Intelligent power grid equipment fault monitoring system and method based on Internet of Things

Abstract

The invention discloses an intelligent power grid equipment fault monitoring system and method based on the Internet of Things, and relates to the technical field of the Internet of Things. Discrete Fourier transform and a Fisher information processing method are adopted to fuse multi-source data, meteorological interference is corrected, and accurate collection of electrical quantity and quantification of environmental influence are achieved; and setting a fault monitoring sliding window and a harmonic analysis time interval, and obtaining harmonic data of different time scales. Extracting harmonic features through slope calculation, and identifying voltage abnormal features; historical data is processed according to a time sequence prediction algorithm to construct a prediction set, voltage abnormity is judged based on a historical data threshold value, intelligent early warning of the voltage state is achieved, the abnormity capturing precision is improved, and when the system detects the voltage abnormity, information of strain section equipment sending out a fault early warning signal is obtained and transmitted to an OTA upgrading terminal. And enabling the OTA upgrading terminal to carry out state identification and data interaction on the upgrading process of the intelligent power grid equipment based on the equipment information.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, specifically to an IoT-based smart grid equipment fault monitoring system and method. Background Technology

[0002] Smart grid equipment is crucial for ensuring the safe and stable operation of the power system. Through real-time monitoring and intelligent control, it improves grid efficiency, reduces operation and maintenance costs, effectively addresses complex challenges such as the integration of new energy sources, and provides core support for the intelligent transformation of the power grid and sustainable energy development.

[0003] In smart grid transmission lines, the electrical characteristics of equipment in different tension sections are affected by factors such as weather and fluctuations in electrical quantities, resulting in varying degrees of attenuation that accumulates continuously. This can easily lead to hidden anomalies, which traditional monitoring methods struggle to accurately detect, resulting in delayed or missed fault identification. If potential equipment problems cannot be detected in a timely manner, the OTA (Over-The-Air) upgrade system may receive incorrect or incomplete equipment ID information. If the upgrade is initiated prematurely when the equipment hardware is abnormal, it can easily lead to program malfunctions, version verification failures, or equipment crashes, affecting the upgrade's effectiveness and the safe and stable operation of the power grid. Therefore, there is an urgent need for an intelligent fault monitoring method that can integrate multi-source data, dynamically correct for environmental influences, and accurately predict equipment status. Summary of the Invention

[0004] The purpose of this invention is to provide a smart grid equipment fault monitoring system and method based on the Internet of Things, so as to solve the problems raised in the prior art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a method for fault monitoring of smart grid equipment based on the Internet of Things, the method comprising the following steps: Step S1: Select any tension section in the smart grid transmission line as the research object, denoted as the target tension section, and set the electrical quantity data acquisition cycle to continuously monitor and obtain the electrical quantity data of the target tension section to construct the first harmonic drop analysis set; Step S1-1: Install power sensors on the towers at both ends of the target tension section to acquire electrical quantity data of the smart grid on the target tension section, and extract the original three-phase voltage signal data of the target tension section based on the electrical quantity data. Step S1-2: Transmit the raw three-phase voltage signal data to the data service terminal via power line carrier communication. The data service terminal is an integrated data hub with the ability to receive, process, store, and transmit data. Step S1-3: Utilize the original three-phase voltage signal data from Discrete Fourier Transform and extract the amplitude and phase data of each harmonic component. Combine the amplitude difference data and phase difference data of each harmonic component into data units, arrange them in chronological order, and construct the first harmonic drop analysis set. The formula for the Discrete Fourier Transform is as follows: ; In the formula, X(k) represents the complex spectrum of the k-th harmonic, including the amplitude |X(k)| and the phase ∠X(k); N represents the number of sampling points, specifically the total number of samples within one electrical quantity acquisition cycle; x(n) represents the discrete-time signal, specifically the n-th sampling point of the original three-phase voltage signal; k represents the frequency index, specifically the corresponding harmonic order, such as k=1 for the fundamental frequency, k=5 for the 5th harmonic; j represents the imaginary unit. By installing power sensors on the towers at both ends of the target tension section to acquire the raw three-phase voltage signal data, and transmitting it to the integrated data center via power line carrier communication, the amplitude and phase data of each harmonic component are extracted using discrete Fourier transform and a first harmonic drop analysis set is constructed. This enables real-time monitoring and accurate analysis of the electrical quantity data of the target tension section, providing a reliable data foundation for subsequent fault monitoring and effectively improving the accuracy and timeliness of power grid equipment fault early warning.

[0006] Step S2: Based on the electrical quantity data acquisition cycle, acquire meteorological data of the target tension section in real time, and use Fisher's information processing method to combine meteorological data for quantitative analysis to construct a meteorological coupling correction model. Analyze and process the first harmonic drop analysis set through the meteorological coupling correction model to obtain the second harmonic drop analysis set. Step S2-1: Install meteorological sensors on the target tension section to acquire meteorological data in the area. The meteorological sensors include wind speed and direction sensors, temperature and humidity sensors, rain gauge sensors, and atmospheric pressure sensors. Transmit the acquired meteorological data to the data service terminal via power line carrier communication. Step S2-2: Construct a meteorological coupling correction model using Fisher's information processing method. Then, construct a multidimensional joint probability density function for the meteorological data uploaded during each electrical quantity data acquisition cycle using the meteorological coupling correction model. The formula for calculating the multidimensional joint probability density function is as follows: ; In the formula, x represents a meteorological data vector with a dimension of 4, corresponding to the acquired meteorological data parameters. Specifically, x1 represents the wind speed parameter; x2 represents the temperature and humidity parameter; x3 represents the rainfall parameter; x4 represents the atmospheric pressure parameter; and θ represents the probability distribution parameter vector, corresponding to the meteorological parameters. i Let x be the type i meteorological parameter. i The distribution parameters; It is represented as a univariate probability density function of the i-th type of meteorological parameter; The formula for Fisher's information processing method is as follows: ; In the formula, In the Fisher information matrix, the dimension is 4×4, and the element Ijk represents the parameter θ. j and θ k The degree of information coupling between them; Represented as the statistical average of a sample of meteorological parameters; log-likelihood function ; Represented as the log-likelihood function θ j First-order partial derivatives, parameter θ j Impact rate on data distribution.

[0007] Step S2-3: Calculate the parameters contained in each set of meteorological data using the Fisher information matrix in the meteorological coupling correction model, and determine the coupling coefficients between the meteorological parameters. The parameters include wind speed parameters, temperature and humidity parameters, rainfall parameters, and atmospheric pressure parameters. The coupling coefficients between meteorological parameters are obtained by standardizing the Fisher information processing method, and the calculation formula is as follows: ; Where R jk This is expressed as the coupling coefficient between meteorological parameters; Step S2-4: Use the calculated coupling coefficients as weighting factors to perform weighted correction on each set of data units in the first harmonic drop analysis set. After weighted correction calculation, the second harmonic drop analysis set of the target tension section is obtained. The formula for calculating the coupling coefficient of meteorological parameters is as follows: ; In the formula, wi represents the weighting factor of each parameter in the meteorological parameters, and the sum of the weighting factors of each parameter in the meteorological parameters is 1; i represents the parameter number in the meteorological parameters, and the value of i ranges from 1 to 4, and the value is an integer. The weighted correction is calculated using the following formula: ; In the formula, Represented as a data unit after weighted correction calculation; It represents the harmonic drop component under the influence of the i-th type of meteorological parameter.

[0008] By installing multiple types of meteorological sensors on the target tension section to collect meteorological data in real time and transmitting it to the terminal, a meteorological coupling correction model is constructed using the Fisher information processing method. The coupling coefficients between meteorological parameters are calculated as weighting factors to weight and correct the first harmonic drop analysis set, resulting in the second harmonic drop analysis set. This effectively quantifies the impact of meteorological factors on electrical quantity data, dynamically corrects environmental interference, improves the adaptability of the analysis set to actual operating conditions, provides a more realistic data foundation for subsequent fault feature extraction, and enhances the robustness of fault monitoring methods to complex external conditions.

[0009] Step S3: Set the fault monitoring sliding window and harmonic analysis time interval. Analyze and process the second harmonic drop analysis set through the harmonic analysis time interval to obtain the maximum value of the first harmonic drop of the target tension section. Analyze and process the fault monitoring sliding window through the harmonic analysis time interval to obtain the maximum value of the second harmonic drop of the target tension section. Step S3-1: Set the fault monitoring sliding window to n consecutive electrical quantity data acquisition cycles; set the harmonic analysis time interval according to the electrical quantity data acquisition frequency of the target tension section within the electrical quantity data acquisition cycle; Step S3-2: Select the electrical quantity data acquisition cycle as the basic unit of the sliding step size of the fault monitoring sliding window, and perform point-by-point sliding processing on the second harmonic drop analysis set; Step S3-3: Select the harmonic analysis time interval as the basic unit, calculate the amplitude difference slope and phase difference slope of the corresponding harmonic components in the second harmonic drop analysis set, and after normalizing the amplitude difference slope and phase difference slope, calculate the comprehensive slope of a single harmonic analysis time interval through weighted fusion; select the maximum comprehensive slope obtained from each harmonic analysis time interval in the second harmonic drop analysis set, and use the comprehensive slope corresponding to the start time of the harmonic analysis time interval minus the comprehensive slope corresponding to the end time to obtain the absolute value of the difference as the first harmonic drop maximum value, which is denoted as the first harmonic drop maximum value. Step S3-4: Simultaneously using the calculation methods of amplitude difference slope and phase difference slope of harmonic components in step S3-3, analyze and calculate the harmonic components within the fault monitoring sliding window to obtain the maximum value of the second harmonic drop of the target tension section. By setting a fault monitoring sliding window and a harmonic analysis time interval, and processing the second harmonic drop analysis set point by point with the electrical quantity data acquisition cycle as the sliding step, the amplitude difference and phase difference slope of the harmonic components are calculated and normalized and weighted to obtain the comprehensive slope. Then, the maximum values ​​of the first and second harmonic drops are extracted, which can realize dynamic quantitative analysis of the trend of harmonic component changes, capture the harmonic drop characteristics at different time scales, and provide multi-dimensional key parameters for subsequent voltage characteristic comprehensive value calculation and fault threshold judgment, effectively improving the ability to capture abnormal changes in electrical quantities in the tension section and the accuracy of fault feature identification.

[0010] Step S4: Calculate the comprehensive voltage characteristic drop value by weighted fusion based on the maximum value of the first harmonic drop and the maximum value of the second harmonic drop; collect electrical quantity data of the target tension section in real time, and combine the electrical quantity fluctuation limit in the smart grid transmission line to screen out the electrical quantity data that meet the conditions in the fault monitoring sliding window to construct the harmonic drop trough set and the harmonic drop peak set; Step S4-1: Based on the weighted fusion calculation, combine the maximum value of the first harmonic drop and the maximum value of the second harmonic drop of the target tension section to obtain the comprehensive voltage characteristic drop value; Step S4-2: Extract the raw three-phase voltage signal data based on the real-time collected electrical quantity data of the target tension section, and analyze and process the data through the data service terminal to obtain the three-phase voltage data, which includes voltage amplitude data and voltage phase data. Step S4-3: Normalize the voltage amplitude data and voltage phase data of the target tension section in real time, and simultaneously select the weighting coefficient used in the comprehensive slope calculation in step S3-3 to calculate the comprehensive voltage characteristic value of the target tension section at the current moment. Step S4-4: Obtain ideal three-phase voltage data and electrical quantity fluctuation limits in the smart grid transmission line. The electrical quantity fluctuation limits include upper and lower limits of electrical quantity fluctuation. Based on the ideal three-phase voltage data, calculate the comprehensive value of ideal voltage characteristics using the weighting coefficients used in the comprehensive slope calculation in step S3-3. Step S4-5: When the real-time obtained comprehensive voltage characteristic value exceeds the ideal comprehensive voltage characteristic value, the comprehensive voltage characteristic value is added to the comprehensive voltage characteristic difference value, and the sum is compared with the upper limit of electrical quantity fluctuation. The specific process is as follows: Step S4-5-1: When the sum exceeds the upper limit of electrical quantity fluctuation, use the sum minus the upper limit of electrical quantity fluctuation to obtain the voltage characteristic increment value, select the current timestamp data as the index, and store the voltage characteristic increment value to construct a set of harmonic drop peaks; Step S4-5-2: When the sum does not exceed the upper limit of electrical quantity fluctuation, the voltage characteristic increment value corresponding to the current moment is set to 0 by default and stored in the harmonic drop peak set. One electrical quantity data acquisition cycle corresponds to one harmonic drop peak set. Step S4-6: When the real-time obtained comprehensive voltage characteristic value does not exceed the ideal comprehensive voltage characteristic value, subtract the comprehensive voltage characteristic value from the comprehensive voltage characteristic drop value, and compare the resulting difference with the lower limit value of electrical quantity fluctuation. The specific process is as follows: Step S4-6-1: When the difference does not exceed the lower limit of electrical quantity fluctuation, subtract the difference from the lower limit of electrical quantity fluctuation to obtain the voltage characteristic deficit difference. Select the current timestamp data as the index and the voltage characteristic deficit value as the value to store and construct a set of harmonic drop valleys. Step S4-6-2: When the difference exceeds the lower limit of electrical quantity fluctuation, the voltage characteristic deficit value corresponding to the current moment is set to 0 by default and stored in the harmonic drop valley set. One electrical quantity data acquisition cycle corresponds to one harmonic drop valley set. By weighted fusion of the maximum values ​​of the first and second harmonic drops, a comprehensive voltage characteristic drop value is obtained. Combined with real-time collected electrical quantity data and normalized processing, the current comprehensive voltage characteristic value is calculated. It is compared with the ideal value and data is filtered according to the electrical quantity fluctuation limit to construct a set of harmonic drop peaks and valleys. This can provide a quantitative comprehensive index for the voltage characteristics of the tension section, dynamically filter abnormal electrical quantity data, accurately construct a dataset reflecting voltage fluctuation characteristics, and effectively improve the identification accuracy of fault characteristics of power grid equipment and the ability to capture abnormal states.

[0011] Step S5: Analyze and process the set of harmonic drop troughs and the set of harmonic drop peaks in the fault monitoring sliding window using a time series prediction algorithm, predict the set of harmonic drop troughs and the set of harmonic drop peaks for the next electrical quantity data acquisition cycle of the target tension section, and monitor the target tension section based on the prediction results. Step S5-1: Obtain the set of harmonic drop troughs and the set of harmonic drop peaks generated by the fault monitoring sliding window over multiple historical sliding steps; Step S5-2: Use the time series prediction algorithm to process each historical harmonic drop trough set to obtain the harmonic drop trough set generated by the next step of the fault monitoring sliding window, which is denoted as the predicted trough set. Step S5-3: Use the SARIMA model to process the historical harmonic drop peak sets respectively, and obtain the harmonic drop peak set generated by the next step of the fault monitoring sliding window, which is denoted as the predicted peak set. Step S5-4: Calculate the proportion of 0 based on the ratio of the number of data points with a value of 0 in each historical harmonic drop trough set to the total number of data points in the corresponding harmonic drop trough set within the sliding step. Step S5-5: Calculate the proportion of 0 based on the ratio of the number of data points with a value of 0 in each historical set of harmonic drop peak values ​​to the total number of data points in the set of harmonic drop peak values ​​within the corresponding sliding step. Steps S5-6: Set the trough threshold according to the proportion of 0 in the historical harmonic drop trough set, and set the peak threshold according to the proportion of 0 in the historical harmonic drop peak set. Step S5-7: Calculate the proportion of 0 in the predicted trough set, and record it as the predicted trough value; and simultaneously calculate the proportion of 0 in the predicted peak set, and record it as the predicted peak value. Based on the mean and standard deviation, the trough threshold and peak threshold are set according to the three sigma principle. The mean is selected plus k times the standard deviation. The value of k can be determined to be between 1 and 3 according to the three sigma principle.

[0012] Step S5-8: Monitor the target tension section based on the prediction results. The specific process is as follows: Step S5-8-1: When the predicted low value does not exceed the low value threshold, it is determined that there is an abnormal voltage shortage in the target tension section, and a fault warning signal is issued; when the predicted low value exceeds the low value threshold, it is determined that the voltage of the target tension section is normal. Step S5-8-2: When the predicted peak value exceeds the peak value threshold, it is determined that there is an abnormal voltage increment in the target tension section, and a fault warning signal is issued; when the predicted peak value does not exceed the peak value threshold, it is determined that the voltage of the target tension section is normal. Step S5-9: Obtain the information of the tension section equipment that issued the fault warning signal, and transmit the tension section equipment information to the OTA upgrade terminal through power line carrier communication, so that the OTA upgrade terminal can perform status identification and data interaction for the upgrade process of smart grid equipment based on the tension section equipment information.

[0013] By acquiring the set of harmonic drop troughs and peaks of historical sliding step sizes, and using time series prediction algorithms to obtain the prediction set for the next step size, the proportion of 0 in the historical and predicted sets is calculated and a threshold is set. Voltage anomalies are identified and warnings are issued based on the comparison between the predicted value and the threshold. This enables dynamic prediction and intelligent judgment of the voltage status of the target tension section. A personalized threshold system is constructed based on historical data characteristics, improving the foresight and accuracy of fault warnings. This provides data support for the early identification and timely handling of abnormal states of power grid equipment, effectively reducing the rate of missed fault detection and maintenance lag. By accurately capturing voltage anomalies in the target tension section, the system can quickly associate unique equipment IDs and synchronize potential equipment fault information to the OTA upgrade system. Based on the received equipment status data, the OTA system suspends the upgrade process for equipment at risk of voltage anomalies, avoiding upgrade failures and program crashes due to unstable hardware performance caused by voltage fluctuations. This achieves intelligent linkage between fault warnings and software upgrades, ensuring the safety and reliability of the upgrade process and improving the overall efficiency and safety of power grid equipment operation and maintenance management.

[0014] Furthermore, an IoT-based smart grid equipment fault monitoring system includes a data acquisition module, a weather correction module, a sliding analysis module, a feature calculation module, and a prediction and early warning module. The data acquisition module is used to acquire electrical quantity data of smart grid transmission lines and construct an analysis set; the meteorological correction module is used to correct the analysis set by combining meteorological data; the sliding analysis module is used to set parameters and calculate harmonic drop values; the feature calculation module is used to calculate voltage feature values ​​and filter data; the prediction and early warning module is used to predict data and determine faults. The output of the data acquisition module is electrically connected to the input of the meteorological correction module; the output of the meteorological correction module is electrically connected to the input of the sliding analysis module; the output of the sliding analysis module is electrically connected to the input of the feature calculation module; and the output of the feature calculation module is electrically connected to the input of the forecast and early warning module. The data acquisition module includes a power acquisition unit and a harmonic set construction unit; the power acquisition unit is used to install sensors to acquire electrical quantity data and transmit it to the terminal; the harmonic set construction unit is used to process voltage signal data to construct a first harmonic drop analysis set; The meteorological correction module includes a meteorological parameter acquisition unit and a coupled correction calculation unit; the meteorological parameter acquisition unit is used to install meteorological sensors to acquire regional meteorological data; the coupled correction calculation unit is used to construct a set of weight factor correction analysis for model calculation. The sliding analysis module includes a window parameter setting unit and a harmonic drop calculation unit; the window parameter setting unit is used to set the sliding window and analysis time interval parameters; the harmonic drop calculation unit is used to calculate the maximum values ​​of the first and second harmonic drops; The feature calculation module includes a comprehensive value calculation unit and a data filtering and construction unit; the comprehensive value calculation unit is used to calculate the comprehensive voltage feature drop value and the current comprehensive value; the data filtering and construction unit is used to construct the set of harmonic drop peaks and valleys; The prediction and early warning module includes a time series prediction unit and a fault threshold judgment unit; the time series prediction unit is used to predict the harmonic drop set data for the next step; the fault threshold judgment unit is used to set the threshold and determine whether to issue an early warning signal.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention deploys electrical and meteorological sensors on the target tension section, and combines discrete Fourier transform and Fisher information processing method to construct a harmonic analysis set of multi-source data fusion and dynamically correct meteorological interference. This enables real-time and accurate acquisition of electrical quantity data of the tension section and quantitative processing of environmental impact, providing a highly reliable data foundation for fault monitoring and effectively improving the effectiveness of data characteristics and the adaptability of analysis models under complex working conditions.

[0016] 2. This invention sets a fault monitoring sliding window and a harmonic analysis time interval, extracts harmonic drop features at multiple time scales based on slope calculation and weighted fusion, and constructs peak and trough sets in combination with electrical quantity fluctuation limits. This enables dynamic screening and structured characterization of voltage anomaly features, providing multi-dimensional key input parameters for subsequent prediction models and effectively improving the accuracy of capturing abnormal equipment states and the ability to identify features.

[0017] 3. This invention uses a time series prediction algorithm to process historical data and construct a prediction set. Based on the characteristics of historical data, it sets personalized thresholds and judges voltage anomalies, realizing forward-looking prediction and intelligent early warning of the voltage status of the target tension section, breaking through the limitation of traditional methods that rely on fixed thresholds. When the system detects an abnormal voltage in the target tension section, it transmits the tension section equipment information to the OTA upgrade terminal in real time, enabling the OTA upgrade terminal to mark the abnormal equipment status, avoiding upgrade risks caused by voltage instability, and improving the reliability of power grid equipment operation and maintenance. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a method for fault monitoring of smart grid equipment based on the Internet of Things according to the present invention. Figure 2 This is a schematic diagram of the structure of an Internet of Things-based smart grid equipment fault monitoring system according to the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Example 1: As Figure 1 As shown, the present invention provides a technical solution, a method for fault monitoring of smart grid equipment based on the Internet of Things, the intelligent management method including the following steps: Step S1: Select any tension section in the smart grid transmission line as the research object, denoted as the target tension section, and set the electrical quantity data acquisition cycle to continuously monitor and obtain the electrical quantity data of the target tension section to construct the first harmonic drop analysis set; Step S1-1: Install power sensors on the towers at both ends of the target tension section to acquire electrical quantity data of the smart grid on the target tension section, and extract the original three-phase voltage signal data of the target tension section based on the electrical quantity data. Step S1-2: Transmit the raw three-phase voltage signal data to the data service terminal via power line carrier communication. The data service terminal is an integrated data hub with the ability to receive, process, store, and transmit data. Step S1-3: Utilize the original three-phase voltage signal data from Discrete Fourier Transform and extract the amplitude and phase data of each harmonic component. Combine the amplitude difference data and phase difference data of each harmonic component into data units, arrange them in chronological order, and construct the first harmonic drop analysis set. In practical implementation, taking a tension section of a 220kV transmission line as an example, the raw three-phase voltage signals are collected in real time by a power sensor. The time-domain signal is converted into the frequency domain using discrete Fourier transform, and the amplitude and phase difference of each harmonic are extracted to construct an analysis set, thereby capturing the harmonic distortion characteristics of the voltage signal. It is important to note that the sensor installation position should ensure the accuracy of signal acquisition, and the sampling frequency should meet the power grid harmonic analysis standards to avoid data distortion due to installation errors or insufficient sampling frequency.

[0021] Step S2: Based on the electrical quantity data acquisition cycle, acquire meteorological data of the target tension section in real time, and use Fisher's information processing method to combine meteorological data for quantitative analysis to construct a meteorological coupling correction model. Analyze and process the first harmonic drop analysis set through the meteorological coupling correction model to obtain the second harmonic drop analysis set. Step S2-1: Install meteorological sensors on the target tension section to acquire meteorological data in the area. The meteorological sensors include wind speed and direction sensors, temperature and humidity sensors, rain gauge sensors, and atmospheric pressure sensors. Transmit the acquired meteorological data to the data service terminal via power line carrier communication. Step S2-2: Construct a meteorological coupling correction model using Fisher's information processing method. Then, construct a multidimensional joint probability density function for the meteorological data uploaded during each electrical quantity data acquisition cycle using the meteorological coupling correction model. Step S2-3: Calculate the parameters contained in each set of meteorological data using the Fisher information matrix in the meteorological coupling correction model, and determine the coupling coefficients between the meteorological parameters. The parameters include wind speed parameters, temperature and humidity parameters, rainfall parameters, and atmospheric pressure parameters. Step S2-4: Use the calculated coupling coefficients as weighting factors to perform weighted correction on each set of data units in the first harmonic drop analysis set. After weighted correction calculation, the second harmonic drop analysis set of the target tension section is obtained. In practical implementation, the Fisher information processing method is used to construct a meteorological coupling correction model. The influence of meteorological parameters such as wind speed, temperature and humidity on electrical quantity data is quantified by a multidimensional joint probability density function. The coupling coefficient between parameters is calculated based on the Fisher information matrix, and this is used as a weighting factor to correct the harmonic drop analysis set, because meteorological factors will change the line impedance and distributed capacitance, thereby affecting the harmonic components.

[0022] Step S3: Set the fault monitoring sliding window and harmonic analysis time interval. Analyze and process the second harmonic drop analysis set through the harmonic analysis time interval to obtain the maximum value of the first harmonic drop of the target tension section. Analyze and process the fault monitoring sliding window through the harmonic analysis time interval to obtain the maximum value of the second harmonic drop of the target tension section. Step S3-1: Set the fault monitoring sliding window to n consecutive electrical quantity data acquisition cycles; set the harmonic analysis time interval according to the electrical quantity data acquisition frequency of the target tension section within the electrical quantity data acquisition cycle; Step S3-2: Select the electrical quantity data acquisition cycle as the basic unit of the sliding step size of the fault monitoring sliding window, and perform point-by-point sliding processing on the second harmonic drop analysis set; Step S3-3: Select the harmonic analysis time interval as the basic unit, calculate the amplitude difference slope and phase difference slope of the corresponding harmonic components in the second harmonic drop analysis set, and after normalizing the amplitude difference slope and phase difference slope, calculate the comprehensive slope of a single harmonic analysis time interval through weighted fusion; select the maximum comprehensive slope obtained from each harmonic analysis time interval in the second harmonic drop analysis set, and use the comprehensive slope corresponding to the start time of the harmonic analysis time interval minus the comprehensive slope corresponding to the end time to obtain the absolute value of the difference as the first harmonic drop maximum value, which is denoted as the first harmonic drop maximum value. Step S3-4: Simultaneously using the calculation methods of amplitude difference slope and phase difference slope of harmonic components in step S3-3, analyze and calculate the harmonic components within the fault monitoring sliding window to obtain the maximum value of the second harmonic drop of the target tension section. In practice, a sliding window with n consecutive acquisition cycles is set to capture short-term data fluctuations. The amplitude difference and phase difference slope are calculated using the harmonic analysis time interval. After normalization and weighted fusion, the comprehensive slope is obtained. The maximum harmonic drop value is determined by the absolute value of the difference between the comprehensive slope at the start and end times, thereby identifying the harmonic abrupt change characteristics during a fault. The sliding window size and time interval are adjusted according to the line length and load characteristics.

[0023] Step S4: Calculate the comprehensive voltage characteristic drop value by weighted fusion based on the maximum value of the first harmonic drop and the maximum value of the second harmonic drop; collect electrical quantity data of the target tension section in real time, and combine the electrical quantity fluctuation limit in the smart grid transmission line to screen out the electrical quantity data that meet the conditions in the fault monitoring sliding window to construct the harmonic drop trough set and the harmonic drop peak set; Step S4-1: Based on the weighted fusion calculation, combine the maximum value of the first harmonic drop and the maximum value of the second harmonic drop of the target tension section to obtain the comprehensive voltage characteristic drop value; Step S4-2: Extract the raw three-phase voltage signal data based on the real-time collected electrical quantity data of the target tension section, and analyze and process the data through the data service terminal to obtain the three-phase voltage data, which includes voltage amplitude data and voltage phase data. Step S4-3: Normalize the voltage amplitude data and voltage phase data of the target tension section in real time, and simultaneously select the weighting coefficient used in the comprehensive slope calculation in step S3-3 to calculate the comprehensive voltage characteristic value of the target tension section at the current moment. Step S4-4: Obtain ideal three-phase voltage data and electrical quantity fluctuation limits in the smart grid transmission line. The electrical quantity fluctuation limits include upper and lower limits of electrical quantity fluctuation. Based on the ideal three-phase voltage data, calculate the comprehensive value of ideal voltage characteristics using the weighting coefficients used in the comprehensive slope calculation in step S3-3. Step S4-5: When the real-time obtained comprehensive voltage characteristic value exceeds the ideal comprehensive voltage characteristic value, the comprehensive voltage characteristic value is added to the comprehensive voltage characteristic difference value, and the sum is compared with the upper limit of electrical quantity fluctuation. The specific process is as follows: Step S4-5-1: When the sum exceeds the upper limit of electrical quantity fluctuation, use the sum minus the upper limit of electrical quantity fluctuation to obtain the voltage characteristic increment value, select the current timestamp data as the index, and store the voltage characteristic increment value to construct a set of harmonic drop peaks; Step S4-5-2: When the sum does not exceed the upper limit of electrical quantity fluctuation, the voltage characteristic increment value corresponding to the current moment is set to 0 by default and stored in the harmonic drop peak set. One electrical quantity data acquisition cycle corresponds to one harmonic drop peak set. Step S4-6: When the real-time obtained comprehensive voltage characteristic value does not exceed the ideal comprehensive voltage characteristic value, subtract the comprehensive voltage characteristic value from the comprehensive voltage characteristic drop value, and compare the resulting difference with the lower limit value of electrical quantity fluctuation. The specific process is as follows: Step S4-6-1: When the difference does not exceed the lower limit of electrical quantity fluctuation, subtract the difference from the lower limit of electrical quantity fluctuation to obtain the voltage characteristic deficit difference. Select the current timestamp data as the index and the voltage characteristic deficit value as the value to store and construct a set of harmonic drop valleys. Step S4-6-2: When the difference exceeds the lower limit of electrical quantity fluctuation, the voltage characteristic deficit value corresponding to the current moment is set to 0 by default and stored in the harmonic drop valley set. One electrical quantity data acquisition cycle corresponds to one harmonic drop valley set. In practice, the maximum values ​​of the first and second harmonic voltage drops are weighted and fused to obtain the comprehensive voltage drop value. By comparing the deviation between the real-time voltage characteristic comprehensive value and the ideal value, and combining the electrical quantity fluctuation limit, such as the voltage amplitude fluctuation limit being 105% of the rated value, the high and low valley sets are screened to locate the voltage anomaly moment. The ideal three-phase voltage data should be set based on the grid rated parameters, and the fluctuation limit must meet the national standard requirements. For example, when the voltage characteristic comprehensive value exceeds the ideal value by 20% at a certain moment, if the sum of the comprehensive voltage drop values ​​exceeds the upper limit, it is stored in the peak value set.

[0024] Step S5: Analyze and process the set of harmonic drop troughs and the set of harmonic drop peaks in the fault monitoring sliding window using a time series prediction algorithm, predict the set of harmonic drop troughs and the set of harmonic drop peaks for the next electrical quantity data acquisition cycle of the target tension section, and monitor the target tension section based on the prediction results. Step S5-1: Obtain the set of harmonic drop troughs and the set of harmonic drop peaks generated by the fault monitoring sliding window over multiple historical sliding steps; Step S5-2: Use the time series prediction algorithm to process each historical harmonic drop trough set to obtain the harmonic drop trough set generated by the next step of the fault monitoring sliding window, which is denoted as the predicted trough set. Step S5-3: Use the time series prediction algorithm to process the historical harmonic drop peak sets respectively, and obtain the harmonic drop peak set generated by the next step of the fault monitoring sliding window, which is denoted as the predicted peak set. Step S5-4: Calculate the proportion of 0 based on the ratio of the number of data points with a value of 0 in each historical harmonic drop trough set to the total number of data points in the corresponding harmonic drop trough set within the sliding step. Step S5-5: Calculate the proportion of 0 based on the ratio of the number of data points with a value of 0 in each historical set of harmonic drop peak values ​​to the total number of data points in the set of harmonic drop peak values ​​within the corresponding sliding step. Steps S5-6: Set the trough threshold according to the proportion of 0 in the historical harmonic drop trough set, and set the peak threshold according to the proportion of 0 in the historical harmonic drop peak set. Step S5-7: Calculate the proportion of 0 in the predicted trough set, and record it as the predicted trough value; and simultaneously calculate the proportion of 0 in the predicted peak set, and record it as the predicted peak value. Step S5-8: Monitor the target tension section based on the prediction results. The specific process is as follows: Step S5-8-1: When the predicted low value does not exceed the low value threshold, it is determined that there is an abnormal voltage shortage in the target tension section, and a fault warning signal is issued; when the predicted low value exceeds the low value threshold, it is determined that the voltage of the target tension section is normal. Step S5-8-2: When the predicted peak value exceeds the peak value threshold, it is determined that there is an abnormal voltage increment in the target tension section, and a fault warning signal is issued; when the predicted peak value does not exceed the peak value threshold, it is determined that the voltage of the target tension section is normal. Step S5-9: Obtain the information of the tension section equipment that issued the fault warning signal, and transmit the tension section equipment information to the OTA upgrade terminal through power line carrier communication, so that the OTA upgrade terminal can perform status identification and data interaction for the upgrade process of smart grid equipment based on the tension section equipment information.

[0025] In practical implementation, time series prediction algorithms such as ARIMA are used. Based on the harmonic drop trough set and harmonic drop peak set obtained from multiple historical sliding step sizes in step S5-1, steps S5-2 and S5-3 are executed respectively to obtain the predicted trough set and predicted peak set. Steps S5-4 and S5-5 calculate the proportion of data points with a value of 0 in each historical harmonic drop trough set and peak set. Based on the calculation results, combined with the power grid equipment operation standards and historical data fluctuation patterns, trough thresholds and peak thresholds are set in step S5-6. The proportion of 0 in the predicted trough set and predicted peak set is calculated, i.e., the predicted trough value and predicted peak value in step S5-7. According to the judgment logic in step S5-8, when the predicted trough value does not exceed the trough threshold or the predicted peak value exceeds the peak threshold, it is determined that there is a voltage anomaly in the target tension section and a fault warning signal is issued. The system transmits the unique identifier of the tension section equipment that issued the fault warning signal, the anomaly type, the anomaly degree, and other equipment information to the equipment management module of the OTA upgrade terminal in real time. Based on the device information, the OTA upgrade terminal marks the smart grid equipment with abnormal voltage as "upgrade temporarily suspended" and interrupts the upgrade process to avoid upgrade failure or equipment damage due to voltage instability. Once the device voltage returns to normal, the OTA upgrade terminal updates the status according to the device information and restarts the upgrade process, realizing closed-loop collaboration between hardware fault monitoring and software upgrade management.

[0026] Example 2, as Figure 2 As shown, the present invention provides an Internet of Things-based smart grid equipment fault monitoring system. The smart management system includes a data acquisition module, a weather correction module, a sliding analysis module, a feature calculation module, and a prediction and early warning module. The data acquisition module is used to acquire electrical quantity data of smart grid transmission lines and construct an analysis set; the meteorological correction module is used to correct the analysis set by combining meteorological data; the sliding analysis module is used to set parameters and calculate harmonic drop values; the feature calculation module is used to calculate voltage feature values ​​and filter data; the prediction and early warning module is used to predict data and determine faults. The output of the data acquisition module is electrically connected to the input of the meteorological correction module; the output of the meteorological correction module is electrically connected to the input of the sliding analysis module; the output of the sliding analysis module is electrically connected to the input of the feature calculation module; and the output of the feature calculation module is electrically connected to the input of the forecast and early warning module. The data acquisition module includes a power acquisition unit and a harmonic set construction unit; the power acquisition unit is used to install sensors to acquire electrical quantity data and transmit it to the terminal; the harmonic set construction unit is used to process voltage signal data to construct a first harmonic drop analysis set; The meteorological correction module includes a meteorological parameter acquisition unit and a coupled correction calculation unit; the meteorological parameter acquisition unit is used to install meteorological sensors to acquire regional meteorological data; the coupled correction calculation unit is used to construct a set of weight factor correction analysis for model calculation. The sliding analysis module includes a window parameter setting unit and a harmonic drop calculation unit; the window parameter setting unit is used to set the sliding window and analysis time interval parameters; the harmonic drop calculation unit is used to calculate the maximum values ​​of the first and second harmonic drops; The feature calculation module includes a comprehensive value calculation unit and a data filtering and construction unit; the comprehensive value calculation unit is used to calculate the comprehensive voltage feature drop value and the current comprehensive value; the data filtering and construction unit is used to construct the set of harmonic drop peaks and valleys; The prediction and early warning module includes a time series prediction unit and a fault threshold judgment unit; the time series prediction unit is used to predict the harmonic drop set data for the next step; the fault threshold judgment unit is used to set the threshold and determine whether to issue an early warning signal.

[0027] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A method for fault monitoring of smart grid equipment based on the Internet of Things, characterized in that: The smart grid equipment fault monitoring method includes the following steps: Step S1: Select any tension section in the smart grid transmission line as the research object, denoted as the target tension section, and set the electrical quantity data acquisition cycle to continuously monitor and obtain the electrical quantity data of the target tension section to construct the first harmonic drop analysis set; Step S2: Based on the electrical quantity data acquisition cycle, acquire meteorological data of the target tension section in real time, and use Fisher's information processing method to combine meteorological data for quantitative analysis to construct a meteorological coupling correction model. Analyze and process the first harmonic drop analysis set through the meteorological coupling correction model to obtain the second harmonic drop analysis set. Step S3: Set the fault monitoring sliding window and harmonic analysis time interval. Analyze and process the second harmonic drop analysis set through the harmonic analysis time interval to obtain the maximum value of the first harmonic drop of the target tension section. Analyze and process the fault monitoring sliding window through the harmonic analysis time interval to obtain the maximum value of the second harmonic drop of the target tension section. Step S4: Calculate the comprehensive voltage characteristic drop value by weighted fusion based on the maximum value of the first harmonic drop and the maximum value of the second harmonic drop; collect electrical quantity data of the target tension section in real time, and combine the electrical quantity fluctuation limit in the smart grid transmission line to screen out the electrical quantity data that meet the conditions in the fault monitoring sliding window to construct the harmonic drop trough set and the harmonic drop peak set; Step S5: Analyze and process the set of harmonic drop troughs and peak values ​​in the fault monitoring sliding window using a time series prediction algorithm to predict the set of harmonic drop troughs and peak values ​​for the next electrical quantity data acquisition cycle of the target tension section, and monitor the target tension section based on the prediction results.

2. The method for fault monitoring of smart grid equipment based on the Internet of Things according to claim 1, characterized in that: The specific steps of step S1 are as follows: Step S1-1: Install power sensors on the towers at both ends of the target tension section to acquire electrical quantity data of the smart grid on the target tension section, and extract the original three-phase voltage signal data of the target tension section based on the electrical quantity data. Step S1-2: Transmit the raw three-phase voltage signal data to the data service terminal via power line carrier communication. The data service terminal is an integrated data hub with the ability to receive, process, store, and transmit data. Steps S1-3: Utilize the original three-phase voltage signal data from Discrete Fourier Transform and extract the amplitude and phase data of each harmonic component. Combine the amplitude difference data and phase difference data of each harmonic component into data units, arrange them in chronological order, and construct the first harmonic drop analysis set.

3. The method for fault monitoring of smart grid equipment based on the Internet of Things according to claim 2, characterized in that: The specific steps of step S2 are as follows: Step S2-1: Install meteorological sensors on the target tension section to acquire meteorological data in the area. The meteorological sensors include wind speed and direction sensors, temperature and humidity sensors, rain gauge sensors, and atmospheric pressure sensors. Transmit the acquired meteorological data to the data service terminal via power line carrier communication. Step S2-2: Construct a meteorological coupling correction model using Fisher's information processing method. Then, construct a multidimensional joint probability density function for the meteorological data uploaded during each electrical quantity data acquisition cycle using the meteorological coupling correction model. Step S2-3: Calculate the parameters contained in each set of meteorological data using the Fisher information matrix in the meteorological coupling correction model, and determine the coupling coefficients between the meteorological parameters. The parameters include wind speed parameters, temperature and humidity parameters, rainfall parameters, and atmospheric pressure parameters. Step S2-4: Use the calculated coupling coefficients as weighting factors to perform weighted correction on each set of data units in the first harmonic drop analysis set. After weighted correction calculation, the second harmonic drop analysis set of the target tension section is obtained.

4. The method for fault monitoring of smart grid equipment based on the Internet of Things according to claim 3, characterized in that: The specific steps of step S3 are as follows: Step S3-1: Set the fault monitoring sliding window to n consecutive electrical quantity data acquisition cycles; set the harmonic analysis time interval according to the electrical quantity data acquisition frequency of the target tension section within the electrical quantity data acquisition cycle; Step S3-2: Select the electrical quantity data acquisition cycle as the basic unit of the sliding step size of the fault monitoring sliding window, and perform point-by-point sliding processing on the second harmonic drop analysis set; Step S3-3: Select the harmonic analysis time interval as the basic unit, calculate the amplitude difference slope and phase difference slope of the corresponding harmonic components in the second harmonic drop analysis set, and after normalizing the amplitude difference slope and phase difference slope, calculate the comprehensive slope of a single harmonic analysis time interval through weighted fusion; select the maximum comprehensive slope obtained from each harmonic analysis time interval in the second harmonic drop analysis set, and use the comprehensive slope corresponding to the start time of the harmonic analysis time interval minus the comprehensive slope corresponding to the end time to obtain the absolute value of the difference as the first harmonic drop maximum value, which is denoted as the first harmonic drop maximum value. Step S3-4: Simultaneously using the calculation methods of amplitude difference slope and phase difference slope of harmonic components in step S3-3, analyze and calculate the harmonic components within the fault monitoring sliding window to obtain the maximum value of the second harmonic drop of the target tension section.

5. The method for fault monitoring of smart grid equipment based on the Internet of Things according to claim 4, characterized in that: The specific steps of step S4 are as follows: Step S4-1: Based on the weighted fusion calculation, combine the maximum value of the first harmonic drop and the maximum value of the second harmonic drop of the target tension section to obtain the comprehensive voltage characteristic drop value; Step S4-2: Extract the raw three-phase voltage signal data based on the real-time collected electrical quantity data of the target tension section, and analyze and process the data through the data service terminal to obtain the three-phase voltage data, which includes voltage amplitude data and voltage phase data. Step S4-3: Normalize the voltage amplitude data and voltage phase data of the target tension section in real time, and simultaneously select the weighting coefficient used in the comprehensive slope calculation in step S3-3 to calculate the comprehensive voltage characteristic value of the target tension section at the current moment.

6. The method for fault monitoring of smart grid equipment based on the Internet of Things according to claim 5, characterized in that: Step S4 also includes: Step S4-4: Obtain ideal three-phase voltage data and electrical quantity fluctuation limits in the smart grid transmission line. The electrical quantity fluctuation limits include upper and lower limits of electrical quantity fluctuation. Based on the ideal three-phase voltage data, calculate the comprehensive value of ideal voltage characteristics using the weighting coefficients used in the comprehensive slope calculation in step S3-3. Step S4-5: When the real-time obtained comprehensive voltage characteristic value exceeds the ideal comprehensive voltage characteristic value, the comprehensive voltage characteristic value is added to the comprehensive voltage characteristic difference value, and the sum is compared with the upper limit of electrical quantity fluctuation. The specific process is as follows: Step S4-5-1: When the sum exceeds the upper limit of electrical quantity fluctuation, use the sum minus the upper limit of electrical quantity fluctuation to obtain the voltage characteristic increment value, select the current timestamp data as the index, and store the voltage characteristic increment value to construct a set of harmonic drop peaks; Step S4-5-2: When the sum does not exceed the upper limit of electrical quantity fluctuation, the voltage characteristic increment value corresponding to the current moment is set to 0 by default and stored in the harmonic drop peak set. One electrical quantity data acquisition cycle corresponds to one harmonic drop peak set. Step S4-6: When the real-time obtained comprehensive voltage characteristic value does not exceed the ideal comprehensive voltage characteristic value, subtract the comprehensive voltage characteristic value from the comprehensive voltage characteristic drop value, and compare the resulting difference with the lower limit value of electrical quantity fluctuation. The specific process is as follows: Step S4-6-1: When the difference does not exceed the lower limit of electrical quantity fluctuation, subtract the difference from the lower limit of electrical quantity fluctuation to obtain the voltage characteristic deficit difference. Select the current timestamp data as the index and the voltage characteristic deficit value as the value to store and construct a set of harmonic drop valleys. Step S4-6-2: When the difference exceeds the lower limit of electrical quantity fluctuation, the voltage characteristic underestimation value corresponding to the current moment is set to 0 by default and stored in the harmonic drop valley set. One electrical quantity data acquisition cycle corresponds to one harmonic drop valley set.

7. The method for fault monitoring of smart grid equipment based on the Internet of Things according to claim 6, characterized in that: The specific steps of step S5 are as follows: Step S5-1: Obtain the set of harmonic drop troughs and the set of harmonic drop peaks generated by the fault monitoring sliding window over multiple historical sliding steps; Step S5-2: Use the time series prediction algorithm to process each historical harmonic drop trough set to obtain the harmonic drop trough set generated by the next step of the fault monitoring sliding window, which is denoted as the predicted trough set. Step S5-3: Use the time series prediction algorithm to process the historical harmonic drop peak sets respectively, and obtain the harmonic drop peak set generated by the next step of the fault monitoring sliding window, which is denoted as the predicted peak set. Step S5-4: Calculate the proportion of 0 based on the ratio of the number of data points with a value of 0 in each historical harmonic drop trough set to the total number of data points in the corresponding harmonic drop trough set within the sliding step. Step S5-5: Calculate the proportion of 0 based on the ratio of the number of data points with a value of 0 in each historical set of harmonic drop peak values ​​to the total number of data points in the set of harmonic drop peak values ​​within the corresponding sliding step. Steps S5-6: Set the trough threshold according to the proportion of 0 in the historical harmonic drop trough set, and set the peak threshold according to the proportion of 0 in the historical harmonic drop peak set. Step S5-7: Calculate the proportion of 0 in the predicted trough set, and record it as the predicted trough value; and simultaneously calculate the proportion of 0 in the predicted peak set, and record it as the predicted peak value. Step S5-8: Monitor the target tension section based on the prediction results. The specific process is as follows: Step S5-8-1: When the predicted low value does not exceed the low value threshold, it is determined that there is an abnormal voltage shortage in the target tension section, and a fault warning signal is issued; when the predicted low value exceeds the low value threshold, it is determined that the voltage of the target tension section is normal. Step S5-8-2: When the predicted peak value exceeds the peak value threshold, it is determined that there is an abnormal voltage increment in the target tension section, and a fault warning signal is issued; when the predicted peak value does not exceed the peak value threshold, it is determined that the voltage of the target tension section is normal. Step S5-9: Obtain the information of the tension section equipment that issued the fault warning signal, and transmit the tension section equipment information to the OTA upgrade terminal through power line carrier communication, so that the OTA upgrade terminal can perform status identification and data interaction for the upgrade process of smart grid equipment based on the tension section equipment information.

8. A smart grid equipment fault monitoring system based on the Internet of Things (IoT), which is applied to the smart grid equipment fault monitoring method based on the Internet of Things (IoT) as described in any one of claims 1-7, characterized in that: The smart grid equipment fault monitoring system includes a data acquisition module, a weather correction module, a sliding analysis module, a feature calculation module, and a prediction and early warning module. The data acquisition module is used to acquire electrical quantity data of smart grid transmission lines and construct an analysis set; the meteorological correction module is used to correct the analysis set in combination with meteorological data; the sliding analysis module is used to set parameters and calculate harmonic drop values. The feature calculation module is used to calculate voltage feature values ​​and filter data; the prediction and early warning module is used to predict data and determine faults. The output of the data acquisition module is electrically connected to the input of the meteorological correction module; the output of the meteorological correction module is electrically connected to the input of the sliding analysis module; the output of the sliding analysis module is electrically connected to the input of the feature calculation module; and the output of the feature calculation module is electrically connected to the input of the forecast and early warning module.

9. A smart grid equipment fault monitoring system based on the Internet of Things according to claim 8, characterized in that: The data acquisition module includes a power acquisition unit and a harmonic set construction unit; the power acquisition unit is used to install sensors to acquire electrical quantity data and transmit it to the terminal; the harmonic set construction unit is used to process voltage signal data to construct a first harmonic drop analysis set; The meteorological correction module includes a meteorological parameter acquisition unit and a coupled correction calculation unit; The meteorological parameter acquisition unit is used to install meteorological sensors to acquire regional meteorological data; the coupling correction calculation unit is used to construct a set of model calculation weight factor correction analysis. The sliding analysis module includes a window parameter setting unit and a harmonic drop calculation unit; the window parameter setting unit is used to set the sliding window and analysis time interval parameters. The harmonic drop calculation unit is used to calculate the maximum values ​​of the first and second harmonic drops.

10. A smart grid equipment fault monitoring system based on the Internet of Things according to claim 8, characterized in that: The feature calculation module includes a comprehensive value calculation unit and a data filtering and construction unit; the comprehensive value calculation unit is used to calculate the comprehensive voltage feature drop value and the current comprehensive value; the data filtering and construction unit is used to construct the set of harmonic drop peaks and valleys; The prediction and early warning module includes a time series prediction unit and a fault threshold judgment unit; the time series prediction unit is used to predict the harmonic drop set data for the next step; the fault threshold judgment unit is used to set the threshold and determine whether to issue an early warning signal.

Images