Switch safety identification method of circuit breaker equipment

Abstract

The invention provides a switch safety identification method of circuit breaker equipment, which belongs to the technical field of switch safety identification of the circuit breaker equipment, and comprises the following steps: acquiring electrical quantity data (voltage and current) and mechanical quantity data (displacement, speed, acceleration, vibration and the like) in a switch opening and closing process; establishing a mathematical equation set of electrical quantity and mechanical quantity, and solving a value range of each parameter through a numerical solution and an analytical solution; carrying out normalization processing on the collected data, and extracting a feature matrix; inputting the characteristic matrix into a pre-trained fault mode identification model, and identifying a fault mode of opening and closing of the circuit breaker switch; and a diagnosis result is output to operation and maintenance personnel, so that a basis is provided for equipment state monitoring and fault early warning. The method provided by the invention solves the technical problems that the operating environment of the circuit breaker is complex and changeable, the external interference factors are numerous, and the real state of the circuit breaker is difficult to accurately reflect by relying on single-type data in the prior art.

Description

Technical Field

[0001] The present invention belongs to the technical field of switch safety identification of circuit breaker equipment, and in particular, relates to a switch safety identification method of circuit breaker equipment. Background Art

[0002] With the rapid development of power systems and the widespread application of smart grids, circuit breakers, as key equipment in medium-voltage distribution networks, have a direct impact on the safe and stable operation of power systems due to their operational reliability. In practical applications, circuit breakers frequently perform opening and closing operations, which can easily lead to various failure modes, such as mechanical jamming, contact erosion, spring failure, etc. These failures not only cause the circuit breaker to fail to work properly, but may also cause more serious accidents, resulting in large-scale power outages and equipment damage. Therefore, it is of great significance to identify the failure mode of circuit breakers in a timely and accurate manner.

[0003] At present, circuit breaker fault diagnosis mainly relies on regular maintenance and on-site manual inspections. This method is not only time-consuming and labor-intensive, but also difficult to detect potential fault hazards in a timely manner. In recent years, with the advancement of sensing technology and signal processing technology, fault diagnosis methods based on vibration signals, acoustic emission signals, and current signals have gradually emerged. However, these methods often only focus on a single type of signal and are difficult to fully reflect the operating status of the circuit breaker. In addition, most of the existing fault diagnosis algorithms are based on traditional signal processing technology and simple classification models. When faced with complex fault modes, the diagnostic accuracy and reliability are often unsatisfactory.

[0004] In actual applications, the operating environment of circuit breakers is complex and changeable, with many external interference factors. It is difficult for a single type of data to accurately reflect the true state of the circuit breaker. At the same time, the failure mode of the circuit breaker is diverse and complex, often manifested as an abnormal combination of multiple parameters. Existing technologies are difficult to effectively handle such complex fault characteristics, resulting in high misdiagnosis and missed diagnosis rates. Summary of the invention

[0005] In view of this, the present invention provides a switch safety identification method for a circuit breaker device, which can solve the technical problem that the operating environment of the circuit breaker is complex and changeable, there are many external interference factors, and the prior art relies on a single type of data to accurately reflect the true state of the circuit breaker.

[0006] The present invention is achieved in that:

[0007] A first aspect of the present invention provides a switch safety identification method for a circuit breaker device, comprising the following steps:

[0008] S10, collecting electrical quantity data and mechanical quantity data during the opening and closing process of the circuit breaker switch, wherein the electrical quantity data includes the three-phase voltage and current waveforms of the circuit breaker, the operating current and voltage waveforms of the switch mechanism, and the current and voltage parameters of the contact opening and closing process; the mechanical quantity data includes the displacement, speed, acceleration curve of the switch mechanism, and the vibration signals of the spring and damper of the circuit breaker;

[0009] S20, establishing a group of electrical quantity equations, including a three-phase voltage equation, a three-phase current equation, a switch mechanism action current equation, a switch mechanism action voltage equation, and a contact opening and closing process current and voltage equation;

[0010] S30, establishing a mechanical quantity equation group, including a switch mechanism displacement equation, a switch mechanism velocity equation, a switch mechanism acceleration equation, a circuit breaker spring vibration equation, and a circuit breaker damper vibration equation;

[0011] S40, solving the electrical quantity equation group to obtain multiple electrical quantity numerical solutions, which are three-phase voltage peak value, three-phase current peak value, maximum value of switch mechanism action current, maximum value of switch mechanism action voltage, maximum value of contact opening and closing process current, and maximum value of contact opening and closing process voltage;

[0012] S50, solving the mechanical quantity equation group to obtain multiple mechanical quantity numerical solutions, which are respectively the maximum displacement of the switch mechanism, the maximum speed of the switch mechanism, the maximum acceleration of the switch mechanism, the vibration frequency of the circuit breaker spring, and the vibration amplitude of the circuit breaker damper;

[0013] S60, analytically solving the electrical quantity equation group to obtain multiple electrical quantity analytical solutions, substituting into preset electrical quantity initial conditions to obtain multiple electrical quantity ranges;

[0014] S70, analytically solving the mechanical quantity equation group to obtain multiple mechanical quantity analytical solutions, substituting into preset mechanical quantity initial conditions to obtain multiple mechanical quantity ranges;

[0015] S80, normalizing the multiple numerical solutions of electrical quantities, the multiple numerical solutions of mechanical quantities, the maximum values ​​and minimum values ​​of the multiple ranges of electrical quantities, and the maximum values ​​and minimum values ​​of the multiple ranges of mechanical quantities so that each parameter value falls within the interval [0,1], concatenating the normalized data and extracting features through principal component analysis or other dimensionality reduction methods to obtain a feature matrix;

[0016] S90, input the feature matrix into a pre-trained circuit breaker switch opening and closing fault mode recognition model to obtain a circuit breaker switch opening and closing fault mode vector, compare the highest fault probability in the fault mode vector with a preset safety threshold, if the highest fault probability is greater than the safety threshold, determine that the circuit breaker switch safety is low, otherwise, determine that the circuit breaker switch safety is high. The safety threshold is preset to 0.8.

[0017] Among them, the multiple analytical solutions of electrical quantities are: three-phase voltage analytical expression, three-phase current analytical expression, switch mechanism action current analytical expression, switch mechanism action voltage analytical expression, contact opening and closing process current analytical expression, contact opening and closing process voltage analytical expression.

[0018] Furthermore, the multiple electrical quantity ranges are: three-phase voltage peak range; three-phase current peak range; switch mechanism action current maximum range; switch mechanism action voltage maximum range; contact opening and closing process current maximum range; contact opening and closing process voltage maximum range.

[0019] Furthermore, the analytical solutions of the multiple electrical quantities are: an analytical expression for three-phase voltage; an analytical expression for three-phase current; an analytical expression for the switching mechanism operating current; an analytical expression for the switching mechanism operating voltage; an analytical expression for the current in the contact opening and closing process; and an analytical expression for the voltage in the contact opening and closing process.

[0020] Furthermore, the multiple mechanical quantity ranges are: a maximum displacement range of the switch mechanism; a maximum speed range of the switch mechanism; a maximum acceleration range of the switch mechanism; a vibration frequency range of the circuit breaker spring; and a vibration amplitude range of the circuit breaker damper.

[0021] The three-phase voltage equation is specifically expressed as:

[0022]

[0023] Where U φ (t) is the voltage of a certain phase changing with time; U m is the fundamental amplitude, which is measured by voltage transformer; ω is the angular frequency, which is equal to 2πf, where f is the system frequency; t is the time variable; θ φ is the initial phase angle, determined by the synchronous measurement device; U n is the amplitude of the nth harmonic, obtained by Fourier analysis; φ n is the initial phase angle of the nth harmonic; N is the highest harmonic number considered; U dc is the DC bias component, detected by the voltage measurement device; τ is the time constant, which is related to the system impedance; ε(t) is the random noise term;

[0024] The three-phase current equation is specifically expressed as:

[0025]

[0026] In the formula, I φ (t) is the time-varying function of the current of a certain phase; I m is the fundamental amplitude, measured by current transformer; φ is the power factor angle, measured by power analyzer; I n is the amplitude of the nth harmonic, obtained by Fourier analysis; φ n is the phase difference of the nth harmonic; I dc is the DC bias component, detected by the current measuring device; τ I is the current time constant, which is related to the system impedance and inductance; K t is the eddy current coefficient, which is related to the circuit breaker core material; η(t) is the random noise term.

[0027] The switch mechanism action current equation is specifically expressed as:

[0028]

[0029] In the formula, I op (t) is the operating current of the switch mechanism; I0 is the initial current, which is determined by the initial state of the operating mechanism; K m is the mass coefficient, which is related to the mass of the moving parts of the operating mechanism; x(t) is the displacement function of the operating mechanism; K v is the velocity damping coefficient, which is related to the friction of the operating mechanism; K s I is the spring stiffness coefficient, which is determined by the spring characteristics of the operating mechanism; f is the Coulomb friction current, which is related to the contact surface material; sign() is the sign function; I e is the eddy current amplitude; τ e is the eddy current time constant; ξ(t) is the random noise term;

[0030] The switching mechanism action voltage equation is specifically expressed as:

[0031]

[0032] Where U op (t) is the operating voltage of the switch mechanism; U0 is the rated operating voltage; R c is the equivalent resistance of the operating circuit; L c is the equivalent inductance of the operating circuit; K b is the back electromotive force coefficient; μ(t) is the voltage fluctuation function, simulating the grid fluctuation; ω v is the voltage fluctuation angular frequency; ζ(t) is the random noise term.

[0033] The current and voltage equation of the contact opening and closing process is specifically expressed as:

[0034]

[0035] In the formula, I c (t), U c (t) is the contact current and voltage; I a , U a are the initial current and voltage amplitudes; τ a , τ c is the current and voltage decay time constant; ω a is the current oscillation angular frequency; I b , U b are the steady-state current and voltage components; τ b , τ d is the current and voltage rise time constant; I arc , U arc are the arc current and voltage characteristic values; K arc is the arc formation rate coefficient; l arc (t) is the arc length function; I0 is a small current constant to prevent division by zero error; γ(t) and δ(t) are random noise terms.

[0036] The displacement equation of the switch mechanism is specifically expressed as:

[0037]

[0038] Where x(t) is the displacement of the switch mechanism; ζ is the damping ratio; ω n is the natural angular frequency; F m (t) is the driving force function; m is the mass of the switch mechanism; F f is the friction force; k is the spring stiffness; A e is the external vibration amplitude; ω e is the external vibration angular frequency;∈(t) is the random disturbance;

[0039] The speed equation of the switch mechanism is specifically expressed as:

[0040]

[0041] Where v(t) is the speed of the switch mechanism; v0 is the initial speed; β is the damping coefficient, which is equal to ζω n ; φ is the phase angle, equal to arctan(ω e / β); ν(t) is the random disturbance of velocity;

[0042] The acceleration equation of the switch mechanism is specifically expressed as:

[0043]

[0044] Where a(t) is the acceleration of the switch mechanism; κ(t) is the random disturbance of acceleration;

[0045] The circuit breaker spring vibration equation is specifically expressed as:

[0046]

[0047] Where y(t) is the spring displacement; ζ s is the spring-damping ratio; ω s is the natural angular frequency of the spring; F s (t) is the external force acting on the spring; m s is the equivalent mass of the spring; A s is the initial vibration amplitude of the spring; α s is the spring vibration attenuation coefficient; λ(t) is the random disturbance of spring vibration;

[0048] The circuit breaker damper vibration equation is specifically expressed as:

[0049]

[0050] Where z(t) is the damper displacement; ζ d is the damping ratio of the damper; ω d is the natural angular frequency of the damper; F d (t) is the external force acting on the damper; m d is the mass of the damper; c d is the nonlinear damping coefficient; A d is the initial vibration amplitude of the damper; α d is the vibration attenuation coefficient of the damper; ρ(t) is the random disturbance of the damper vibration.

[0051] Furthermore, the training step of the circuit breaker switch opening and closing fault mode recognition model specifically includes:

[0052] Establishing a training data set, specifically obtaining multiple sets of electrical quantity data and mechanical quantity data during the opening and closing process of circuit breaker switches with known faults, and generating a feature matrix in accordance with steps S10 to S80, wherein the input of the training is the generated feature matrix, and the output of the training is the known fault;

[0053] Network training, using the training data set to train a multi-layer perceptron network to obtain the circuit breaker switch opening and closing fault mode recognition model.

[0054] Compared with the prior art, the switch safety identification method of the circuit breaker equipment provided by the present invention establishes a set of mathematical models that comprehensively reflect the operating status of the circuit breaker by comprehensively collecting and analyzing electrical quantity data and mechanical quantity data. The method not only considers traditional electrical parameters such as voltage and current, but also introduces mechanical parameters such as the displacement, speed, acceleration of the switch mechanism, and vibration signals of the spring and damper, thereby realizing all-round monitoring of the circuit breaker status.

[0055] By establishing a set of electrical quantity equations and a set of mechanical quantity equations, this method can accurately describe the dynamic characteristics of the circuit breaker during the opening and closing process. By numerically and analytically solving these equations, not only the specific values ​​of the key parameters are obtained, but also the theoretical range of variation of these parameters. This dual solution method greatly improves the accuracy and reliability of fault diagnosis.

[0056] The present invention adopts advanced data processing and feature extraction technology. Through normalization processing and principal component analysis, complex multi-dimensional data is converted into a low-dimensional feature matrix, which effectively extracts the key features of the fault, reduces data redundancy, and improves the efficiency of subsequent fault identification.

[0057] In the fault identification stage, this method uses a multi-layer perceptron neural network model, which has strong nonlinear mapping and self-learning capabilities and can effectively handle complex fault modes. By outputting fault mode vectors and their probabilities, it provides an intuitive and reliable decision-making basis for operation and maintenance personnel.

[0058] Compared with the prior art, the method of the present invention has the following significant improvements:

[0059] 1. Comprehensive monitoring of circuit breaker status is realized, which improves the comprehensiveness and accuracy of fault diagnosis.

[0060] 2. By establishing an accurate mathematical model, the ability to extract fault features and the accuracy of fault mode identification are improved.

[0061] 3. The use of advanced data processing and machine learning technologies has improved the intelligence level and adaptability of fault diagnosis.

[0062] 4. Ability to identify complex failure modes and provide probability assessment, providing strong support for predictive maintenance.

[0063] In summary, the present invention solves the technical problem that the operating environment of the circuit breaker is complex and changeable, there are many external interference factors, and the prior art is difficult to accurately reflect the real state of the circuit breaker by relying on a single type of data. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A flow chart of the method provided by the present invention. DETAILED DESCRIPTION

[0065] In order to make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0066] like Figure 1 As shown, it is a flow chart of a switch safety identification method of a circuit breaker device provided by the present invention. The method comprises the following steps: wherein, the following steps are included:

[0067] S10, collecting electrical quantity data and mechanical quantity data during the opening and closing process of the circuit breaker switch, the electrical quantity data including the three-phase voltage and current waveforms of the circuit breaker, the operating current and voltage waveforms of the switch mechanism, and the current and voltage parameters of the contact opening and closing process; the mechanical quantity data including the displacement, speed, acceleration curve of the switch mechanism, and the vibration signals of the spring and damper of the circuit breaker;

[0068] S20, establishing a group of electrical quantity equations, including a three-phase voltage equation, a three-phase current equation, a switch mechanism action current equation, a switch mechanism action voltage equation, and a contact opening and closing process current and voltage equation;

[0069] S30, establishing a mechanical quantity equation group, including a switch mechanism displacement equation, a switch mechanism velocity equation, a switch mechanism acceleration equation, a circuit breaker spring vibration equation, and a circuit breaker damper vibration equation;

[0070] S40, solving the electrical quantity equation group to obtain multiple electrical quantity numerical solutions, which are three-phase voltage peak value, three-phase current peak value, maximum value of switch mechanism action current, maximum value of switch mechanism action voltage, maximum value of contact opening and closing process current, and maximum value of contact opening and closing process voltage;

[0071] S50, solving the mechanical quantity equation group to obtain multiple mechanical quantity numerical solutions, which are the maximum displacement of the switch mechanism, the maximum speed of the switch mechanism, the maximum acceleration of the switch mechanism, the vibration frequency of the circuit breaker spring, and the vibration amplitude of the circuit breaker damper;

[0072] S60, analytically solving the electrical quantity equation group to obtain multiple electrical quantity analytical solutions, substituting into preset electrical quantity initial conditions to obtain multiple electrical quantity ranges;

[0073] S70, analytically solving the mechanical quantity equation group to obtain multiple mechanical quantity analytical solutions, substituting into preset mechanical quantity initial conditions to obtain multiple mechanical quantity ranges;

[0074] S80, normalizing multiple numerical solutions of electrical quantities, multiple numerical solutions of mechanical quantities, maximum values ​​and minimum values ​​of multiple ranges of electrical quantities, and maximum values ​​and minimum values ​​of multiple ranges of mechanical quantities so that each parameter value falls within the interval [0,1], concatenating the normalized data and extracting features through principal component analysis or other dimensionality reduction methods to obtain a feature matrix;

[0075] S90. Input the feature matrix into a pre-trained circuit breaker switch opening and closing fault mode recognition model to obtain the circuit breaker switch opening and closing fault mode vector, and compare the highest fault probability in the fault mode vector with a preset safety threshold. If the highest fault probability is greater than the safety threshold, it is determined that the circuit breaker switch safety is low; otherwise, it is determined that the circuit breaker switch safety is high. The safety threshold is preset to 0.8.

[0076] The specific implementation methods of the above steps are described in detail below:

[0077] The specific implementation method of step S10 is: first, install high-precision voltage transformers and current transformers to collect three-phase voltage and current waveform data of the circuit breaker. Secondly, install Hall current sensors and voltage sensors on the switch mechanism to collect the action current and voltage waveforms of the switch mechanism. Thirdly, install micro current and voltage sensors at the contacts to collect the current and voltage parameters of the contact opening and closing process. Then, install displacement sensors, speed sensors and acceleration sensors on the switch mechanism to collect the displacement, speed and acceleration curves of the switch mechanism. Finally, install vibration sensors on the springs and dampers of the circuit breaker to collect vibration signals. The sampling frequency is set to 10kHz and the sampling time is 1 second. A low-pass filter is used to eliminate high-frequency noise during data acquisition, and the cut-off frequency is set to 5kHz. After analog-to-digital conversion, the collected data is stored in the cache for subsequent processing. The purpose of this step is to obtain comprehensive data during the operation of the circuit breaker to provide a basis for subsequent analysis.

[0078] The specific implementation method of step S20 is: based on the data collected in step S10, establish a group of electrical quantity equations. First, use Fourier transform to analyze the three-phase voltage and current waveforms, extract the fundamental and harmonic components, and establish three-phase voltage equations and three-phase current equations. Among them, considering the influence of the 5th, 7th, and 11th harmonics, the harmonic number N is 11. Then, according to the action characteristics of the switch mechanism, the action current equation and action voltage equation of the switch mechanism are established, considering the influence of inductance, resistance and back electromotive force. Finally, based on the physical model of the contact opening and closing process, the current and voltage equations of the contact opening and closing process are established, considering the arc characteristics and contact material characteristics. The parameters in the equation group are obtained by fitting the measured data by the least squares method. The purpose of this step is to establish a mathematical model that accurately reflects the electrical characteristics of the circuit breaker, laying the foundation for subsequent analysis.

[0079] The specific implementation method of step S30 is: based on the data collected in step S10, a set of mechanical quantity equations is established. First, the displacement, velocity, and acceleration data of the switch mechanism are fitted using the least squares method to establish the displacement equation, velocity equation, and acceleration equation of the switch mechanism. Among them, the influence of friction, spring force, and driving force are considered. Then, the vibration signal of the circuit breaker spring is subjected to time-frequency analysis, the main vibration modes are extracted, and the vibration equation of the circuit breaker spring is established. Finally, the vibration signal of the circuit breaker damper is subjected to nonlinear system identification, and the vibration equation of the circuit breaker damper is established, considering the nonlinear damping characteristics. The parameters in the set of equations are obtained by genetic algorithm optimization to minimize the error between the model prediction value and the measured value. The purpose of this step is to establish a mathematical model that accurately reflects the mechanical characteristics of the circuit breaker and provide a basis for subsequent analysis.

[0080] The specific implementation method of step S40 is: numerically solve the electrical quantity equations established in step S20. First, for the three-phase voltage equation and the three-phase current equation, directly substitute the time variable to calculate the peak value. Then, for the switch mechanism action current equation and the action voltage equation, the fourth-order Runge-Kutta method is used for numerical integration, and the time step is 0.1ms. Finally, for the contact opening and closing process current and voltage equation, since the equation is rigid, the backward difference method is used to solve it, and the time step is adaptively adjusted in the range of 0.01ms to 0.1ms. During the solution process, the Newton-Raphson method is used to process nonlinear terms. Through numerical solution, the three-phase voltage peak value, the three-phase current peak value, the maximum value of the switch mechanism action current, the maximum value of the switch mechanism action voltage, the maximum value of the contact opening and closing process current, and the maximum value of the contact opening and closing process voltage are obtained. The purpose of this step is to obtain the specific values ​​of key electrical parameters during the operation of the circuit breaker and provide a quantitative basis for fault diagnosis.

[0081] The specific implementation of step S50 is: numerically solve the mechanical quantity equations established in step S30. First, for the displacement equation, velocity equation and acceleration equation of the switch mechanism, the fifth-order Runge-Kutta method with variable step size is used for numerical integration, the initial step size is 0.1ms, and the error tolerance is set to 1e-6. Then, for the vibration equation of the circuit breaker spring, the Newmark-β method is used to solve it, β is 0.25, γ is 0.5, and the time step is 0.1ms. Finally, for the vibration equation of the circuit breaker damper, due to the existence of strong nonlinearity, the prediction correction method is used to solve it, the prediction step uses the fourth-order Adams-Bashforth formula, the correction step uses the fourth-order Adams-Moulton formula, and the time step is 0.1ms. Through numerical solution, the maximum displacement of the switch mechanism, the maximum speed of the switch mechanism, the maximum acceleration of the switch mechanism, the vibration frequency of the circuit breaker spring, and the vibration amplitude of the circuit breaker damper are obtained. The purpose of this step is to obtain the specific values ​​of the key mechanical parameters during the operation of the circuit breaker, and provide a quantitative basis for fault diagnosis.

[0082] The specific implementation method of step S60 is: analytically solve the set of electrical quantity equations established in step S20. First, for the three-phase voltage equation and the three-phase current equation, write the analytical solution directly. Then, for the action current equation and the action voltage equation of the switch mechanism, the Laplace transform method is used to solve them, considering the influence of the initial conditions. Finally, for the current and voltage equations of the contact opening and closing process, the perturbation method is used for approximate analytical solution, and expanded to the second-order terms. The parameters in the analytical solution are obtained by fitting the measured data by the least squares method. Then, the preset initial conditions of the electrical quantities are substituted into the analytical solution to obtain multiple electrical quantity ranges. The setting of the initial conditions is based on the statistical analysis of a large amount of historical data, such as the range of the initial phase angle of the three-phase voltage is [-π / 6,π / 6], and the range of the initial current of the switch mechanism is [0.8In,1.2In], where In is the rated current. The purpose of this step is to obtain the theoretical variation range of the electrical parameters of the circuit breaker and provide a benchmark for the identification of abnormal conditions.

[0083] The specific implementation method of step S70 is: analytically solve the mechanical quantity equations established in step S30. First, for the switch mechanism displacement equation, velocity equation and acceleration equation, the characteristic root method is used to solve the homogeneous equations, and then the constant variation method is used to solve the non-homogeneous equations. Then, for the circuit breaker spring vibration equation, the separation of variables method is used to solve it, considering the influence of boundary conditions. Finally, for the circuit breaker damper vibration equation, the harmonic balance method is used for approximate analytical solution, retaining the third-order harmonic term. The parameters in the analytical solution are obtained by fitting the measured data through the particle swarm optimization algorithm. Then, the preset mechanical quantity initial conditions are substituted into the analytical solution to obtain multiple mechanical quantity ranges. The setting of the initial conditions is based on the equipment specifications and historical operation data, such as the range of the initial displacement of the switch mechanism is [-1mm, 1mm], and the range of the initial compression of the spring is [0.9L0, 1.1L0], where L0 is the free length of the spring. The purpose of this step is to obtain the theoretical variation range of the mechanical parameters of the circuit breaker and provide a benchmark for the identification of abnormal conditions.

[0084] The specific implementation of step S80 is: first, the multiple electrical quantity numerical solutions and the multiple mechanical quantity numerical solutions obtained in steps S40 and S50 are subjected to maximum and minimum normalization processing, and each parameter value is mapped to the interval [0,1]. Then, for the multiple electrical quantity ranges and the multiple mechanical quantity ranges obtained in steps S60 and S70, the maximum and minimum values ​​are taken and respectively subjected to maximum and minimum normalization processing. Next, the normalized data are combined into a feature vector in a predetermined order. In order to reduce the dimension of the feature vector and extract the main features, the principal component analysis (PCA) method is used for feature extraction. In the PCA process, the covariance matrix of the feature vector is first calculated, and then the eigenvalues ​​and eigenvectors of the covariance matrix are solved, and the first k principal components with a cumulative contribution rate of 95% are selected, and the original eigenvectors are projected onto these k principal components to obtain the feature matrix after dimensionality reduction. If the PCA effect is not ideal, the kernel principal component analysis (KPCA) method is used to extract nonlinear features using the Gaussian kernel function. The purpose of this step is to convert complex multi-dimensional data into a low-dimensional feature matrix that is easy to analyze, thereby improving the efficiency and accuracy of subsequent fault identification.

[0085] The specific implementation of step S90 is: first, construct a multi-layer perceptron neural network as a circuit breaker switch opening and closing fault mode recognition model. The neural network includes an input layer, two hidden layers and an output layer. The number of neurons in the input layer is the same as the number of columns of the feature matrix, the number of neurons in the first hidden layer is 1.5 times that of the input layer, the number of neurons in the second hidden layer is 0.75 times that of the input layer, and the number of neurons in the output layer is the same as the number of predefined fault modes. The hidden layer uses the ReLU activation function, and the output layer uses the Softmax activation function. Then, the feature matrix obtained in step S80 is input into the neural network, and after forward propagation calculation, the result of the output layer, i.e., the fault mode vector, is obtained. Each element represents the probability of the corresponding fault mode. Finally, the fault mode vector is output to the operation and maintenance personnel, and the three fault modes with the highest probability and their corresponding probability values ​​are given at the same time. If the highest probability value is lower than the preset threshold (such as 0.7), it is prompted that there may be an unknown fault mode, and it is recommended to conduct further inspection. The purpose of this step is to quickly identify the fault mode of the circuit breaker based on the extracted features, and provide a decision-making basis for the operation and maintenance personnel.

[0086] The following is a detailed description of each equation:

[0087] 1. Three-phase voltage equation:

[0088]

[0089] Where U φ (t) is the voltage of a certain phase changing with time; U m is the fundamental amplitude, which is measured by a voltage transformer; ω is the angular frequency, which is equal to 2πf, where f is the system frequency (usually 50Hz or 60Hz); t is the time variable; θ φ is the initial phase angle, determined by the synchronous measurement device; U n is the amplitude of the nth harmonic, obtained through Fourier analysis; θ n is the initial phase angle of the nth harmonic; N is the highest harmonic number considered; U dc is the DC bias component, which is detected by the voltage measurement device; τ is the time constant, which is related to the system impedance; ε(t) is the random noise term, which obeys the normal distribution N(0,σ 2 ).

[0090] 2. Three-phase current equation:

[0091]

[0092] In the formula, I φ (t) is the time-varying function of the current of a certain phase; I m is the fundamental amplitude, measured by current transformer; φ is the power factor angle, measured by power analyzer; In is the amplitude of the nth harmonic, obtained by Fourier analysis; φ n is the phase difference of the nth harmonic; I dc is the DC bias component, detected by the current measuring device; τ I is the current time constant, which is related to the system impedance and inductance; K t is the eddy current coefficient, which is related to the circuit breaker core material; η(t) is the random noise term, which obeys the normal distribution

[0093] 3. Switch mechanism action current equation:

[0094]

[0095] In the formula, I op (t) is the operating current of the switch mechanism; I0 is the initial current, which is determined by the initial state of the operating mechanism; K m is the mass coefficient, which is related to the mass of the moving parts of the operating mechanism; x(t) is the displacement function of the operating mechanism; K v is the velocity damping coefficient, which is related to the friction of the operating mechanism; K s I is the spring stiffness coefficient, which is determined by the spring characteristics of the operating mechanism; f is the Coulomb friction current, which is related to the contact surface material; sign() is the sign function; I e is the eddy current amplitude; τ e is the eddy current time constant; ξ(t) is the random noise term, which obeys the normal distribution

[0096] 4. Switch mechanism action voltage equation:

[0097]

[0098] Where U op (t) is the operating voltage of the switch mechanism; U0 is the rated operating voltage; R c is the equivalent resistance of the operating circuit; L c is the equivalent inductance of the operating circuit; K b is the back electromotive force coefficient; μ(t) is the voltage fluctuation function, simulating the grid fluctuation; ω v is the voltage fluctuation angular frequency; ζ(t) is the random noise term, which obeys the normal distribution

[0099] 5. Current and voltage equations during contact opening and closing process:

[0100]

[0101] In the formula, I c (t), U c(t) is the contact current and voltage; I a , U a are the initial current and voltage amplitudes; τ a , τ c is the current and voltage decay time constant; ω a is the current oscillation angular frequency; I b , U b are the steady-state current and voltage components; τ b , τ d is the current and voltage rise time constant; I arc , U arc is the characteristic value of arc current and voltage; K arc is the arc formation rate coefficient; l arc (t) is the arc length function; I0 is a small current constant to prevent zero division errors; γ(t) and δ(t) are random noise terms that obey the normal distribution.

[0102] 6. Switch mechanism displacement equation:

[0103]

[0104] Where x(t) is the displacement of the switch mechanism; ζ is the damping ratio; ω n is the natural angular frequency; F m (t) is the driving force function; m is the mass of the switch mechanism; F f is the friction force; k is the spring stiffness; A e is the external vibration amplitude; ω e is the external vibration angular frequency; ∈(t) is a random disturbance, which obeys the normal distribution

[0105] 7. Switching mechanism speed equation:

[0106]

[0107] Where v(t) is the speed of the switch mechanism; v0 is the initial speed; β is the damping coefficient, which is equal to ζω n ; φ is the phase angle, equal to arctan(ω e / β); ν(t) is the random disturbance of velocity, which obeys the normal distribution

[0108] 8. Switch mechanism acceleration equation:

[0109]

[0110] Where a(t) is the acceleration of the switch mechanism; k(t) is the random disturbance of acceleration, which follows a normal distribution.

[0111] 9. Circuit breaker spring vibration equation:

[0112]

[0113] Where y(t) is the spring displacement; ζ s is the spring-damping ratio; ω s is the natural angular frequency of the spring; F s (t) is the external force acting on the spring; m s is the equivalent mass of the spring; A s is the initial vibration amplitude of the spring; α s is the spring vibration attenuation coefficient; λ(t) is the random disturbance of spring vibration, which obeys the normal distribution

[0114] 10. Circuit breaker damper vibration equation:

[0115]

[0116] Where z(t) is the damper displacement; ζ d is the damping ratio of the damper; ω d is the natural angular frequency of the damper; F d (t) is the external force acting on the damper; m d is the mass of the damper; c d is the nonlinear damping coefficient; A d is the initial vibration amplitude of the damper; α d is the vibration attenuation coefficient of the damper; ρ(t) is the random disturbance of the damper vibration, which obeys the normal distribution

[0117] These equations take into account a variety of complex factors, including nonlinear effects, random disturbances, and system coupling. Each parameter has a physical meaning and can be measured, calculated, or estimated. These equations can more fully describe the electrical and mechanical behavior of the circuit breaker during the opening and closing process.

[0118] The numerical solution process for each equation is described in detail below. The classical fourth-order method (RK4) in the Runge-Kutta method will be used to solve these differential equations. The following is the solution process for each equation:

[0119] 1. Three-phase voltage equation:

[0120] This is an explicit function that does not need to be solved numerically. The voltage value can be directly calculated at a given time point.

[0121] 2. Three-phase current equation:

[0122] It is also an explicit function and can be calculated directly.

[0123] 3. Switch mechanism action current equation:

[0124] This equation depends on the displacement function x(t), and the displacement equation needs to be solved first.

[0125] 4. Switch mechanism action voltage equation:

[0126] This equation also depends on the displacement function x(t) and the operating current I o p(t), we need to solve these two functions first.

[0127] 5. Current and voltage equations during contact opening and closing process:

[0128] These two equations are coupled and need to be solved simultaneously. They can be considered as a two-dimensional system:

[0129] Let y1 = I c ,y2=U c , then:

[0130]

[0131] Solve using the RK4 method:

[0132] Given the initial condition y1(0) = I c (0),y2(0)=U c (0), time step h, for each time step:

[0133]

[0134] 6. Switch mechanism displacement equation:

[0135] This is a second-order ordinary differential equation, which can be transformed into a system of first-order equations:

[0136] Let y1=x,y2=dx / dt, then:

[0137]

[0138] Use the RK4 method to solve, the steps are the same as above.

[0139] 7. Switching mechanism speed equation:

[0140] This equation can be obtained directly from the solution of the displacement equation, that is, v(t)=y2(t).

[0141] 8. Switch mechanism acceleration equation:

[0142] The acceleration can be found by numerically differentiating the velocity. Using the central difference method:

[0143] Where h is the time step.

[0144] 9. Circuit breaker spring vibration equation:

[0145] Similarly, convert the second-order equation into a system of first-order equations:

[0146] Let y1=y,y2=dy / dt, then:

[0147]

[0148] Use the RK4 method to solve, the steps are the same as above.

[0149] 10. Circuit breaker damper vibration equation:

[0150] The same transformation into a first-order system of equations:

[0151] Let y1=z,y2=dz / dt, then:

[0152]

[0153]

[0154] Use the RK4 method to solve, the steps are the same as above.

[0155] For each equation, you need:

[0156] 1. Set initial conditions

[0157] 2. Choose an appropriate time step h

[0158] 3. Define the time interval [0,T]

[0159] 4. Apply the RK4 method to each time step

[0160] 5. Store the results of each time step

[0161] 6. Interpolate as needed to obtain higher precision results

[0162] Optionally, this process needs to be implemented using a programming language (such as Python, MATLAB, etc.) to facilitate iterative calculations and data processing. In practical applications, issues such as numerical stability, computational efficiency, and error control also need to be considered.

[0163] The following provides the process of solving the analytical solution for each equation, as well as the process of calculating the range. Due to the complexity of some equations, it may not be possible to obtain a complete analytical solution. In these cases, approximate solutions or partial analytical solutions will be provided.

[0164] 1. Analytical solution of the three-phase voltage equation:

[0165]

[0166] This equation is already in analytical form and does not require further solution.

[0167] Range calculation: Three-phase voltage peak range: [U m -3σ,U m +3σ], where σ is the standard deviation of ε(t);

[0168] For example, the default initial condition is: m =8200V(peak),f=50Hz,U d c=100V,τ=0.01s; considering the influence of harmonics and DC components, the actual three-phase voltage peak range is: [7800V,8600V].

[0169] 2. Analytical solution of the three-phase current equation:

[0170]

[0171] This equation contains derivative terms and requires solving a differential equation:

[0172] Let y(t) = I φ (t), then:

[0173]

[0174] Using the integrating factor method, the integrating factor is The solution is:

[0175]

[0176] Where C is the integration constant, which is determined by the initial conditions.

[0177] Range calculation: Three-phase current peak range: [I m -3σ I ,I m +3σ I ], where σ I is the standard deviation of η(t);

[0178] For example, the default initial condition is: I m =1000A(peak), I d c=10A,τ I =0.005s,K t =0.001s; considering the influence of harmonics and DC components, the actual three-phase current peak range is: [950A, 1050A].

[0179] 3. Analytical solution of the switch mechanism action current equation:

[0180]

[0181] This equation depends on the displacement function x(t), so we need to solve the displacement equation first. Assuming we already have an analytical solution for x(t), we can substitute it into this equation.

[0182] Due to the nonlinear nature of the sign function, a complete analytical solution may be difficult to obtain. You can consider the form of a piecewise function:

[0183] When dx(t) / dt>0:

[0184]

[0185] When dx(t) / dt<0:

[0186]

[0187] Range calculation: Maximum range of switch mechanism operating current: [I0+I e -3σ o p,I0+I e +K m ·a max +K v ·v max +K s ·x max +I f +3σ o p]; where a max ,v max ,x max are the maximum acceleration, velocity and displacement respectively, v o p is the standard deviation of ξ(t);

[0188] For example, the preset initial conditions are: I0=50A,K m =0.1kg,K v =10N·s / m,K s =1000N / m,I f =5A,I e =20A,τ e =0.01s, the actual maximum operating current range of the switch mechanism is: [65A, 150A].

[0189] 4. Analytical solution of the switch mechanism action voltage equation:

[0190]

[0191] This equation depends on I op (t) and x(t). Assuming that the analytical solutions of these two functions have been obtained, substitute them into this equation.

[0192] For μ(t), assume it is a constant μ. Then the analytical solution is:

[0193]

[0194] Range calculation: Maximum range of switch mechanism action voltage: [U0-R c I max -L c (dI / dt) max -K b ·a max -μU0-3σ U ,U0+3σ U ] Among them I max ,(dI / dt) max ,a max are the maximum current, maximum current change rate and maximum acceleration, σ U is the standard deviation of ζ(t);

[0195] For example, the preset initial conditions are: U0=220V,R c =0.1Ω,L c =0.001H,K b =0.05V·s2 / m,μ=0.05,ω v =100πrad / s; then the maximum operating voltage range of the switch mechanism is: [180V, 230V].

[0196] 5. Analytical solution of the current and voltage equations during the contact opening and closing process:

[0197] Current equation:

[0198] I c (t) = I a e -t / τa sin(ω a t)+I b (1-e -t / τb )+I arc tanh(K arc t)+γ(t);

[0199] This equation is already in analytical form and does not need to be solved further.

[0200] Voltage equation:

[0201]

[0202] This equation depends on I c (t) and l arc (t). Assume that arc (t) = l0 + varc ·t, where l0 is the initial arc length, v arc is the arc extension speed.

[0203] Range calculation: Maximum current range of contact opening and closing process: [I b -3σ γ ,I a +I b +I arc +3σ γ ], where σ γ is the standard deviation of γ(t); the maximum voltage range of the contact opening and closing process: [U b -3σ δ ,U a +U b +U arc ·l max / (I min +I0)+3σ δ ], where σ δ is the standard deviation of δ(t), l max is the maximum arc length, I min is the minimum current value;

[0204] For example, the default initial condition is: I a =1000A,τ a =0.005s,ω a =1000πrad / s,I b =500A,τ b =0.01s,I arc =100A,K arc =100s-1, U a =1000V,τ c =0.005s,U b =500V,τ d =0.01s,U arc =20V,l0=0.001m,v arc =10m / s, I0=1A, then the maximum current range of the actual contact opening and closing process is: [450A, 1650A].

[0205] 6. Analytical solution of the switch mechanism displacement equation:

[0206]

[0207] This is a nonlinear second-order differential equation, and a complete analytical solution is difficult to obtain. Linear approximation or piecewise linearization can be considered.

[0208] Assume that the friction term can be approximated as linear damping, that is:

[0209] make Then the equation is simplified to:

[0210]

[0211] This is a standard second-order linear differential equation, and its analytical solution is:

[0212]

[0213] in C1 and C2 are determined by the initial conditions, x p (t) is a particular solution, depending on the specific form of F(t).

[0214] Range calculation: Maximum displacement range of the switch mechanism: [-x max ,x max ], where x max It can be obtained by solving the equation numerically;

[0215] For example, the preset initial conditions are: ζ=0.1,ω n =100rad / s,m=10kg,F f =50N,k=10000N / m,A e =0.1m,ω e =50rad / s; then the maximum displacement range of the switch mechanism is: [-0.05m, 0.05m].

[0216] 7. Analytical solution of the switch mechanism speed equation:

[0217] Velocity is the derivative of displacement, so:

[0218]

[0219] Range calculation: Maximum speed range of the switch mechanism: [-v max ,v max ], where v max The equation can be solved by numerical method to obtain, for example, the maximum speed range of the switch mechanism is [-5 m / s, 5 m / s].

[0220] 8. Analytical solution of the acceleration equation of the switch mechanism:

[0221] Acceleration is the derivative of velocity and can be obtained by differentiating the velocity equation again. Since the expression is relatively complicated, the specific form is omitted here.

[0222] Range calculation: Maximum acceleration range of the switch mechanism: [-a max ,a max ], where a maxIt can be solved by numerical method to obtain, for example, [-500 m / s 2 ,500 m / s 2 ].

[0223] 9. Analytical solution of circuit breaker spring vibration equation:

[0224]

[0225] This is a linear second-order differential equation whose homogeneous solutions are:

[0226]

[0227] in D_1 and D_2 are determined by the initial conditions.

[0228] The specific solution depends on the specific form of F_s(t). Assuming F_s(t) = F_0 (constant), the specific solution form is:

[0229]

[0230] Where B1 and B2 are constants that can be substituted into the original equation to solve. λ (t) is the particular solution caused by the random perturbation λ(t).

[0231] The complete solution is: y(t) = y h (t)+y p (t);

[0232] Range calculation: Circuit breaker spring vibration frequency range: [ω s (1-3σ ω ),ω s (1+3σ ω )], where σ ω is the standard deviation of the frequency;

[0233] For example, the default initial condition is: s =0.05,ω s =200rad / s,m s =1kg,F0=100N,A s =0.01m,α s =10s - 1, then the vibration frequency range of the circuit breaker spring is [190 rad / s, 210 rad / s].

[0234] 10. Analytical solution of circuit breaker damper vibration equation:

[0235]

[0236] This is a nonlinear equation and a complete analytical solution is difficult to obtain. We can consider a linear approximation:

[0237] Assume that the nonlinear damping term can be approximated by linear damping:

[0238] The simplified equation is:

[0239]

[0240] The homogeneous solution is:

[0241]

[0242] in E_1 and E_2 are determined by the initial conditions.

[0243] The particular solution depends on the specific form of F_d(t). Assuming F_d(t) = F_d0 (constant), the particular solution is:

[0244]

[0245] Where C1 and C2 are constants that can be substituted into the original equation to solve. p (t) is a particular solution caused by the random perturbation ρ(t).

[0246] The complete solution is: z(t) = z h (t)+z p (t);

[0247] Range calculation: Circuit breaker damper vibration amplitude range: [-z max ,z max ], where z max It can be obtained by solving the equation numerically;

[0248] For example, the default initial condition is: d =0.1,ω d =150rad / s,m d =2kg,F d 0=200N,c d =0.5N·s 2 / m 2 ,A d =0.02m,α d =5s - 1; the actual vibration amplitude range of the circuit breaker damper is [-0.03 m, 0.03 m].

[0249] Summarize:

[0250] 1. For linear equations, complete analytical solutions can be obtained.

[0251] 2. For nonlinear equations, linearization approximation is used to obtain approximate analytical solutions.

[0252] 3. In practical applications, it may be necessary to combine numerical methods to obtain more accurate solutions and range estimates.

[0253] 4. The range calculation takes into account the impact of random disturbances, and the 3σ principle is usually used to estimate the range.

[0254] 5. The preset initial conditions and parameter values ​​are based on assumptions and need to be adjusted according to the specific circuit breaker characteristics in actual applications.

[0255] 6. For complex coupled systems, it may be necessary to consider the mutual influence between the equations, which may require more advanced mathematical tools or numerical methods to handle.

[0256] These analytical solutions and range calculations provide a theoretical basis for the fault mode identification of circuit breaker switch opening and closing. In practical applications, it may be necessary to combine experimental data and statistical analysis to further optimize the model and parameter estimation.

[0257] Specifically, the principle of the present invention is:

[0258] First, this method adopts the idea of ​​multi-source data fusion. By simultaneously collecting electrical quantity data and mechanical quantity data, comprehensive monitoring of the circuit breaker status is achieved. Electrical quantity data includes three-phase voltage, current, and electrical parameters of the switch mechanism and contacts, which can reflect the electrical performance and breaking capacity of the circuit breaker. Mechanical quantity data includes the motion parameters of the switch mechanism and the vibration signals of the spring and damper, which can reflect the mechanical properties and operating characteristics of the circuit breaker. The comprehensive analysis of multi-source data can more comprehensively and accurately reflect the operating status of the circuit breaker, effectively avoiding the one-sidedness and inaccuracy that may be caused by a single data source.

[0259] Secondly, this method establishes an accurate mathematical model. By establishing a set of electrical quantity equations and a set of mechanical quantity equations, the dynamic characteristics of the circuit breaker during the opening and closing process are accurately described. These equations take into account a variety of complex factors, such as nonlinear effects, random disturbances, and system coupling, and can more realistically reflect the physical process of the circuit breaker. The numerical and analytical solutions of the equations not only obtain the specific values ​​of the key parameters, but also the theoretical range of variation of these parameters. This dual solution method provides more fault characteristic information, which is conducive to improving the accuracy and reliability of fault diagnosis.

[0260] Thirdly, this method uses advanced data processing and feature extraction techniques. Through normalization, the influence of dimensions between different parameters is eliminated, making each feature comparable in subsequent analysis. The application of principal component analysis (PCA) technology realizes dimensionality reduction and feature extraction of high-dimensional data. PCA can find the main direction of change in the data, retain the most important information, and remove redundancy and noise. This not only reduces the computational complexity of subsequent processing, but also improves the representativeness and discrimination of the features.

[0261] Finally, this method uses a multilayer perceptron neural network as a fault identification model. Neural networks have powerful nonlinear mapping and self-learning capabilities and can effectively handle complex fault modes. Through multi-layer structures and nonlinear activation functions, neural networks can learn and represent complex feature combinations, adapt to the diversity and complexity of circuit breaker fault modes, and enable the model to directly output the probabilities of various fault modes, providing reliable quantitative indicators for fault diagnosis.

[0262] The technical solution of the present invention is logical, which is mainly reflected in the following aspects:

[0263] 1. From data collection to feature extraction and then to fault identification, a complete fault diagnosis process is constructed, with each link closely connected and logically rigorous.

[0264] 2. The electrical and mechanical characteristics of the circuit breaker are comprehensively considered, which conforms to the working principle and failure mechanism of the circuit breaker.

[0265] 3. The mathematical model and data processing method adopted have a solid theoretical basis and can effectively extract fault characteristics.

[0266] 4. The machine learning model used is suitable for handling complex nonlinear problems and matches the needs of circuit breaker fault diagnosis.

[0267] In summary, the technical solution of the present invention constructs a comprehensive, accurate and efficient circuit breaker fault diagnosis system through the organic combination of multi-source data fusion, precise mathematical modeling, advanced data processing and intelligent recognition algorithm, which can effectively solve the problems existing in the prior art and improve the accuracy and reliability of circuit breaker fault diagnosis.

[0268] In order to better understand and implement the present invention, an embodiment of a specific application scenario of the present invention is provided below: a distribution station has 10 circuit breakers of model ZN63A-12 and rated current of 1250A.

[0269] 1. Installation and configuration of data acquisition system

[0270] First, the company installed a data acquisition system on each circuit breaker. The system included the following sensors:

[0271] Three-phase voltage transformer: model JDZX9-10, transformation ratio 10000 / 100V;

[0272] Three-phase current transformer: model LZZBJ9-10, ratio 1250 / 5A;

[0273] Switch mechanism displacement sensor: model WY-LD, range 0-100mm;

[0274] Switch mechanism speed sensor: model WY-CS, range ±10m / s;

[0275] Switch mechanism acceleration sensor: Model WY-CA, range ±500m / s 2 ;

[0276] Spring vibration sensor: model WY-SV, frequency range 0-1000Hz;

[0277] Damper vibration sensor: model WY-DV, amplitude range is ±50mm;

[0278] All sensors are connected to a central data acquisition unit, model WY-DAQ1000. The sampling rate of the unit is set to 10kHz to ensure that fast transient changes during circuit breaker operation are captured.

[0279] 2. Data Collection Process

[0280] In a routine operation test, the operator opened and closed the circuit breaker No. 1. The data acquisition system recorded various parameters during the entire process. The following is part of the collected data (the data has been preliminarily processed and the sampling interval is 1ms) as shown in Table 1:

[0281] Table 1 Collection data table

[0282]

[0283]

[0284] (Note: The complete data set contains 1000 data points, covering the entire opening and closing process)

[0285] 3. Equation establishment and solution

[0286] Based on the collected data, the system established a group of electrical quantity equations and a group of mechanical quantity equations.

[0287] These equations were solved using a numerical method (fourth-order Runge–Kutta method), and both numerical and analytical solutions for each parameter were obtained.

[0288] 4. Determine the parameter range

[0289] By substituting the preset initial conditions into the analytical solution, the system obtains the range of each parameter:

[0290] Three-phase voltage peak range: [7800V, 8600V];

[0291] Three-phase current peak range: [950A, 1050A];

[0292] Maximum range of switch mechanism operating current: [65A, 150A];

[0293] Maximum range of switch mechanism operating voltage: [180V, 230V];

[0294] Maximum current range during contact opening and closing: [450A, 1650A];

[0295] Maximum voltage range of contact opening and closing process: [450V, 1600V];

[0296] Maximum displacement range of the switch mechanism: [-0.05m, 0.05m];

[0297] Maximum speed range of the switch mechanism: [-5 m / s, 5 m / s];

[0298] Maximum acceleration range of the switch mechanism: [-500 m / s 2 ,500 m / s 2 ];

[0299] Circuit breaker spring vibration frequency range: [190 rad / s, 210 rad / s];

[0300] Circuit breaker damper vibration amplitude range: [-0.03 m, 0.03 m];

[0301] 5. Data normalization and feature extraction

[0302] The system normalizes the obtained numerical solutions and the maximum and minimum values ​​of the range so that all parameter values ​​fall within the interval [0,1]. Then, the system uses the principal component analysis (PCA) method to extract features from the normalized data.

[0303] The PCA analysis results show that the first five principal components can explain 95% of the data variance. The system retains these five principal components and forms a 5x1000 feature matrix.

[0304] 6. Application of Fault Mode Recognition Model

[0305] A circuit breaker switch opening and closing fault mode recognition model based on support vector machine (SVM) has been trained. The model can identify the following 10 fault modes:

[0306] Normal operation, mechanical jamming, spring fatigue, damper failure, contact erosion, operating mechanism failure, auxiliary switch failure, coil failure, secondary circuit failure, insulation deterioration.

[0307] The system inputs the extracted feature matrix into this pre-trained SVM model. The model outputs a 10-dimensional failure mode vector:

[0308] [0.02,0.92,0.01,0.01,0.01,0.01,0.01,0.01,0.00,0.00];

[0309] This vector represents the probability of various failure modes. According to the output results, the system determines that the operating state of circuit breaker No. 1 is mechanical jamming with a probability of 92%. Since the set safety threshold is 0.8, it means that the circuit breaker has low safety.

[0310] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A switch safety identification method for a circuit breaker device, characterized in that: The following steps are involved: S10, collecting electrical quantity data and mechanical quantity data during the opening and closing process of the circuit breaker switch, wherein the electrical quantity data includes: three-phase voltage and current waveform of the circuit breaker, operating current and voltage waveform of the switch mechanism, and current and voltage parameters of the opening and closing process of the contact; the mechanical quantity data includes displacement, speed, acceleration curve of the switch mechanism, and vibration signals of the spring and damper of the circuit breaker; S20, establishing an electrical quantity equation group, including: a three-phase voltage equation, a three-phase current equation, a switch mechanism action current equation, a switch mechanism action voltage equation, and a contact opening and closing process current and voltage equation; S30, establishing a mechanical quantity equation group, including: a switch mechanism displacement equation, a switch mechanism velocity equation, a switch mechanism acceleration equation, a circuit breaker spring vibration equation, and a circuit breaker damper vibration equation; S40, solving the electrical quantity equation group to obtain multiple electrical quantity numerical solutions, which are three-phase voltage peak value, three-phase current peak value, maximum value of switch mechanism action current, maximum value of switch mechanism action voltage, maximum value of contact opening and closing process current, and maximum value of contact opening and closing process voltage; S50, solving the mechanical quantity equation group to obtain multiple mechanical quantity numerical solutions, which are respectively the maximum displacement of the switch mechanism, the maximum speed of the switch mechanism, the maximum acceleration of the switch mechanism, the vibration frequency of the circuit breaker spring, and the vibration amplitude of the circuit breaker damper; S60, analytically solving the electrical quantity equation group to obtain multiple electrical quantity analytical solutions, substituting into preset electrical quantity initial conditions to obtain multiple electrical quantity ranges; S70, analytically solving the mechanical quantity equation group to obtain multiple mechanical quantity analytical solutions, substituting into preset mechanical quantity initial conditions to obtain multiple mechanical quantity ranges; S80, normalizing the multiple numerical solutions of electrical quantities, the multiple numerical solutions of mechanical quantities, the maximum and minimum values ​​of the multiple ranges of electrical quantities, and the maximum and minimum values ​​of the multiple ranges of mechanical quantities so that each parameter value falls within the interval [0,1], concatenating the normalized data and extracting features through principal component analysis or dimensionality reduction method to obtain a feature matrix; S90. Input the feature matrix into a pre-trained circuit breaker switch opening and closing fault mode recognition model to obtain a circuit breaker switch opening and closing fault mode vector, and compare the highest fault probability in the fault mode vector with a pre-set safety threshold. If the highest fault probability is greater than the safety threshold, it is determined that the circuit breaker switch safety is low; otherwise, it is determined that the circuit breaker switch safety is high.

2. A switch safety identification method for a circuit breaker device according to claim 1, characterized in that: The multiple analytical solutions of electrical quantities are: an analytical expression for three-phase voltage, an analytical expression for three-phase current, an analytical expression for switch mechanism action current, an analytical expression for switch mechanism action voltage, an analytical expression for contact opening and closing process current, and an analytical expression for contact opening and closing process voltage.

3. A switch safety identification method for a circuit breaker device according to claim 2, characterized in that: The multiple electrical quantity ranges are: three-phase voltage peak range, three-phase current peak range, switch mechanism action current maximum range, switch mechanism action voltage maximum range, contact opening and closing process current maximum range, and contact opening and closing process voltage maximum range.

4. A switch safety identification method for circuit breaker equipment according to claim 3, characterized in that: The multiple analytical solutions of electrical quantities are: an analytical expression for three-phase voltage, an analytical expression for three-phase current, an analytical expression for switch mechanism action current, an analytical expression for switch mechanism action voltage, an analytical expression for contact opening and closing process current, and an analytical expression for contact opening and closing process voltage.

5. A switch safety identification method for circuit breaker equipment according to claim 4, characterized in that: The multiple mechanical quantity ranges are: a maximum displacement range of a switch mechanism, a maximum speed range of a switch mechanism, a maximum acceleration range of a switch mechanism, a vibration frequency range of a circuit breaker spring, and a vibration amplitude range of a circuit breaker damper.

6. A switch safety identification method for circuit breaker equipment according to claim 5, characterized in that: The three-phase voltage equation is specifically expressed as: Where U φ (t) is the voltage of a certain phase changing with time; U m is the fundamental amplitude, which is measured by voltage transformer; ω is the angular frequency, which is equal to 2πf, where f is the system frequency; t is the time variable; θ φ is the initial phase angle, determined by the synchronous measurement device; U n is the amplitude of the nth harmonic, obtained through Fourier analysis; θ n is the initial phase angle of the nth harmonic; N is the highest harmonic number considered; U dc is the DC bias component, detected by the voltage measurement device; τ is the time constant, which is related to the system impedance; ε(t) is the random noise term; The three-phase current equation is specifically expressed as: In the formula, I φ (t) is the time-varying function of the current of a certain phase; I m is the fundamental amplitude, measured by current transformer; φ is the power factor angle, measured by power analyzer; I n is the amplitude of the nth harmonic, obtained by Fourier analysis; φ n is the phase difference of the nth harmonic; I dc is the DC bias component, detected by the current measuring device; τ I is the current time constant, which is related to the system impedance and inductance; K t is the eddy current coefficient, which is related to the circuit breaker core material; η(t) is the random noise term.

7. A switch safety identification method for a circuit breaker device according to claim 6, characterized in that: The switch mechanism action current equation is specifically expressed as: In the formula, I op (t) is the operating current of the switch mechanism; I0 is the initial current, which is determined by the initial state of the operating mechanism; K m is the mass coefficient, which is related to the mass of the moving parts of the operating mechanism; x(t) is the displacement function of the operating mechanism; K v is the velocity damping coefficient, which is related to the friction of the operating mechanism; K s I is the spring stiffness coefficient, which is determined by the spring characteristics of the operating mechanism; f is the Coulomb friction current, which is related to the contact surface material; sign() is the sign function; I e is the eddy current amplitude; τ e is the eddy current time constant; ξ(t) is the random noise term; The switching mechanism action voltage equation is specifically expressed as: Where U op (t) is the operating voltage of the switch mechanism; U0 is the rated operating voltage; R c is the equivalent resistance of the operating circuit; L c is the equivalent inductance of the operating circuit; K b is the back electromotive force coefficient; μ(t) is the voltage fluctuation function, simulating the grid fluctuation; ω v is the voltage fluctuation angular frequency; ζ(t) is the random noise term.

8. A switch safety identification method for circuit breaker equipment according to claim 7, characterized in that: The current and voltage equation of the contact opening and closing process is specifically expressed as: In the formula, I c (t), U c (t) is the contact current and voltage; I a , U a are the initial current and voltage amplitudes; τ a , τ c is the current and voltage decay time constant; ω a is the current oscillation angular frequency; I b , U b are the steady-state current and voltage components; τ b , τ d is the current and voltage rise time constant; I arc , U arc is the characteristic value of arc current and voltage; K arc is the arc formation rate coefficient; l arc (t) is the arc length function; I0 is a small current constant to prevent division by zero error; γ(t) and δ(t) are random noise terms.

9. A switch safety identification method for circuit breaker equipment according to claim 8, characterized in that: The displacement equation of the switch mechanism is specifically expressed as: Where x(t) is the displacement of the switch mechanism; ζ is the damping ratio; ω n is the natural angular frequency; F m (t) is the driving force function; m is the mass of the switch mechanism; F f is the friction force; k is the spring stiffness; A e is the external vibration amplitude; ω e is the external vibration angular frequency;∈(t) is the random disturbance; The speed equation of the switch mechanism is specifically expressed as: Where v(t) is the speed of the switch mechanism; v0 is the initial speed; β is the damping coefficient, which is equal to ζω n ; φ is the phase angle, equal to arctan(ω e / β); ν(t) is the random disturbance of velocity; The acceleration equation of the switch mechanism is specifically expressed as: Where a(t) is the acceleration of the switch mechanism; κ(t) is the random disturbance of acceleration; The circuit breaker spring vibration equation is specifically expressed as: Where y(t) is the spring displacement; ζ s is the spring-damping ratio; ω s is the natural angular frequency of the spring; F s (t) is the external force acting on the spring; m s is the equivalent mass of the spring; A s is the initial vibration amplitude of the spring; α s is the spring vibration attenuation coefficient; λ(t) is the random disturbance of spring vibration; The circuit breaker damper vibration equation is specifically expressed as: Where z(t) is the damper displacement; ζ d is the damping ratio of the damper; ω d is the natural angular frequency of the damper; F d (t) is the external force acting on the damper; m d is the mass of the damper; c d is the nonlinear damping coefficient; A d is the initial vibration amplitude of the damper; α d is the vibration attenuation coefficient of the damper; ρ(t) is the random disturbance of the damper vibration.

10. A switch safety identification method for circuit breaker equipment according to claim 9, characterized in that: The training steps of the circuit breaker switch opening and closing fault mode recognition model specifically include: Establishing a training data set, specifically obtaining multiple sets of electrical quantity data and mechanical quantity data during the opening and closing process of circuit breaker switches with known faults, and generating a feature matrix in accordance with steps S10 to S80, wherein the input of the training is the generated feature matrix, and the output of the training is the known fault; Network training, using the training data set to train a multi-layer perceptron network to obtain the circuit breaker switch opening and closing fault mode recognition model.

Images