is performed on the equipment to obtain the data fluctuation prediction function.

[0074] For each cycle of the vibration signal, obtain the equipment disturbance coefficient and vibration trend coefficient for that cycle, and then obtain the vibration trend coefficient of the previous cycle. Substitute the equipment disturbance coefficient for that cycle into the fluctuation prediction function to obtain the fluctuation prediction value for that cycle.

[0075] The objective function is determined based on the difference between the predicted fluctuation value and the vibration trend coefficient of two adjacent periods. An optimization algorithm is used to iterate the objective function until its minimum value is obtained. The vibration signal corresponding to the minimum value is then used as the corrected vibration signal for that period.

[0076] The objective function of the optimization algorithm is:

[0077] , Let v be the vibration trend coefficient of the x-th cycle. This represents the vibration trend coefficient for the (x-1)th period. This represents the predicted fluctuation value for the x-th period. This represents the target value in the x-th period and the v-th iteration.

[0078] Under a preset number of iterations, the vibration signal corresponding to the minimum value of the objective function at the corresponding iteration number is the corrected vibration signal. In this embodiment, the number of iterations is 100.

[0079] The corrected vibration signal is used to dynamically monitor the fault status of power equipment using vibration testing. Vibration testing is a well-known technology in the field of non-electrical quantity detection of power equipment, and will not be elaborated further.

[0080] This completes the dynamic monitoring of the power equipment.

[0081] Based on the same inventive concept as the above method, this embodiment of the invention also provides a real-time dynamic monitoring system for power equipment systems based on the Internet of Things, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described methods for real-time dynamic monitoring of power equipment systems based on the Internet of Things.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

[0083] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for real-time dynamic monitoring of power equipment systems based on the Internet of Things, characterized in that, The method includes the following steps: Acquire vibration signals, voltage data, current data, and equipment temperature over a time interval for each monitoring point in each cycle; Vibration signals within a cycle are used to form a vibration sequence. The vibration sequence is then converted into a spectrum and fitted to obtain a fitting curve function. For each monitoring point, the coefficients of variation of the vibration sequences in all previous cycles are sorted to obtain the trend sequence of the cycle. The correlation between vibration sequences and the correlation between trend sequences among monitoring points in the same cycle are recorded as the horizontal correlation measure and the vertical correlation measure, respectively. The amplitude change trend is determined by the fitted curve function. The vibration trend coefficient is determined by the horizontal and vertical correlation measures among all monitoring points in each cycle, as well as the amplitude, fluctuation, and amplitude change trend of the vibration sequences of all monitoring points in each cycle. Periodic clustering is performed based on vibration trend coefficients; voltage data, current data, and equipment temperature are denoted as data types; for each data type, the changes in data within the cluster and the changes in all data are denoted as local change trends and overall change trends, respectively; the equipment disturbance coefficient for each period is obtained based on the local change trends and overall change trends, and the equipment disturbance coefficient is positively correlated with the difference between the overall change trend and the local change trend; The equipment disturbance coefficient is fitted to determine the fluctuation prediction function. Based on the fluctuation prediction function and the difference in vibration trend coefficient between adjacent periods, the vibration signal is corrected using an optimization algorithm to obtain the corrected vibration signal, and dynamic monitoring is completed. The process of acquiring the corrected vibration signal is as follows: The equipment disturbance coefficients for each period are substituted into the fluctuation prediction function to obtain the fluctuation prediction value for each period; the objective function of the optimization algorithm is: , Let v be the vibration trend coefficient of the x-th cycle. This represents the vibration trend coefficient for the (x-1)th period. This represents the predicted fluctuation value for the x-th period. This represents the target value of the v-th iteration in the x-th period; Under a preset number of iterations, the vibration signal corresponding to the minimum value of the objective function at the corresponding iteration number is taken as the corrected vibration signal.

2. The method for real-time dynamic monitoring of power equipment systems based on the Internet of Things as described in claim 1, characterized in that, The method for obtaining the fitting curve function by converting the vibration sequence into a spectrum diagram is as follows: The vibration sequence of each monitoring point in each cycle is used as input, and the corresponding spectrum is obtained by Fourier transform. In the spectrum, the frequency is used as the horizontal axis and the corresponding amplitude is used as the vertical axis. Polynomial fitting is performed to output the fitted curve function.

3. The method for real-time dynamic monitoring of power equipment systems based on the Internet of Things as described in claim 1, characterized in that, The vibration trend coefficient is positively correlated with both the horizontal and vertical correlation measures; and negatively correlated with the amplitude, fluctuation, and amplitude change trend of the vibration sequence.

4. The method for real-time dynamic monitoring of power equipment systems based on the Internet of Things as described in claim 3, characterized in that, The amplitude of the vibration sequence is the mean of all peaks in the vibration sequence, the fluctuation of the vibration sequence is the variance of all amplitudes in the vibration sequence, and the amplitude change trend of the vibration sequence is the proportion of data points with positive derivatives in the fitted curve function.

5. The method for real-time dynamic monitoring of power equipment systems based on the Internet of Things as described in claim 1, characterized in that, The equipment temperature for each cycle is the equipment temperature within the time interval of that cycle.

6. The method for real-time dynamic monitoring of power equipment systems based on the Internet of Things as described in claim 1, characterized in that, The method for recording changes in data within clusters and changes in all data as local and overall trends, respectively, is as follows: The periods of each data type in each cluster are sorted in chronological order. The data values ​​are used as the x-axis and the corresponding time series are used as the y-axis. Polynomial fitting is performed to obtain the polynomial function of each data type, which is denoted as the first polynomial function. Sort all periods of each data type in chronological order, use the data values ​​as the x-axis and the corresponding time series as the y-axis, and perform polynomial fitting to obtain the polynomial function of each data type, which is denoted as the second polynomial function. Integrating the first and second polynomial functions respectively yields the overall trend and the local trend.

7. The method for real-time dynamic monitoring of power equipment systems based on the Internet of Things as described in claim 6, characterized in that, The method for obtaining the overall trend and local trend by integrating the first polynomial function and the second polynomial function respectively is as follows: The overall trend is the area of ​​the integral of the second polynomial function; The expression for the local trend of change is: , This represents the sum of the vibration trend coefficients in the j-th cluster. This represents the sum of all vibration trend coefficients. Let the integral area of ​​the first polynomial function in the j-th cluster be the area of ​​the m-th data type. This represents the local change trend of the m-th data type in the i-th period.

8. A real-time dynamic monitoring system for power equipment based on the Internet of Things, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the real-time dynamic monitoring method for an Internet of Things-based power equipment system as described in any one of claims 1-7.

Images