Loop closing monitoring system for low-voltage power distribution network based on IoT

Abstract

The invention discloses an IoT (Internet of Things)-based low-voltage power distribution network loop closing monitoring system, which belongs to the technical field of power distribution network monitoring, and can ensure that the power grid states of two transformer areas are compatible before loop closing operation by monitoring key electric parameters such as voltage difference and phase angle difference in real time. Therefore, loop closing failure or power grid accidents caused by mismatching of the electric parameters are avoided. By automatically calculating the loop closing impact current, the steady-state current, the phase angle difference and the distribution transformer load rate difference, the system can provide a scientific judgment basis for operators, so that the loop closing operation is more accurate and reliable. Operators can know the state of the power grid in real time through the remote on-line monitoring system, so that the working efficiency is improved, and the safety risk of the operators is reduced at the same time. Through digital support, the system can quickly respond to the change of the power grid, timely adjust the loop closing operation, and improve the operation efficiency and power supply reliability of the power grid. The power failure time caused by loop closing operation is reduced through real-time monitoring and rapid judgment, and the power utilization experience of a user is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of distribution network monitoring, and specifically to an IoT-based low-voltage distribution network closed-loop monitoring system. Background Art

[0002] The quality of life of the people is closely related to safe and stable power supply. At present, the low-voltage distribution network in urban areas has the characteristics of high load density, concentrated power consumption, and high power supply reliability requirements. In order to further reduce the user's perception of power outages, low-voltage ring-closing power conversion technology has become an important research topic. With the rapid development of Internet of Things (IoT) technology, intelligent monitoring and control of low-voltage distribution networks have become an important part of the modernization of power systems.

[0003] However, the low-voltage distribution network has problems such as complex substation topology and chaotic phase sequence of outgoing lines. Operators are often unable to judge in advance whether the low-voltage loop closing conditions are met. In addition, at the moment of loop closing, a large impact current may be generated, affecting the stable operation of the upper power supply, or a large circulating current may be generated after the loop closing is stable. This may cause electrical equipment or lines to overload, and even cause the relay protection device to trip, thus seriously affecting the safe operation of the entire power grid. Summary of the invention

[0004] The purpose of the present invention is to provide an IoT-based low-voltage distribution network closed-loop monitoring system in order to solve the above-mentioned problems.

[0005] The technical solution adopted by the present invention is as follows: a closed-loop monitoring system for low-voltage distribution network based on IoT, characterized in that: the system comprises: a perception layer module, a communication network layer module and an application layer module;

[0006] The perception layer module is internally provided with a sensor module and a processor module;

[0007] The communication network layer module is internally provided with a wireless communication network and a server;

[0008] The application layer module is internally provided with a client module and a data analysis module

[0009] The sensor module is responsible for collecting current, voltage and phase electrical parameter data, and an edge computing unit is provided inside the sensor module;

[0010] The processor module processes and analyzes the data collected by the sensor and controls the loop closing operation;

[0011] Communication module: transmits the data processed by the processor to the network layer;

[0012] The wireless communication network connects the perception layer module and the application layer module and is responsible for data transmission;

[0013] The server receives data from the perception layer module, stores, processes and analyzes it, and transmits the result to the application layer module;

[0014] The client module receives data transmitted by the server and displays it to the user, allowing the user to perform a loop closing operation;

[0015] The data analysis module analyzes the data transmitted by the server and provides a visual display.

[0016] In a preferred embodiment, the sensor module uses a Hall sensor, which is installed at the outlet of the distribution transformer to collect current data in real time;

[0017] The sensor module uses a voltage transformer, which is installed at the outlet of the distribution transformer to collect voltage data in real time;

[0018] The sensor module uses a digital phase measuring instrument, which is installed at the outlet of the distribution transformer to collect phase data of voltage or current in real time;

[0019] The sensor module is used for sensors of 0.4kV low-voltage distribution network through a real-time acquisition unit, including high-precision current transformer, voltage transformer and phase sensor, for real-time acquisition of electrical parameters of low-voltage outgoing lines of two substations, including voltage difference and phase angle difference.

[0020] In a preferred embodiment, the processor module uses an STM32 single-chip microcomputer, which is responsible for data processing, control logic and communication functions; filters, amplifies and converts current, voltage and phase data to remove noise and improve data accuracy and reliability; the processor module determines whether the closing conditions are met according to the closing model and preset conditions, including the pressure difference is less than 5% and the phase angle difference is less than 2°, and controls the closing operation, including closing and opening;

[0021] The processor module transmits the processed data to the server through the 5G module and receives control instructions from the server.

[0022] In a preferred embodiment, the communication module transmits the data processed by the processor module to the server, and receives control instructions from the server.

[0023] In a preferred embodiment, the wireless communication network uses the 5G network of the mobile operator to connect the perception layer module and the server to achieve wireless transmission of data;

[0024] The server deploys the closing loop judgment logic, and automatically calculates whether the closing loop conditions are met based on the real-time collected pressure difference, phase angle difference, closing loop impact current, steady-state current and load rate difference data of the distribution transformers on both sides. When it is monitored that the closing loop conditions are not met, the client module can issue an early warning message to remind the operating personnel to pay attention to safety.

[0025] In a preferred embodiment, the sensor interface of the edge computing unit is responsible for connecting with the current, voltage, and phase sensors to receive data collected by the sensors;

[0026] The edge computing unit converts the analog signal collected by the sensor into a digital signal through an analog-to-digital converter for subsequent processing;

[0027] The edge computing unit filters the digital signal through a filter to remove noise and improve data accuracy.

[0028] In a preferred embodiment, the edge computing unit performs an abnormality detection module through a current overload detection algorithm, specifically including:

[0029] S1: Data acquisition: The current sensor collects current data and transmits it to the data acquisition module;

[0030] S2: Analog-to-digital conversion: The data acquisition module converts the analog current signal into a digital signal;

[0031] S3: Filtering: The data processing module filters the digital current signal to remove noise;

[0032] S4: Effective value calculation: The data processing module calculates the effective value of the current, and the formula is the same as described above;

[0033] S5: Current change rate calculation: The data processing module calculates the change rate of the effective value of the current at two adjacent sampling points. The formula is as follows:

[0034] dI_eff=(I_eff(t+Δt)-I_eff(t)) / Δt;

[0035] Wherein, dI_eff is the current change rate, unit: ampere / second (A / s), I_eff(t+Δt) is the effective value of the current at time t+Δt, I_eff(t) is the effective value of the current at time t, Δt is the sampling time interval, unit: second (s);

[0036] S6: Threshold determination: The abnormality detection module compares the calculated current change rate with a preset current change rate threshold;

[0037] S7: Early warning: If the current change rate exceeds the threshold, the early warning module issues a red early warning and transmits the early warning information to the server and client through the communication module.

[0038] In a preferred embodiment, the edge computing unit uses the ARIMA model in time series analysis to predict the closed-loop fault of the IoT low-voltage distribution network, and the calculation process includes:

[0039] S1: Data collection: Collect historical fault data of IoT low-voltage distribution network loop;

[0040] S2: Data preprocessing: Stabilize the low-voltage distribution network loop data and remove outliers;

[0041] The first-order difference formula for stabilizing the closed-loop data of the low-voltage distribution network and removing outliers is:

[0042] X′ t =X t -X t-1 ; where X t The value of the time series at time t, that is, the observed value at the current closed loop data time point; X t-1 : The value of the closed-loop data time series at time t-1, that is, the observation value of the previous closed-loop data time point;

[0043] The L lag operator, when applied to a closed loop data time series, k X t =X t-k , that is, L to the power of k acts on X t The value at time tk;

[0044] d: The order of difference, which indicates the number of times the closed loop data time series is differentiated to make the series stable

[0045] S3: Model identification: Use ACF and PACF plots to determine the parameters (p, d, q) of the ARIMA model, where ACF measures the correlation between the closed loop data time series and its lagged values; PACF: measures the partial correlation between the time series and its lagged values, removing the influence of the intermediate lagged terms;

[0046] The identification formula of the autocorrelation function (ACF) is:

[0047] ρ k is the autocorrelation coefficient of lag k; γ k is the autocovariance of lag k; γ0 is the variance of the series;

[0048] The identification formula of the partial autocorrelation function (PACF) is:

[0049] in:

[0050] φk is the partial autocorrelation coefficient at lag k;

[0051] φkj is the partial autocorrelation coefficient of lag j;

[0052] S4: Model estimation: The model parameters were estimated using the maximum likelihood estimation method;

[0053] The formula for solving the maximum value of the following log-likelihood function for the ARIMA model is:

[0054] where θ is the vector of model parameters; X is the observed data; T is the number of observations; σ2 is the variance of the error term; εt is the residual at time t;

[0055] S5: Model testing: residual analysis was used to test the validity of the model;

[0056] S6: Prediction: Use the model to predict future failures;

[0057] The ARIMA model is represented by ARIMA(p,d,q), where:

[0058] p is the order of the autoregressive term;

[0059] d is the number of differences;

[0060] q is the order of the moving average term;

[0061] The calculation formula of the ARIMA model includes:

[0062] The calculation formula for the autoregressive (AR) part is: Where Yt is the observed value at time point t; c is the constant term; Φi is the autoregressive coefficient; t is the white noise error term;

[0063] The calculation formula for the differential part is:

[0064] in is a difference operator;

[0065] B is the lag operator; BYt=Yt-1;

[0066] d is the number of differences;

[0067] The calculation formula for the moving average (MA) part is:

[0068] Where μ is the mean of the time series;

[0069] Θi is the moving average coefficient;

[0070] t and ti are error terms.

[0071] In a preferred embodiment, the client module displays the data collected by the perception layer module in the form of charts and curves, and allows the user to perform closing operations according to system prompts, including closing and opening the switch; at the same time, the client module also has an early warning function. When it is detected that the closing conditions are not met, an early warning message is issued to remind the user to pay attention to safety.

[0072] In a preferred embodiment, the data analysis module is combined with the edge computing unit, and uses the ARIMA model in time series analysis to perform trend analysis on current, voltage, and phase data to predict possible problems in the future, including loop failures for the 0.4kV low-voltage distribution network; the data analysis module displays the data analysis results in the form of charts and curves, so that users can intuitively understand the operating status of the 0.4kV low-voltage distribution network.

[0073] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0074] 1. In the present invention, the edge computing unit can process data instantly at the source of data generation, reducing the bandwidth and delay required for data transmission to the cloud or server, thereby achieving rapid response to electrical parameters such as current, voltage, and phase. Since the edge computing unit can perform preliminary data processing locally, only necessary data or analysis results need to be uploaded to the server, which greatly reduces the amount of data transmitted over the network and reduces the network load. By performing data processing on the edge computing unit, even if the network connection is unstable or interrupted, the system can still maintain basic monitoring and control functions locally, improving the reliability of the system. The edge computing unit can process sensitive data locally, reducing the risk of data transmission on the network and improving data security.

[0075] 2. In the present invention, the edge computing unit runs the current change rate detection algorithm and the ARIMA model of time series analysis. These algorithms can perform fault prediction and anomaly detection locally, so that the system can make intelligent decisions faster, customize data processing and analysis algorithms, and provide higher flexibility. Cost-effectiveness: By processing data at the edge, the dependence on the central server can be reduced, thereby reducing the cost of data storage and computing. The edge computing unit can receive control instructions from the server and perform closing operations, such as closing and opening, which makes remote operation possible and improves the manageability of the system. The introduction of the edge computing unit makes the IoT-based low-voltage distribution network closing monitoring system more efficient, reliable, and secure, and has better adaptability and economy.

[0076] 3. In the present invention, by real-time monitoring of key electrical parameters such as pressure difference and phase angle difference, the system can ensure that the power grid states of the two substations are compatible before the closing operation, thereby avoiding closing failure or power grid accidents caused by mismatching electrical parameters. By automatically calculating the closing impact current, steady-state current, phase angle difference and distribution transformer load rate difference, the system can provide scientific judgment basis for operators, making the closing operation more accurate and reliable. Operators can understand the power grid state in real time through the remote online monitoring system without on-site operation, which improves work efficiency and reduces personnel safety risks. Through digital support, the system can quickly respond to power grid changes and adjust the closing operation in time, thereby improving the operation efficiency and power supply reliability of the power grid. Real-time monitoring and rapid judgment reduce the power outage time caused by the closing operation and improve the user's power experience. By predicting and analyzing the possible impact of the closing operation, the system can detect potential faults in advance, take preventive measures, and avoid the occurrence of power grid accidents. Automated monitoring and judgment reduce manual intervention and reduce the maintenance cost of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] Figure 1 is the overall system block diagram of the present invention;

[0078] Figure 2 This is a schematic diagram of the edge computing unit process principle in the present invention. DETAILED DESCRIPTION

[0079] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0080] Reference Figure 1-2 ,

[0081] Based on IoT low-voltage distribution network closed-loop monitoring system, the system includes: perception layer module, communication network layer module and application layer module;

[0082] The perception layer module is internally provided with a sensor module and a processor module;

[0083] The communication network layer module is internally provided with a wireless communication network and a server;

[0084] The application layer module is internally configured with a client module and a data analysis module.

[0085] The sensor module is responsible for collecting electrical parameter data such as current, voltage, and phase, and an edge computing unit is set inside the sensor module.

[0086] The processor module processes and analyzes the data collected by the sensor and controls the closed-loop operation.

[0087] Communication module: transmits data processed by the processor to the network layer.

[0088] The wireless communication network connects the perception layer module and the application layer module and is responsible for data transmission.

[0089] The server receives data from the perception layer module, stores, processes and analyzes it, and transmits the results to the application layer module.

[0090] The client module receives the data transmitted by the server and displays it to the user, allowing the user to perform loop closing operations.

[0091] The data analysis module analyzes the data transmitted by the server and provides a visual display.

[0092] The sensor module uses a Hall sensor, which is installed at the output end of the distribution transformer to collect current data in real time.

[0093] The sensor module uses a voltage transformer and is installed at the output end of the distribution transformer to collect voltage data in real time.

[0094] The sensor module uses a digital phase measuring instrument, which is installed at the outlet of the distribution transformer to collect phase data of voltage or current in real time;

[0095] The sensor module is used for sensors of 0.4kV low-voltage distribution network through real-time acquisition unit, including high-precision current transformer, voltage transformer and phase sensor, which are used to collect the electrical parameters of low-voltage outgoing lines of two substations in real time, including voltage difference and phase angle difference.

[0096] The processor module uses an STM32 microcontroller, which is responsible for data processing, control logic and communication functions. It filters, amplifies and converts current, voltage and phase data to remove noise and improve data accuracy and reliability. The processor module determines whether the closing conditions are met based on the closing model and preset conditions, such as the pressure difference is less than 5% and the phase angle difference is less than 2°, and controls the closing operation, such as closing and opening.

[0097] The processor module transmits the processed data to the server through the 5G module and receives control instructions from the server.

[0098] The communication module transmits the data processed by the processor module to the server and receives control instructions from the server.

[0099] The wireless communication network uses the mobile operator's 5G network to connect the perception layer module and the server to achieve wireless transmission of data;

[0100] The server deploys the closing loop judgment logic, and automatically calculates whether the closing loop conditions are met based on the real-time collected data such as pressure difference, phase angle difference, closing loop impact current, steady-state current, and load rate difference of the distribution transformers on both sides. When it is monitored that the closing loop conditions are not met, the client module can issue an early warning message to remind the operating personnel to pay attention to safety.

[0101] Sensor interface of the edge computing unit: responsible for connecting with current, voltage, phase and other sensors, and receiving data collected by the sensors.

[0102] The edge computing unit converts the analog signals collected by the sensor into digital signals through an analog-to-digital converter for subsequent processing.

[0103] The edge computing unit filters the digital signal through a filter to remove noise and improve data accuracy.

[0104] The edge computing unit uses the current overload detection algorithm to perform an anomaly detection module, which specifically includes:

[0105] S1: Data acquisition: The current sensor collects current data and transmits it to the data acquisition module.

[0106] S2: Analog-to-digital conversion: The data acquisition module converts the analog current signal into a digital signal.

[0107] S3: Filtering: The data processing module filters the digital current signal to remove noise.

[0108] S4: Effective value calculation: The data processing module calculates the effective value of the current, and the formula is the same as before.

[0109] S5: Current change rate calculation: The data processing module calculates the change rate of the effective value of the current at two adjacent sampling points. The formula is as follows:

[0110] dI_eff=(I_eff(t+Δt)-I_eff(t)) / Δt;

[0111] Wherein, dI_eff is the current change rate, unit: ampere / second (A / s), I_eff(t+Δt) is the effective value of the current at time t+Δt, I_eff(t) is the effective value of the current at time t, Δt is the sampling time interval, unit: second (s).

[0112] S6: Threshold determination: The abnormality detection module compares the calculated current change rate with a preset current change rate threshold.

[0113] S7: Early warning: If the current change rate exceeds the threshold, the early warning module issues a red early warning and transmits the early warning information to the server and client through the communication module.

[0114] The edge computing unit uses the ARIMA model in time series analysis to predict IoT low-voltage distribution network loop failures. The calculation process includes:

[0115] S1: Data collection: Collect historical fault data of IoT low-voltage distribution network loop;

[0116] S2: Data preprocessing: Stabilize the low-voltage distribution network loop data and remove outliers;

[0117] The first-order difference formula for stabilizing the closed-loop data of the low-voltage distribution network and removing outliers is:

[0118] X′ t =X t -X t-1 ; where X t The value of the time series at time t, that is, the observed value at the current closed loop data time point; X t-1 : The value of the closed-loop data time series at time t-1, that is, the observation value of the previous closed-loop data time point;

[0119] The L lag operator, when applied to a closed loop data time series, k X t =X t-k , that is, L to the power of k acts on X t The value at time tk;

[0120] d: The order of difference, which indicates the number of times the closed loop data time series is differentiated to make the series stable

[0121] S3: Model identification: Use ACF and PACF plots to determine the parameters (p, d, q) of the ARIMA model, where ACF measures the correlation between the closed loop data time series and its lagged values; PACF: measures the partial correlation between the time series and its lagged values, removing the influence of the intermediate lagged terms;

[0122] The identification formula of the autocorrelation function (ACF) is:

[0123] ρ k is the autocorrelation coefficient of lag k; γ k is the autocovariance of lag k; γ0 is the variance of the series;

[0124] The identification formula of the partial autocorrelation function (PACF) is:

[0125] in:

[0126] φk is the partial autocorrelation coefficient at lag k;

[0127] φkj is the partial autocorrelation coefficient of lag j;

[0128] S4: Model estimation: The model parameters were estimated using the maximum likelihood estimation method;

[0129] The formula for solving the maximum value of the following log-likelihood function for the ARIMA model is:

[0130] where θ is the vector of model parameters; X is the observed data; T is the number of observations; σ2 is the variance of the error term; εt is the residual at time t;

[0131] S5: Model testing: residual analysis was used to test the validity of the model;

[0132] S6: Prediction: Use the model to predict future failures;

[0133] The ARIMA model is represented by ARIMA(p,d,q), where:

[0134] p is the order of the autoregressive term;

[0135] d is the number of differences;

[0136] q is the order of the moving average term;

[0137] The calculation formula of the ARIMA model includes:

[0138] The calculation formula for the autoregressive (AR) part is: Where Yt is the observed value at time point t; c is the constant term; Φi is the autoregressive coefficient; t is the white noise error term;

[0139] The calculation formula for the differential part is:

[0140] in is a difference operator;

[0141] B is the lag operator; BYt=Yt-1;

[0142] d is the number of differences;

[0143] The calculation formula for the moving average (MA) part is:

[0144] Where μ is the mean of the time series;

[0145] Θi is the moving average coefficient;

[0146] t and ti are error terms.

[0147] The client module displays the data collected by the perception layer module in the form of charts, curves, etc., and allows users to perform closing operations according to system prompts, such as closing and opening the circuit breaker, etc. At the same time, the client module also has an early warning function. When it detects that the closing conditions are not met, it issues an early warning message to remind users to pay attention to safety.

[0148] The data analysis module combines with the edge computing unit and uses the ARIMA model in time series analysis to perform trend analysis on current, voltage, phase and other data to predict possible problems in the future, including loop failures for the 0.4kV low-voltage distribution network. The data analysis module displays the data analysis results in the form of charts, curves, etc., so that users can intuitively understand the operating status of the 0.4kV low-voltage distribution network.

[0149] In the present invention, the edge computing unit can process data instantly at the source of data generation, reducing the bandwidth and delay required for data transmission to the cloud or server, thereby achieving a rapid response to electrical parameters such as current, voltage, and phase. Since the edge computing unit can perform preliminary data processing locally, only necessary data or analysis results need to be uploaded to the server, which greatly reduces the amount of data transmitted over the network and reduces the network load. By performing data processing on the edge computing unit, even if the network connection is unstable or interrupted, the system can still maintain basic monitoring and control functions locally, thereby improving the reliability of the system. The edge computing unit can process sensitive data locally, reducing the risk of data transmission over the network and improving data security.

[0150] In the present invention, the edge computing unit runs the current change rate detection algorithm and the ARIMA model of time series analysis. These algorithms can perform fault prediction and anomaly detection locally, so that the system can make intelligent decisions faster, customize data processing and analysis algorithms, and provide higher flexibility. Cost-effectiveness: By processing data at the edge, the dependence on the central server can be reduced, thereby reducing the cost of data storage and computing. The edge computing unit can receive control instructions from the server and perform closing operations, such as closing and opening, which makes remote operation possible and improves the manageability of the system. The introduction of the edge computing unit makes the IoT-based low-voltage distribution network closing monitoring system more efficient, reliable, and safe, and has better adaptability and economy.

[0151] In the present invention, by real-time monitoring of key electrical parameters such as pressure difference and phase angle difference, the system can ensure that the power grid states of the two substations are compatible before the closing operation, thereby avoiding closing failure or power grid accidents caused by mismatching electrical parameters. By automatically calculating the closing impact current, steady-state current, phase angle difference and distribution transformer load rate difference, the system can provide a scientific basis for judgment for the operator, making the closing operation more accurate and reliable. The operator can understand the power grid state in real time through the remote online monitoring system without on-site operation, which improves work efficiency and reduces personnel safety risks. Through digital support, the system can quickly respond to power grid changes and adjust the closing operation in time, thereby improving the operation efficiency and power supply reliability of the power grid. Real-time monitoring and rapid judgment reduce the power outage time caused by the closing operation and improve the user's power experience. By predicting and analyzing the possible impact of the closing operation, the system can detect potential faults in advance, take preventive measures, and avoid the occurrence of power grid accidents. Automated monitoring and judgment reduce manual intervention and reduce the maintenance cost of the power grid.

[0152] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0153] The above description enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. IoT-based low-voltage distribution network closed-loop monitoring system, characterized by: The system comprises: a perception layer module, a communication network layer module and an application layer module; The perception layer module is internally provided with a sensor module and a processor module; The communication network layer module is internally provided with a wireless communication network and a server; The application layer module is internally provided with a client module and a data analysis module The sensor module is responsible for collecting current, voltage and phase electrical parameter data, and an edge computing unit is provided inside the sensor module; The processor module processes and analyzes the data collected by the sensor and controls the loop closing operation; Communication module: transmits the data processed by the processor to the network layer; The wireless communication network connects the perception layer module and the application layer module and is responsible for data transmission; The server receives data from the perception layer module, stores, processes and analyzes it, and transmits the result to the application layer module; The client module receives data transmitted by the server and displays it to the user, allowing the user to perform a loop closing operation; The data analysis module analyzes the data transmitted by the server and provides a visual display.

2. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The sensor module uses a Hall sensor, which is installed at the outlet of the distribution transformer to collect current data in real time; The sensor module uses a voltage transformer, which is installed at the outlet of the distribution transformer to collect voltage data in real time; The sensor module uses a digital phase measuring instrument, which is installed at the outlet of the distribution transformer to collect phase data of voltage or current in real time; The sensor module is used for sensors of 0.4kV low-voltage distribution network through a real-time acquisition unit, including high-precision current transformer, voltage transformer and phase sensor, for real-time acquisition of electrical parameters of low-voltage outgoing lines of two substations, including voltage difference and phase angle difference.

3. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The processor module uses an STM32 single-chip microcomputer, which is responsible for data processing, control logic and communication functions; it filters, amplifies and converts current, voltage and phase data to remove noise and improve data accuracy and reliability; the processor module determines whether the closing conditions are met according to the closing model and preset conditions, including the pressure difference is less than 5% and the phase angle difference is less than 2°, and controls the closing operation, including closing and opening; The processor module transmits the processed data to the server through the 5G module and receives control instructions from the server.

4. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The communication module transmits the data processed by the processor module to the server and receives control instructions from the server.

5. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The wireless communication network uses the 5G network of the mobile operator to connect the perception layer module and the server to achieve wireless transmission of data; The server deploys the closing loop judgment logic, and automatically calculates whether the closing loop conditions are met based on the real-time collected pressure difference, phase angle difference, closing loop impact current, steady-state current and load rate difference data of the distribution transformers on both sides. When it is monitored that the closing loop conditions are not met, the client module can issue an early warning message to remind the operating personnel to pay attention to safety.

6. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The sensor interface of the edge computing unit is responsible for connecting with the current, voltage and phase sensors and receiving the data collected by the sensors; The edge computing unit converts the analog signal collected by the sensor into a digital signal through an analog-to-digital converter for subsequent processing; The edge computing unit filters the digital signal through a filter to remove noise and improve data accuracy.

7. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The edge computing unit performs an abnormality detection module through a current overload detection algorithm, specifically including: S1: Data acquisition: The current sensor collects current data and transmits it to the data acquisition module; S2: Analog-to-digital conversion: The data acquisition module converts the analog current signal into a digital signal; S3: Filtering: The data processing module filters the digital current signal to remove noise; S4: Effective value calculation: The data processing module calculates the effective value of the current, and the formula is the same as described above; S5: Current change rate calculation: The data processing module calculates the change rate of the effective value of the current at two adjacent sampling points. The formula is as follows: dI_eff=(I_eff(t+Δt)-I_eff(t)) / Δt; Wherein, dI_eff is the current change rate, unit: ampere / second (A / s), I_eff(t+Δt) is the effective value of the current at time t+Δt, I_eff(t) is the effective value of the current at time t, Δt is the sampling time interval, unit: second (s); S6: Threshold determination: The abnormality detection module compares the calculated current change rate with a preset current change rate threshold; S7: Early warning: If the current change rate exceeds the threshold, the early warning module issues a red early warning and transmits the early warning information to the server and client through the communication module.

8. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The edge computing unit uses the ARIMA model in time series analysis to predict IoT low-voltage distribution network loop failures. The calculation process includes: S1: Data collection: Collect historical fault data of IoT low-voltage distribution network loop; S2: Data preprocessing: Stabilize the low-voltage distribution network loop data and remove outliers; The first-order difference formula for stabilizing the closed-loop data of the low-voltage distribution network and removing outliers is: t =X t -X t-1 ; where X t The value of the time series at time t, that is, the observed value at the current closed loop data time point; X t-1 : The value of the closed-loop data time series at time t-1, that is, the observation value of the previous closed-loop data time point; The L lag operator, when applied to a closed loop data time series, k X t =X t-k , that is, L to the power of k acts on X t The value at time tk; d: The order of difference, which indicates the number of times the closed loop data time series is differentiated to make the series stable S3: Model identification: Use ACF and PACF plots to determine the parameters (p, d, q) of the ARIMA model, where ACF measures the correlation between the closed loop data time series and its lagged values; PACF: measures the partial correlation between the time series and its lagged values, removing the influence of the intermediate lagged terms; The identification formula of the autocorrelation function (ACF) is: ρ k is the autocorrelation coefficient of lag k; γ k is the autocovariance of lag k; γ0 is the variance of the series; The identification formula of the partial autocorrelation function (PACF) is: in: φk is the partial autocorrelation coefficient at lag k; φkj is the partial autocorrelation coefficient of lag j; S4: Model estimation: The model parameters were estimated using the maximum likelihood estimation method; The formula for solving the maximum value of the following log-likelihood function for the ARIMA model is: where θ is the vector of model parameters; X is the observed data; T is the number of observations; σ2 is the variance of the error term; εt is the residual at time t; S5: Model testing: residual analysis was used to test the validity of the model; S6: Prediction: Use the model to predict future failures; The ARIMA model is represented by ARIMA(p,d,q), where: p is the order of the autoregressive term; d is the number of differences; q is the order of the moving average term; The calculation formula of the ARIMA model includes: The calculation formula for the autoregressive (AR) part is: Where Yt is the observed value at time point t; c is the constant term; Φi is the autoregressive coefficient; t is the white noise error term; The calculation formula for the differential part is: in is a difference operator; B is the lag operator; BYt=Yt-1; d is the number of differences; The calculation formula for the moving average (MA) part is: Where μ is the mean of the time series; Θi is the moving average coefficient; t and ti are error terms.

9. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The client module displays the data collected by the perception layer module in the form of charts and curves, and allows the user to perform closing operations according to the system prompts, including closing and opening the switch; at the same time, the client module also has an early warning function. When it is detected that the closing conditions are not met, an early warning message is issued to remind the user to pay attention to safety.

10. The IoT-based low-voltage distribution network closed-loop monitoring system according to claim 1, characterized in that: The data analysis module is combined with the edge computing unit and uses the ARIMA model in time series analysis to perform trend analysis on current, voltage, and phase data to predict possible problems in the future, including loop failures for the 0.4kV low-voltage distribution network; the data analysis module displays the data analysis results in the form of charts and curves.

Images