Method for identifying operating state of equipment based on intelligent circuit breaker and intelligent circuit breaker

Abstract

The present application relates to the technical field of intelligent circuit breaker, especially a device running state recognition method based on intelligent circuit breaker and intelligent circuit breaker; when data collected by the intelligent circuit breaker is synchronized, first, a time stamp is generated for the data collected by each sensor; the collected data is encapsulated into a data packet together with the time stamp, and is sent to a data processing unit integrated with the intelligent circuit breaker; then the data processing unit performs data alignment operation on the data packet; wherein the improved spline interpolation method is used for data alignment operation; and a physical model constraint condition is established, a linear equation set is constructed according to the physical model constraint condition, an interpolation coefficient is solved, and a cubic polynomial function is determined; then time-aligned data points are generated according to the cubic polynomial function; by introducing the physical model constraint condition, it is ensured that the interpolation result conforms to the actual physical process, and the reliability of subsequent analysis and state recognition is improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent circuit breaker technology, and in particular to a method for identifying the operating status of equipment based on an intelligent circuit breaker and an intelligent circuit breaker. Background Technology

[0002] In recent years, with the development of the Internet of Things (IoT) and artificial intelligence technologies, smart circuit breakers have gradually become an important development direction in the field of industrial automation. Smart circuit breakers not only possess the basic protection functions of traditional circuit breakers, but also integrate various sensors and intelligent processing units, enabling real-time monitoring of equipment operating status and achieving fault early warning and intelligent control through data analysis.

[0003] However, existing smart circuit breakers typically rely on a central clock signal from the smart sensors to synchronize the acquisition actions of all sensors when synchronizing multiple types of collected data. While this method ensures time consistency in data acquisition, it also has limitations: the clocks of different sensors may drift due to factors such as temperature and power fluctuations, leading to inconsistent data acquisition times; in distributed systems, communication delays between sensors and the central processing unit may affect the accuracy of data synchronization; and if the central clock signal malfunctions, the synchronization mechanism of the entire system will fail. Meanwhile, existing technologies also employ interpolation algorithms for data synchronization, such as cubic spline interpolation; however, these interpolation algorithms may cause distortion of the interpolation curve, thus affecting the data synchronization effect.

[0004] Meanwhile, existing technologies generally use artificial neural network models to identify the operating status of equipment. In artificial neural network models, ReLU activation function is usually selected as the activation function, which may lead to the dead ReLU problem and affect the accuracy of identification. Summary of the Invention

[0005] The purpose of this invention is to provide a method for identifying the operating status of equipment based on a smart circuit breaker and a smart circuit breaker, so as to solve the problems mentioned in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution:

[0007] A method for identifying the operating status of equipment based on a smart circuit breaker includes the following steps:

[0008] S1: The current, voltage, and 2nd-50th harmonic data of the device are collected using the sensors integrated in the intelligent circuit breaker;

[0009] S2: Perform data synchronization operation on the current, voltage, and 2nd-50th harmonic data; S2 specifically involves:

[0010] S2.1: Generate a timestamp for the data collected by each sensor;

[0011] S2.2: The collected current, voltage, and 2nd-50th harmonic data are encapsulated together with a timestamp into a data packet and sent to the data processing unit integrated in the smart circuit breaker;

[0012] S2.3: The data processing unit performs data alignment on the data packet; an improved spline interpolation method is used for the data alignment operation;

[0013] S3: Perform feature extraction on the current, voltage, and 2nd-50th harmonic data after data synchronization to obtain feature extraction vectors;

[0014] S4: Establish a device operating status identification model;

[0015] S5: Input the feature extraction vector into the device operation status recognition model to obtain the device operation status recognition result.

[0016] As a preferred embodiment of the present invention, in step S2.3, the data alignment operation using the improved spline interpolation method specifically involves:

[0017] Analyze the timestamp distribution to determine the time interval Δt between data points;

[0018] The interpolation node is dynamically selected based on the time interval Δt.

[0019] In each interpolation interval, an improved spline interpolation method is used to generate time-aligned data points;

[0020] The method of generating time-aligned data points using the improved spline interpolation method is specifically as follows:

[0021] Establish a cubic spline interpolation polynomial;

[0022] Determine the constraints of the physical model;

[0023] Construct a system of linear equations based on the constraints of the physical model, and solve for the interpolation coefficients. a k , b k， c k , d k Thus, the cubic polynomial function is determined;

[0024] Time-aligned data points are generated based on the cubic polynomial function.

[0025] As a preferred embodiment of the present invention, the cubic polynomial S k (t The mathematical expression of ) is:

[0026] ;

[0027] In the formula, a k b k c k d k The coefficients to be solved are t; t is time. k This is the starting point of the interpolation interval;

[0028] The physical model constraints are as follows:

[0029] ;

[0030] Among them, y k Let y be the sampled value at time k. k+1 Let y be the sampled value at time k+1. k+2 The sampled value at time k+2;

[0031] The formula for the system of linear equations is:

[0032] .

[0033] As a preferred embodiment of the present invention, in step S2.2, each data packet includes a sensor type, a data value, and a timestamp.

[0034] As a preferred embodiment of the present invention, S3 specifically involves: performing feature extraction operations on the current, voltage, and 2nd-50th harmonic data that have been synchronized with the data, extracting current harmonic variation features, voltage waveform variation features, and 2nd-50th harmonic variation trends, and forming a feature extraction vector.

[0035] As a preferred embodiment of the present invention, the current harmonic variation characteristics include harmonic ratio characteristics and harmonic energy characteristics; the voltage waveform variation characteristics include peak value, RMS value, and waveform factor; the 2nd-50th harmonic variation trend includes the 2nd-50th current variation trend and the 2nd-50th voltage harmonic variation trend.

[0036] In a preferred embodiment of the present invention, in step S4, the device operating status recognition model is an artificial neural network model; the artificial neural network model comprises an input layer, a hidden layer, an output layer, and an activation function.

[0037] As a preferred embodiment of the present invention, the activation function is a dynamic gating activation function; its function expression is:

[0038] ;

[0039] In the formula,σ It is the sigmoid function, and tanh is the hyperbolic tangent function. α and β These are learnable parameters, and ⊙ represents element-wise product.

[0040] As a preferred embodiment of the present invention, in step S5, the device operating status identification result includes normal operation, overload, short circuit, and fault.

[0041] In a preferred embodiment of the present invention, in step S1, the current data of the device is collected using a current transformer integrated in the intelligent circuit breaker; and the voltage data of the device is collected using a voltage sensor integrated in the intelligent circuit breaker.

[0042] The present invention also includes a smart circuit breaker, using the above-described method for identifying the operating status of a device based on a smart circuit breaker, comprising the following modules:

[0043] The acquisition module is used to acquire the current, voltage, and 2nd-50th harmonic data of the device;

[0044] The synchronization module is used to perform data synchronization operations on the current, voltage, and 2nd-50th harmonic data;

[0045] The feature extraction module is used to perform feature extraction operations on the current, voltage, and 2nd-50th harmonic data after data synchronization to obtain feature extraction vectors;

[0046] The processor is used to establish a device operating status recognition model and input the feature extraction vector into the device operating status recognition model to obtain the device operating status recognition result.

[0047] Compared with the prior art, the present invention has the following advantages:

[0048] In synchronizing data collected by a smart circuit breaker, this invention first generates a timestamp for the data collected by each sensor. The collected current, voltage, and 2nd-50th harmonic data are then encapsulated together with the timestamps into a data packet and sent to the integrated data processing unit of the smart circuit breaker. The data processing unit then performs data alignment on the data packet, employing an improved spline interpolation method. Furthermore, physical model constraints are established, and a system of linear equations is constructed based on these constraints to solve for the interpolation coefficients, thereby determining the cubic polynomial function. Time-aligned data points are then generated based on the cubic polynomial function. By introducing physical model constraints, the interpolation results are ensured to conform to the actual physical process, improving the reliability of subsequent analysis and state identification.

[0049] Meanwhile, this invention improves the existing ReLU activation function when using an artificial neural network model as the equipment operation status recognition model. By combining a gating mechanism and a nonlinear activation function, and learning the α and β parameters, the shape and nonlinearity of the activation function can be adaptively adjusted. Furthermore, by incorporating the sigmoid and hyperbolic tangent functions, it exhibits better smoothness, contributing to the stability of the optimization process. Sufficient nonlinearity is also introduced, enabling the network to learn and represent complex functions. By designing an improved artificial neural network model, including creating new activation functions, the accuracy of equipment operation status recognition can be improved. The new Dynamically Gated Activation Function (DGA), by introducing adaptability and smoothness, provides the neural network with stronger expressive power and optimization stability. These improvements contribute to building a more powerful and flexible equipment operation status recognition system, thereby improving the accuracy of recognition. Attached Figure Description

[0050] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0051] Figure 1 A flowchart of a device operation status identification method based on a smart circuit breaker provided in an embodiment of the present invention.

[0052] Figure 2 This is a flowchart of a data synchronization operation for the current, voltage, and 2nd-50th harmonic data provided in an embodiment of the present invention. Detailed Implementation

[0053] 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.

[0054] The concepts involved in this application will first be described with reference to the accompanying drawings. It should be noted that the following descriptions of various concepts are only for the purpose of making the content of this application easier to understand and do not constitute a limitation on the scope of protection of this application; furthermore, the embodiments and features in the embodiments of this application can be combined with each other unless otherwise specified. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0055] Example 1

[0056] As attached Figure 1 As shown, the present invention provides a method for identifying the operating status of equipment based on a smart circuit breaker, comprising the following steps:

[0057] S1: The current, voltage, and 2nd-50th harmonic data of the device are collected using the sensors integrated in the intelligent circuit breaker;

[0058] The device employs a high-precision current transformer (CT) integrated into the intelligent circuit breaker to collect current data. The current transformer can convert current data into voltage data suitable for processing, and features high precision and high dynamic range. Generally, based on the operating characteristics of the device, the current transformer's acquisition frequency is set to 51200Hz to capture rapidly changing current data.

[0059] The voltage data of the device is collected by a high-precision voltage sensor integrated into the intelligent circuit breaker. The voltage sensor can convert the voltage data into voltage data suitable for processing. The voltage sensor should have high precision, high linearity and good insulation performance. The voltage sensor is used in conjunction with the current transformer to ensure that voltage changes during device operation can be monitored in real time. The acquisition frequency of the voltage sensor is consistent with that of the current data, ranging from 100Hz to 1000Hz. This ensures time synchronization of current and voltage data, which is convenient for subsequent power calculation and analysis.

[0060] In addition, the data that the intelligent circuit breaker can collect includes: current, voltage, residual current (leakage current), frequency, and temperature.

[0061] Sampling technology: High-frequency sampling technology, 1024 samples per cycle, with each cycle lasting 20ms.

[0062] The calculated data includes: active power, reactive power, power factor, 2nd-50th order voltage harmonics, 2nd-50th order current harmonics, total harmonic distortion, three-phase imbalance, capacitive leakage current, resistive leakage current, cumulative active energy, and cumulative reactive energy.

[0063] S2: Perform data synchronization operation on the current, voltage and 2nd-50th harmonic data;

[0064] In existing technologies, data synchronization typically relies on a central clock signal from smart sensors to synchronize the acquisition actions of all sensors. While this method ensures the consistency of data acquisition time, it also has some limitations: the clocks of different sensors may drift due to factors such as temperature and power fluctuations, leading to inconsistent data acquisition times; in distributed systems, communication delays between sensors and the central processing unit may affect the accuracy of data synchronization; and if the central clock signal malfunctions, the synchronization mechanism of the entire system will fail. Therefore, based on the above-mentioned technological status, this embodiment proposes a distributed data synchronization scheme based on timestamps and data interpolation. This scheme achieves high-precision data synchronization by attaching a high-precision timestamp to each sensor and performing time calibration.

[0065] Specifically, as shown in the attached document Figure 2 As shown, S2 specifically refers to:

[0066] S2.1: Generate a timestamp for the data collected by each sensor;

[0067] S2.2: The collected current, voltage, and 2nd-50th harmonic data are encapsulated together with a timestamp into a data packet and sent to the data processing unit integrated in the smart circuit breaker;

[0068] Each data packet includes the sensor type, data value, and timestamp.

[0069] The data packet is transmitted to the data processing unit using a high-speed communication protocol (such as high-speed serial communication), wherein the high-speed communication protocol supports data packet integrity and verification functions to prevent data transmission errors.

[0070] S2.3: The data processing unit performs data alignment operation on the data packet;

[0071] This invention employs an improved spline interpolation method for data alignment. While traditional spline interpolation methods can achieve smooth data interpolation, they may face problems such as insufficient interpolation accuracy, high computational complexity, and unclear physical meaning of the interpolation results in high-precision data synchronization scenarios. Therefore, this invention proposes an improved spline interpolation method aimed at improving interpolation accuracy, reducing computational complexity, and ensuring that the physical meaning of the interpolation results is consistent with the actual application scenario.

[0072] Specifically, the data alignment operation using the improved spline interpolation method is as follows:

[0073] Analyze the timestamp distribution to determine the time interval Δt between data points;

[0074] The interpolation node is dynamically selected based on the time interval Δt.

[0075] Specifically, for regions where the time interval is greater than the preset time interval, the interpolation node density is increased; for regions where the time interval is less than or equal to the preset time interval, the interpolation node density is decreased.

[0076] In each interpolation interval, an improved spline interpolation method is used to generate time-aligned data points;

[0077] Specifically:

[0078] Establish a cubic spline interpolation polynomial;

[0079] Wherein, the cubic polynomial S k ( t The mathematical expression of ) is:

[0080] ;

[0081] In the formula, a k b k c k d k The coefficients to be solved are t; t is time. k This is the starting point of the interpolation interval;

[0082] Determine the constraints of the physical model;

[0083] The physical model constraints are used to ensure that the interpolation results are not only mathematically smooth and continuous, but also physically consistent with the actual physical process. By introducing physical laws related to the measured physical quantity, the physical model constraints limit the shape of the interpolation curve, thereby improving the accuracy and reliability of the interpolation results.

[0084] The physical model constraints are as follows:

[0085] ;

[0086] Among them, y k Let y be the sampled value at time k. k+1 Let y be the sampled value at time k+1. k+2 The sampled value at time k+2.

[0087] The first and second equations in the above system of equations are used to ensure that the spline interpolation function is at node t. k and t k+1 The function value at point y and the given data point y k and y k+1 Matching; the third and fourth equations in the above system of equations ensure that the spline function matches at the nodes. t k and t k+1The first derivative (i.e., the slope) at these points is continuous. This means that the slope of the curve does not change abruptly at these points, thus ensuring the smoothness of the curve. These slopes are estimated based on the differences between adjacent data points, reflecting the trend of data variation around these points.

[0088] Construct a system of linear equations based on the constraints of the physical model, and solve for the interpolation coefficients. a k , b k， c k , d k Thus, the cubic polynomial function is determined;

[0089] The formula for the system of linear equations is:

[0090] ;

[0091] Based on the above system of linear equations, the coefficient 'a' is obtained by solving for each interpolation interval. k b k c k， d k Thus, the cubic polynomial function is determined;

[0092] Time-aligned data points are generated based on the cubic polynomial function.

[0093] The interpolation result S k (t) serves as the time-aligned data point for subsequent data processing and state identification. By introducing physical model constraints, the interpolation results are ensured to conform to the actual physical process, thus improving the reliability of subsequent analysis and state identification.

[0094] S3: Perform feature extraction on the current, voltage, and 2nd-50th harmonic data after data synchronization to obtain feature extraction vectors;

[0095] Feature extraction is performed on the current, voltage, and 2nd-50th harmonic data after data synchronization to extract current harmonic variation features, voltage waveform variation features, and 2nd-50th harmonic variation trends.

[0096] Current harmonics are a phenomenon in power systems where the current waveform deviates from an ideal sinusoidal waveform. They are typically caused by nonlinear loads, such as rectifiers and frequency converters. The characteristics of current harmonic variations can reflect the operating status of equipment and the health of the power system. These characteristics include harmonic ratio characteristics and harmonic energy characteristics. The process of extracting current harmonic variation characteristics is as follows:

[0097] The current data is subjected to a fast Fourier transform to obtain a frequency domain representation of the current data, thereby identifying and quantifying the content of each harmonic.

[0098] The ratio of each harmonic to the fundamental frequency is calculated to obtain the harmonic ratio characteristics, which are used to assess the severity of harmonics.

[0099] The energy of each harmonic is calculated to obtain the harmonic energy characteristics, which are used to assess the impact of harmonics on equipment.

[0100] Voltage waveform variation characteristics can reflect the stability and power supply quality of a power system. Changes in voltage waveform may be caused by load variations, system faults, or interference; the process of voltage waveform variation characteristics is as follows:

[0101] Time-domain analysis was performed on the voltage data to extract voltage waveform variation characteristics;

[0102] The voltage waveform variation characteristics include peak value, RMS value, waveform factor, etc.

[0103] The variation trend of the 2nd to 50th harmonics can reflect the operating status of key components inside the equipment; the variation trend of the 2nd to 50th harmonics includes the variation trend of the 2nd to 50th current harmonics and the variation trend of the 2nd to 50th voltage harmonics.

[0104] The above features are combined into a feature extraction vector. Through feature extraction, the feature extraction vector can extract key information reflecting the operating status of the equipment from the original data, providing strong support for status identification.

[0105] S4: Establish a device operating status identification model;

[0106] The device operation status recognition model is an artificial neural network model; the artificial neural network model consists of an input layer, a hidden layer, an output layer, and an activation function.

[0107] The inputs of the input layer include current harmonic variation characteristics, voltage waveform variation characteristics, and the variation trend of the 2nd to 50th harmonics.

[0108] The number of hidden layers is typically set to 2 to 4, and in this embodiment, it is 4. The number of neurons in each hidden layer can vary, typically ranging from several hundred to several thousand. In this embodiment, a grid search method is used to determine the optimal number of neurons in each hidden layer.

[0109] The output of the output layer represents the device's operating status, and the output dimension of the output layer is determined based on the number of categories of the device's operating status. For example, if there are... m If there are several states, then the output layer has... m One neuron.

[0110] Furthermore, activation functions are key to nonlinear mapping in neural networks, introducing nonlinear characteristics that enable networks to learn and represent complex functions. In existing technologies, ReLU activation function is a commonly used activation function; however, ReLU activation function may lead to the "dead ReLU" problem. This embodiment proposes an improved ReLU activation function to solve the "dead ReLU" problem.

[0111] The improved activation function is a Dynamic Gated Activation Function (DGA); the DGA is a novel activation function that combines gating mechanisms and nonlinear activation; its function expression is:

[0112] ;

[0113] In the formula, σ It is the sigmoid function, and tanh is the hyperbolic tangent function. α and β These are learnable parameters, and ⊙ represents element-wise product.

[0114] The proposed dynamic gated activation function (DGA) adaptively adjusts its shape and nonlinearity by learning the α and β parameters. Furthermore, by combining the sigmoid and hyperbolic tangent functions, it exhibits good smoothness, contributing to the stability of the optimization process. The introduction of sufficient nonlinearity allows the network to learn and represent complex functions. By designing improved artificial neural network models, including creating new activation functions, the accuracy of equipment operating status recognition can be improved. The new DGA, through its adaptiveness and smoothness, provides the neural network with stronger expressive power and optimization stability. These improvements contribute to building a more robust and flexible equipment operating status recognition system.

[0115] The training process of the artificial neural network model is an existing technology. This embodiment focuses on the model architecture of the artificial neural network model, and its training process will not be discussed in detail in this embodiment.

[0116] S5: Input the feature extraction vector into the device operation status recognition model to obtain the device operation status recognition result.

[0117] The equipment operating status identification results include normal operation, overload, short circuit, and fault.

[0118] Example 2

[0119] like Figure 2As shown, an intelligent circuit breaker, using the above-described method for identifying the operating status of an intelligent circuit breaker, includes the following modules:

[0120] The acquisition module is used to acquire the current, voltage, and 2nd-50th harmonic data of the device;

[0121] The synchronization module is used to perform data synchronization operations on the current, voltage, and 2nd-50th harmonic data;

[0122] The feature extraction module is used to perform feature extraction operations on the current, voltage, and 2nd-50th harmonic data after data synchronization to obtain feature extraction vectors;

[0123] The processor is used to establish a device operating status recognition model and input the feature extraction vector into the device operating status recognition model to obtain the device operating status recognition result.

[0124] Example 3

[0125] This embodiment includes a computer-readable storage medium storing a data processing program, which is executed by a processor as a device operation status identification method based on a smart circuit breaker according to Embodiment 1.

[0126] The embodiments and / or implementation methods described above are merely preferred embodiments and / or implementation methods for implementing the technology of the present invention, and are not intended to limit the implementation methods of the technology of the present invention in any way. Any person skilled in the art can make some modifications or alterations to other equivalent embodiments without departing from the scope of the technical means disclosed in the content of the present invention, but they should still be regarded as the technology or embodiments that are substantially the same as the present invention.

[0127] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. The above descriptions are only preferred embodiments of this application. It should be noted that due to the limitations of written expression, while there are objectively infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of this application, and can also combine the above technical features in an appropriate manner. These improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of this application.

Claims

1. A method for identifying the operating status of equipment based on a smart circuit breaker, characterized in that, Includes the following steps: S1: The current, voltage, and 2nd-50th harmonic data of the device are collected using the sensors integrated in the intelligent circuit breaker; S2: Perform data synchronization operation on the current, voltage, and 2nd-50th harmonic data; S2 specifically involves: S2.1: Generate a timestamp for the data collected by each sensor; S2.2: The collected current, voltage, and 2nd-50th harmonic data are encapsulated together with a timestamp into a data packet and sent to the data processing unit integrated in the smart circuit breaker; S2.3: The data processing unit performs data alignment on the data packet; an improved spline interpolation method is used for the data alignment operation; In step S2.3, the data alignment operation using the improved spline interpolation method specifically involves: Analyze the timestamp distribution to determine the time interval Δt between data points; The interpolation node is dynamically selected based on the time interval Δt. In each interpolation interval, an improved spline interpolation method is used to generate time-aligned data points; The method of generating time-aligned data points using the improved spline interpolation method is specifically as follows: Establish a cubic spline interpolation polynomial; Determine the constraints of the physical model; Construct a system of linear equations based on the constraints of the physical model, and solve for the interpolation coefficients. a k , b k， c k , d k， Thus, the cubic polynomial function is determined; Time-aligned data points are generated based on the cubic polynomial function; S3: Perform feature extraction on the current, voltage, and 2nd-50th harmonic data after data synchronization to obtain feature extraction vectors; S4: Establish a device operating status identification model; S5: Input the extracted feature vector into the device operating status recognition model to obtain the device operating status recognition result; The cubic polynomial function S k ( t The mathematical expression of ) is: ； In the formula, a k b k c k d k The coefficients to be solved are t; t is time. k This is the starting point of the interpolation interval; The physical model constraints are as follows: ； Among them, y k Let y be the sampled value at time k. k+1 Let y be the sampled value at time k+1. k+2 The sampled value at time k+2; The formula for the system of linear equations is: 。 2. The equipment operation status identification method based on intelligent circuit breakers according to claim 1, characterized in that, Specifically, S3 involves performing feature extraction on the synchronized current, voltage, and 2nd-50th harmonic data to extract current harmonic variation features, voltage waveform variation features, and the variation trend of the 2nd-50th harmonics. The feature extraction vector is composed of these vectors.

3. The equipment operation status identification method based on intelligent circuit breakers according to claim 2, characterized in that, The current harmonic variation characteristics include harmonic ratio characteristics and harmonic energy characteristics; The voltage waveform variation characteristics include peak value, RMS value, and waveform factor; the 2nd-50th harmonic variation trend includes the 2nd-50th current harmonic variation trend and the 2nd-50th voltage harmonic variation trend.

4. The equipment operation status identification method based on intelligent circuit breakers according to claim 1, characterized in that, In step S4, the device operating status recognition model is an artificial neural network model; the artificial neural network model includes an input layer, a hidden layer, an output layer, and an activation function.

5. The equipment operation status identification method based on intelligent circuit breakers according to claim 4, characterized in that, The activation function is a dynamic gated activation function; its function expression is: ； In the formula, σ It is the sigmoid function, and tanh is the hyperbolic tangent function. α and β These are learnable parameters, and ⊙ represents element-wise product.

6. The equipment operation status identification method based on intelligent circuit breakers according to claim 1, characterized in that, In S5, the equipment operating status identification result includes normal operation, overload, short circuit, and fault.

7. The equipment operation status identification method based on intelligent circuit breakers according to claim 1, characterized in that, In step S1, the current data of the device is collected using the current transformer integrated in the smart circuit breaker; the voltage data of the device is collected using the voltage sensor integrated in the smart circuit breaker.

8. An intelligent circuit breaker, characterized in that, The intelligent circuit breaker uses the equipment operating status identification method as described in any one of claims 1-7, including the following modules: The acquisition module is used to acquire the current, voltage, and 2nd-50th harmonic data of the device; The synchronization module is used to perform data synchronization operations on the current, voltage, and 2nd-50th harmonic data; The feature extraction module is used to perform feature extraction operations on the current, voltage, and 2nd-50th harmonic data after data synchronization to obtain feature extraction vectors; The processor is used to establish a device operating status recognition model and input the feature extraction vector into the device operating status recognition model to obtain the device operating status recognition result.

Images