An intelligent diagnosis method for abnormal operation state of power distribution network
Patent Information
- Application Number
- CN202611467471.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-21
- Publication Date
- 2026-10-27
AI Technical Summary
[0005]有鉴于此，本发明实施例提供了一种配电网的异常运行状态智能诊断方法，以解决如何对配电网的异常状态进行准确诊断的问题
[0038]在本发明中，首先获取当前周期的等效线性电导和等效分布电容，进而获取当前周期内每一时刻的瞬态非线性残余电流，有效剔除了幅值较高的线性容性充放电电流和线性欧姆泄漏电流的干扰，孤立出主要由高压电弧相变或非线性负载引发的非线性物理响应，解决了传统基于幅值检测的常规方法被容性充放电电流覆盖的问题；然后获取当前周期与当前周期的上一周期之间的波形互相关系数，利用所述波形互相关系数，获取当前周期的瞬态热力学混沌指数，将微观的波形形态畸变体现为宏观的热力学混沌物理特征，能够准确识别出由等离子体热力学混沌演变引发的非线性电流，从而有效区分合规的充电桩稳态谐波与具有安全风险的高阻抗接地电弧；最后，获取当前周期的非线性残余电流积分特征值，利用当前周期的非线性残余电流积分特征值和当前周期的瞬态热力学混沌指数，构建双重物理边界判定机制，既避免了对所有微小级别的随机波动均进行混沌特征的确诊与判定而增加底层控制器的算力负荷，又防范了量测噪声或环境电磁干扰引发的虚假跳闸报警风险，显著提高了配电网异常运行状态诊断的准确性、安全性与可靠性。
Smart Images
Figure QLYQS_1 
Figure QLYQS_7 
Figure QLYQS_13
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to an intelligent diagnostic method for abnormal operating conditions of a power distribution network. Background Technology
[0002] In modern power distribution network operations, the use of long-distance underground cables is increasing, and a large number of power electronic loads, such as electric vehicle charging piles and AC / DC rectifier equipment, are connected to the end of the distribution network. When abnormal conditions occur, such as tree branches touching the line or damage to contaminated insulators (insulators with dirt on their surface), high-voltage breakdown discharge is likely to occur at the fault point, forming a high-impedance grounding arc, which brings serious safety hazards.
[0003] Existing technologies primarily diagnose abnormal operating conditions of distribution networks by collecting electrical signals from distribution network lines, extracting the amplitude of the fundamental component, and comparing it with a preset threshold. However, these technologies have the following drawbacks in practical applications: First, long-distance underground cables have significant inherent distributed capacitance to the ground. Normal capacitor charging and discharging processes generate high-amplitude linear capacitive charging and discharging currents. Combined with linear ohmic leakage currents flowing through trees or polluted insulators, these current components, conforming to linear physical laws, directly overshadow the weak nonlinear physical response caused by high-voltage breakdown. This renders conventional amplitude-based detection methods ineffective, failing to identify weak high-impedance grounding arcs. Second, some improved solutions, designed to circumvent interference from linear capacitive charging and discharging currents, attempt to extract only nonlinear current characteristics for diagnosis. However, in modern highly electronic distribution networks, compliant electric vehicle charging piles and other power electronic loads also inject nonlinear current components into the grid during normal operation. Relying solely on the magnitude of the extracted nonlinear current cannot distinguish between the steady-state harmonics of compliant charging piles and high-impedance grounding arcs posing safety risks, easily triggering false trip alarms and affecting the accuracy and reliability of the diagnosis.
[0004] Therefore, how to accurately diagnose abnormal conditions in the power distribution network has become an urgent problem to be solved. Summary of the Invention
[0005] In view of this, embodiments of the present invention provide an intelligent diagnostic method for abnormal operating states of a power distribution network, in order to solve the problem of how to accurately diagnose abnormal states of a power distribution network.
[0006] This invention provides an intelligent diagnostic method for abnormal operating conditions of a power distribution network, the method comprising the following steps:
[0007] The transient zero-sequence voltage and transient zero-sequence current of the distribution network at each moment are obtained, and the preset time period up to the current moment is recorded as the current period.
[0008] Based on the changing characteristics of transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, the equivalent linear conductance and equivalent distributed capacitance of the current cycle are obtained. Based on the equivalent linear conductance and equivalent distributed capacitance of the current cycle, as well as the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, the transient nonlinear residual current at each moment in the current cycle is obtained.
[0009] Obtain the transient nonlinear residual current at each moment in the previous cycle of the current cycle. Based on the correlation characteristics between the transient nonlinear residual current at each moment in the current cycle and the transient nonlinear residual current at each moment in the previous cycle of the current cycle, obtain the waveform cross-correlation coefficient between the current cycle and the previous cycle of the current cycle. Use the waveform cross-correlation coefficient to obtain the transient thermodynamic chaos index of the current cycle.
[0010] Based on the transient nonlinear residual current at each moment in the current cycle, the integral characteristic value of the nonlinear residual current in the current cycle is obtained. Using the integral characteristic value of the nonlinear residual current in the current cycle and the transient thermodynamic chaos index in the current cycle, the distribution network anomaly diagnosis is performed.
[0011] Preferably, based on the variation characteristics of the transient zero-sequence voltage and transient zero-sequence current at each moment within the current cycle, the equivalent linear conductance and equivalent distributed capacitance of the current cycle are obtained, including:
[0012] Based on the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, obtain the equivalent linear conductance of the current cycle;
[0013] For any moment within the current period, the difference between the transient zero-sequence voltage at any moment and the transient zero-sequence voltage at the previous moment is obtained to obtain the transient zero-sequence voltage difference. The interval time between any moment and the previous moment is obtained, and the ratio of the transient zero-sequence voltage difference to the interval time is obtained to obtain the derivative of the transient zero-sequence voltage at any moment.
[0014] Obtain the derivative of the transient zero-sequence voltage at each moment in the current cycle. Based on the derivatives of the transient zero-sequence current and the transient zero-sequence voltage at each moment in the current cycle, obtain the equivalent distributed capacitance of the current cycle.
[0015] Preferably, the equivalent linear conductance for the current period is obtained based on the transient zero-sequence voltage and transient zero-sequence current at each moment within the current period, including:
[0016] The formula for calculating the equivalent linear conductance of the current period is:
[0017] ;
[0018] in, The equivalent linear conductance for the current period, For the current period, the first The transient zero-sequence current at time t. For the current period, the first The transient zero-sequence voltage at time t is the current time, and T is the preset time period.
[0019] Preferably, the equivalent distributed capacitance of the current period is obtained based on the derivatives of the transient zero-sequence current and the transient zero-sequence voltage at each moment within the current period, including:
[0020] The formula for calculating the equivalent distributed capacitance of the current period is:
[0021] ;
[0022] in, The equivalent distributed capacitance for the current period. For the current period, the first The transient zero-sequence current at time t. For the current period, the first The transient zero-sequence voltage at time t. For the current period, the first The derivative of the transient zero-sequence voltage at time t, where t is the current time and T is the preset time period.
[0023] Preferably, the transient nonlinear residual current at each moment in the current period is obtained based on the equivalent linear conductance and equivalent distributed capacitance of the current period, as well as the transient zero-sequence voltage and transient zero-sequence current at each moment in the current period, including:
[0024] For any moment within the current cycle, the product of the equivalent linear conductance of the current cycle and the transient zero-sequence voltage at any moment is obtained to obtain the purely resistive leakage current at any moment.
[0025] The product of the equivalent distributed capacitance of the current period and the derivative of the transient zero-sequence voltage at any given moment is obtained to get the purely capacitive charging and discharging current at any given moment.
[0026] The difference between the transient zero-sequence current at any given time and the purely resistive leakage current at any given time is obtained to obtain a first difference. The difference between the first difference and the purely capacitive charging and discharging current at any given time is obtained to obtain the transient nonlinear residual current at any given time.
[0027] Preferably, based on the correlation characteristics between the transient nonlinear residual current at each moment in the current cycle and the transient nonlinear residual current at each moment in the previous cycle, the waveform cross-correlation coefficient between the current cycle and the previous cycle is obtained, including:
[0028] The formula for calculating the cross-correlation coefficient between the current period and the previous period is:
[0029] ;
[0030] in, This represents the cross-correlation coefficient between the current period and the previous period. For the current period, the first The transient nonlinear residual current at time t. The first cycle in the previous cycle of the current cycle The transient nonlinear residual current at time t is the current time and T is the preset time period.
[0031] Preferably, the transient thermodynamic chaos index of the current period is obtained using the cross-correlation coefficient of the waveform, including:
[0032] The difference between the constant 1 and the cross-correlation coefficient of the waveform is obtained to obtain the transient thermodynamic chaos index of the current period.
[0033] Preferably, the integral characteristic value of the nonlinear residual current for the current period is obtained based on the transient nonlinear residual current at each moment within the current period, including:
[0034] Based on the current period, the square of the transient nonlinear residual current at each moment within the current period is integrated to obtain the integral characteristic value of the nonlinear residual current in the current period.
[0035] Preferably, distribution network anomaly diagnosis is performed using the integral characteristic value of the nonlinear residual current in the current period and the transient thermodynamic chaos index in the current period, including:
[0036] If the integral characteristic value of the nonlinear residual current in the current period is greater than the preset threshold value of the integral characteristic value of the nonlinear residual current, and the transient thermodynamic chaos index in the current period is greater than the preset threshold value of the transient thermodynamic chaos index, then the distribution network is confirmed to have a high-impedance grounding arc anomaly.
[0037] The beneficial effects of the embodiments of the present invention compared with the prior art are as follows:
[0038] In this invention, the equivalent linear conductance and equivalent distributed capacitance of the current cycle are first obtained, and then the transient nonlinear residual current at each moment within the current cycle is obtained. This effectively eliminates the interference of high-amplitude linear capacitive charging and discharging currents and linear ohmic leakage currents, isolating the nonlinear physical response mainly caused by high-voltage arc phase transitions or nonlinear loads. This solves the problem that conventional methods based on amplitude detection are covered by capacitive charging and discharging currents. Then, the waveform cross-correlation coefficient between the current cycle and the previous cycle is obtained. Using this cross-correlation coefficient, the transient thermodynamic chaos index of the current cycle is obtained, transforming microscopic waveform distortion into macroscopic thermodynamic chaotic physical characteristics. It can accurately identify nonlinear currents caused by chaotic evolution of plasma thermodynamics, thus effectively distinguishing between compliant steady-state harmonics of charging piles and high-impedance grounding arcs with safety risks. Finally, it obtains the integral characteristic value of the nonlinear residual current in the current period, and uses the integral characteristic value of the nonlinear residual current in the current period and the transient thermodynamic chaos index in the current period to construct a dual physical boundary judgment mechanism. This avoids increasing the computing load on the underlying controller by diagnosing and judging the chaotic characteristics of all small-level random fluctuations, and also prevents the risk of false trip alarms caused by measurement noise or environmental electromagnetic interference. It significantly improves the accuracy, safety and reliability of abnormal operation status diagnosis of distribution network. Detailed Implementation
[0039] The embodiments of this disclosure are described in detail below. It should be noted that the terms "first," "second," etc., used in this specification are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those described herein. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure.
[0040] To illustrate the technical solution of the present invention, specific embodiments are described below.
[0041] Embodiment 1 of the present invention provides an intelligent diagnostic method for abnormal operating states of a power distribution network, the method comprising:
[0042] Step S101: Obtain the transient zero-sequence voltage and transient zero-sequence current of the distribution network at each moment, and record the preset time period up to the current moment as the current period.
[0043] In modern power distribution network operations, existing technologies primarily diagnose abnormal operating conditions by collecting electrical signals from power distribution lines, extracting the fundamental frequency component amplitude, and comparing it with a preset threshold. However, existing amplitude detection methods are susceptible to interference from the linear capacitive charging and discharging current and linear ohmic leakage current of long-distance cables, making it impossible to detect weak high-impedance grounding arcs. Furthermore, when extracting nonlinear features, existing technologies cannot distinguish between nonlinear currents injected by compliant power electronic loads and high-impedance grounding arcs posing safety risks, easily generating false trip alarms and affecting the accuracy and reliability of the diagnosis.
[0044] Therefore, after acquiring relevant data from the distribution network, this embodiment of the invention establishes a data monitoring cycle, obtains the equivalent linear conductance and equivalent distributed capacitance of the current cycle based on the relevant data, and then obtains the transient nonlinear residual current at each moment within the current cycle, effectively eliminating interference from linear capacitive charging and discharging currents and linear ohmic leakage currents with high amplitudes; then, it obtains the waveform cross-correlation coefficient between the current cycle and the previous cycle, and then obtains the transient thermodynamic chaos index of the current cycle, effectively distinguishing between compliant charging pile steady-state harmonics and high-impedance grounding arcs with safety risks; finally, it obtains the integral characteristic value of the nonlinear residual current of the current cycle, and uses the integral characteristic value of the nonlinear residual current of the current cycle and the transient thermodynamic chaos index of the current cycle to construct a dual physical boundary judgment mechanism, significantly improving the accuracy, safety, and reliability of abnormal operation state diagnosis of the distribution network.
[0045] First, relevant data from the distribution network is acquired. Specifically, analog electrical signals are collected through zero-sequence voltage and current transformers at the feeder outlet of the distribution network. After passing through a low-pass anti-aliasing filter circuit, the signals are synchronously converted into digital signals using an analog-to-digital converter at a preset sampling time step, driven by the high-frequency clock of the underlying field-programmable gate array (FPGA). This allows the acquisition of the transient zero-sequence voltage and transient zero-sequence current at each moment at the feeder outlet of the distribution network. The preset sampling time step is calibrated and acquired based on the Nyquist sampling theorem and the capacity limitations of the underlying memory, with a range of 100 microseconds to 250 microseconds. In this embodiment, to fully reconstruct the real waveform containing higher harmonics and distortion characteristics within the framework of the Nyquist sampling theorem, while also considering the capacity limitations of the underlying memory, the preset sampling time step is precisely set to 200 microseconds, corresponding to a sampling frequency of 5000 Hz (5 kHz). This is not a limitation and can be set according to the specific implementation scenario.
[0046] After obtaining the transient zero-sequence voltage and transient zero-sequence current of the distribution network at each moment, the preset time period up to the current moment is recorded as the current cycle. This provides complete data support for identifying compliant nonlinear loads and high-impedance grounding arcs with safety risks in subsequent processes. In this embodiment, the preset time period includes 100 sampling moments, or 0.02 seconds, which ensures the fidelity of waveform details in subsequent calculations while controlling the computational load of the underlying hardware. This is not limited here and can be set according to the specific implementation scenario.
[0047] Step S102: Based on the changing characteristics of the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, obtain the equivalent linear conductance and equivalent distributed capacitance of the current cycle. Based on the equivalent linear conductance and equivalent distributed capacitance of the current cycle, as well as the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, obtain the transient nonlinear residual current at each moment in the current cycle.
[0048] Because long-distance underground cables have significant inherent distributed capacitance to the ground, normal capacitive charging and discharging will generate linear capacitive charging and discharging currents with high amplitudes. At the same time, there are also linear ohmic leakage currents flowing through trees or polluted insulators in the line. These current components that conform to the linear physical laws will cover the weak nonlinear physical response caused by high-voltage breakdown, causing conventional methods based on amplitude detection to fail. Therefore, it is necessary to separate the above two linear physical components from the total measured current, so as to extract the transient nonlinear residual current mainly caused by high-voltage arc phase transition or nonlinear load, in order to avoid the interference of linear capacitive charging and discharging currents and linear ohmic leakage currents.
[0049] Since periodic sinusoidal voltage signals have the objective law that their own waveform and their own time derivative waveform cancel each other out to zero after undergoing a complete standard power frequency cycle, after obtaining the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, the equivalent linear conductance and equivalent distributed capacitance of the current cycle can be obtained based on the changing characteristics of the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle.
[0050] The method for obtaining the equivalent linear conductance and equivalent distributed capacitance of the current period based on the changing characteristics of the transient zero-sequence voltage and transient zero-sequence current at each moment within the current period is as follows:
[0051] (1) Obtain the equivalent linear conductance of the current period based on the transient zero-sequence voltage and transient zero-sequence current at each moment in the current period.
[0052] By integrating and projecting the transient zero-sequence current at each moment onto the transient zero-sequence voltage at each moment over the entire cycle, the integral of the projection of the transient zero-sequence current onto the transient zero-sequence voltage direction filters out the capacitive component and retains only the purely resistive equivalent linear conductance component, thus obtaining the equivalent linear conductance for the current cycle. The formula for calculating the equivalent linear conductance for the current cycle is:
[0053] ;
[0054] in, The equivalent linear conductance for the current period is expressed in Siemens units. For the current period, the first The transient zero-sequence current at time t is expressed in amperes. For the current period, the first The transient zero-sequence voltage at time t is in volts; t is the current time in seconds; and T is the preset time interval in seconds.
[0055] (2) Obtain the derivative of the transient zero-sequence voltage at each moment in the current cycle.
[0056] Specifically, for any moment within the current period, the difference between the transient zero-sequence voltage at that moment and the transient zero-sequence voltage at the previous moment is obtained to obtain the transient zero-sequence voltage difference. The interval between the moment and the previous moment is obtained, and the ratio of the transient zero-sequence voltage difference to the interval is obtained to obtain the derivative of the transient zero-sequence voltage at that moment. Similarly, the derivative of the transient zero-sequence voltage at each moment within the current period is obtained.
[0057] (3) Obtain the equivalent distributed capacitance of the current period based on the derivatives of the transient zero-sequence current and the transient zero-sequence voltage at each moment in the current period.
[0058] By integrating the transient zero-sequence current at each moment onto the derivative of the transient zero-sequence voltage at each moment over the entire cycle, the integral of the projection of the transient zero-sequence current onto the time derivative of the transient zero-sequence voltage filters out the resistive component and retains only the purely capacitive equivalent distributed capacitance component, thus obtaining the equivalent linear conductance for the current cycle. The formula for calculating the equivalent distributed capacitance for the current cycle is:
[0059] ;
[0060] in, Let be the equivalent distributed capacitance for the current period, with dimensions in farads; For the current period, the first The transient zero-sequence current at time t is expressed in amperes. For the current period, the first The transient zero-sequence voltage at time t, with dimensions in volts; For the current period, the first The derivative of the transient zero-sequence voltage at time t is in volts per second; t is the current time in seconds; T is the preset time period in seconds.
[0061] Furthermore, based on Kirchhoff's current node law, and according to the equivalent linear conductance and equivalent distributed capacitance of the current cycle, as well as the transient zero-sequence voltage and transient zero-sequence current at each moment within the current cycle, the transient nonlinear residual current at each moment within the current cycle is obtained. This effectively eliminates the interference from high-amplitude linear capacitive charging and discharging currents and linear ohmic leakage currents, isolating the nonlinear physical response mainly caused by high-voltage arc phase transitions or nonlinear loads. Kirchhoff's current node law is existing technology and will not be elaborated upon here.
[0062] The method for obtaining the transient nonlinear residual current at each moment in the current period, based on the equivalent linear conductance and equivalent distributed capacitance of the current period, as well as the transient zero-sequence voltage and transient zero-sequence current at each moment in the current period, is as follows:
[0063] For any moment within the current cycle, the product of the equivalent linear conductance of the current cycle and the transient zero-sequence voltage at any moment is obtained to obtain the purely resistive leakage current at any moment.
[0064] The product of the equivalent distributed capacitance of the current period and the derivative of the transient zero-sequence voltage at any given moment is obtained to get the purely capacitive charging and discharging current at any given moment.
[0065] The difference between the transient zero-sequence current at any given time and the purely resistive leakage current at any given time is obtained to obtain a first difference. The difference between the first difference and the purely capacitive charging and discharging current at any given time is obtained to obtain the transient nonlinear residual current at any given time.
[0066] In one implementation, the number of times within the current period... For example, at time 1, the first The formula for calculating the transient nonlinear residual current at time t is:
[0067] ;
[0068] in, For the current period, the first The transient nonlinear residual current at time t is expressed in amperes. For the current period, the first The transient zero-sequence current at time t is expressed in amperes. The equivalent linear conductance for the current period is expressed in Siemens units. Let τ be the transient zero-sequence voltage at time τ within the current period, with the dimension of volts; Let be the equivalent distributed capacitance for the current period, with dimensions in farads; Let be the derivative of the transient zero-sequence voltage at time τ within the previous period, with dimensions in volts per second.
[0069] It should be noted that the formula for calculating transient nonlinear residual current is obtained by removing the normal physical waveform that follows the linear Ohm's law and the linear charging and discharging law of capacitors. The larger the value of transient nonlinear residual current, the more significant the nonlinear physical response in the distribution network. It directly reflects the degree to which the distribution network system deviates from the pure linear steady state, and provides a reliable calculation basis for subsequent analysis of the nonlinear characteristics of high-impedance grounding arcs or power electronic equipment.
[0070] Similarly, the transient nonlinear residual current at each moment within the current period is obtained.
[0071] Step S103: Obtain the transient nonlinear residual current at each moment in the previous cycle of the current cycle. Based on the correlation characteristics between the transient nonlinear residual current at each moment in the current cycle and the transient nonlinear residual current at each moment in the previous cycle of the current cycle, obtain the waveform cross-correlation coefficient between the current cycle and the previous cycle of the current cycle. Use the waveform cross-correlation coefficient to obtain the transient thermodynamic chaos index of the current cycle.
[0072] In modern, highly electronic power distribution networks, the massive number of electric vehicle charging piles and AC / DC rectifier equipment, as typical nonlinear loads, also inject nonlinear current components into the power grid during normal operation. The magnitude of the transient nonlinear residual current alone cannot distinguish between the compliant charging pile steady-state harmonics and the high-impedance grounding arc that poses a safety risk.
[0073] Based on the physical behavior differences between the chaotic characteristics of plasma thermodynamics and the solid-state phase-locked characteristics of silicon-based semiconductors, the current waveforms of power electronic devices driven by solid-state semiconductor forced phase-locking are highly repeatable in adjacent standard power frequency cycles. However, the current waveforms of high-impedance grounding arcs, which are affected by external wind and thermal convection interference, will inevitably undergo physical distortion and waveform decorrelation in adjacent standard power frequency cycles.
[0074] Therefore, this embodiment obtains the transient nonlinear residual current at each moment in the previous cycle of the current cycle. Based on the correlation characteristics between the transient nonlinear residual current at each moment in the current cycle and the transient nonlinear residual current at each moment in the previous cycle of the current cycle, the waveform cross-correlation coefficient between the current cycle and the previous cycle of the current cycle is obtained. Then, using the waveform cross-correlation coefficient, the transient thermodynamic chaos index of the current cycle is obtained, mapping the microscopic waveform distortion to the macroscopic thermodynamic chaotic physical characteristics, thereby realizing the identification of high-impedance grounding arcs.
[0075] The formula for calculating the cross-correlation coefficient between the current period and the previous period is as follows:
[0076] ;
[0077] in, The cross-correlation coefficient between the current period and the previous period is dimensionless. For the current period, the first The transient nonlinear residual current at time t is expressed in amperes. The first cycle in the previous cycle of the current cycle The transient nonlinear residual current at time t is measured in amperes; t is the current time in seconds; and T is the preset time period in seconds.
[0078] It should be noted that, The inner product of the transient nonlinear residual currents in two adjacent cycles was calculated. The square integral values of the two cycles were calculated separately and normalized by taking the square root, so that the cross-correlation coefficient of the waveform was mapped to a reasonable range. When the transient nonlinear residual current of the current cycle originates from electronic loads controlled by hardware phase-locked loops, such as electric vehicle charging piles, the waveform has overlapping characteristics in two adjacent cycles, making the cross-correlation coefficient of the waveform approach 1. When the transient nonlinear residual current of the current cycle originates from a high-impedance grounding arc with thermodynamic chaotic physical phenomena, the random phase transition of air gap breakdown discharge causes waveform distortion, resulting in a decrease in the cross-correlation coefficient of the waveform.
[0079] Furthermore, the method for obtaining the transient thermodynamic chaos index of the current period using the cross-correlation coefficient of the waveform is as follows:
[0080] The difference between the constant 1 and the cross-correlation coefficient of the waveform is obtained to obtain the transient thermodynamic chaos index of the current period.
[0081] In one embodiment, the formula for calculating the transient thermodynamic chaos index of the current period is:
[0082] ;
[0083] in, The transient thermodynamic chaos index for the current period is dimensionless. It is the cross-correlation coefficient of the waveform between the current period and the previous period, and is dimensionless.
[0084] It should be noted that, The larger the value, the more likely the transient nonlinear residual current of the current cycle originates from electronic loads controlled by hardware phase-locked loops, such as electric vehicle charging piles. In such cases, the waveforms overlap in two adjacent cycles, causing the cross-correlation coefficient of the waveforms to approach 1, and the transient thermodynamic chaos index of the current cycle to approach 0. The smaller the value, the more obvious the periodic decorrelation characteristic of the waveform, corresponding to the more intense the plasma thermodynamic chaotic evolution, and the larger the transient thermodynamic chaotic index of the current period.
[0085] For example, suppose a normally functioning electric vehicle charging station is connected to the end of a power distribution network. The transient nonlinear residual current generated in the current cycle exhibits regular steady-state harmonic distortion, and the waveforms of the two cycles overlap significantly. The cross-correlation coefficient obtained by substituting these values into the calculation is as high as 0.99, while the corresponding transient thermodynamic chaos index is only 0.01. Conversely, if a high-impedance grounding arc fault occurs caused by a tree branch touching the power distribution line, the arc is affected by the wind blowing and is interrupted. In the previous cycle, the arc burned for 10 milliseconds, while in the current cycle, it only burned for 5 milliseconds. The two discharge waveforms have random truncation and morphological differences. The cross-correlation coefficient obtained by substituting these values into the calculation drops to 0.4, while the corresponding transient thermodynamic chaos index rises to 0.6. The numerical difference in the transient thermodynamic chaos index distinguishes between two types of current sources with different physical properties.
[0086] Thus, the transient thermodynamic chaos index for the current period is obtained.
[0087] Step S104: Based on the transient nonlinear residual current at each moment in the current cycle, obtain the integral characteristic value of the nonlinear residual current in the current cycle, and use the integral characteristic value of the nonlinear residual current in the current cycle and the transient thermodynamic chaos index in the current cycle to perform distribution network anomaly diagnosis.
[0088] Even when the power distribution network is operating normally and without large nonlinear loads, it may still experience irregular numerical fluctuations due to measurement noise or environmental electromagnetic interference. If all minor random fluctuations are diagnosed and judged for chaotic characteristics, it will increase the computing load on the underlying controller and pose a risk of false trip alarms.
[0089] Therefore, this embodiment obtains the integral characteristic value of the nonlinear residual current for the current period based on the transient nonlinear residual current at each moment within the current period. It then uses this integral characteristic value and the transient thermodynamic chaos index for the current period to perform distribution network anomaly diagnosis. By constructing a dual physical boundary judgment mechanism, the chaos judgment branch is activated only when the accumulated nonlinear characteristics reach a level sufficient to trigger a distribution network safety risk. This avoids increasing the computing load on the underlying controller by diagnosing and judging chaotic characteristics for all minor random fluctuations, and also prevents the risk of false trip alarms caused by measurement noise or environmental electromagnetic interference, thus ensuring the safety and reliability of the final isolation action.
[0090] The method for obtaining the integral characteristic value of the nonlinear residual current for the current period based on the transient nonlinear residual current at each moment within the current period is as follows:
[0091] Based on the current period, the square of the transient nonlinear residual current at each moment within the current period is integrated to obtain the integral characteristic value of the nonlinear residual current in the current period.
[0092] In one embodiment, the formula for calculating the integral characteristic value of the nonlinear residual current in the current period is:
[0093] ;
[0094] in, The integral characteristic value of the nonlinear residual current for the current period is expressed in ampere-square seconds. For the current period, the first The transient nonlinear residual current at time t is measured in amperes; t is the current time in seconds; and T is the preset time period in seconds.
[0095] It should be noted that by squaring the transient nonlinear residual current of the current period and integrating it over time, the transient current amplitude fluctuation is transformed into a macroscopic physical index characterizing the intensity. The larger the value, the more significant the dissipation characteristics of nonlinear harmonics or high-impedance grounding arc injection systems in the distribution network.
[0096] Furthermore, the following method is used for distribution network anomaly diagnosis using the integral characteristic value of the nonlinear residual current in the current period and the transient thermodynamic chaotic index in the current period:
[0097] If the integral characteristic value of the nonlinear residual current in the current period is greater than the preset threshold value of the integral characteristic value of the nonlinear residual current, a chaotic judgment branch is triggered (i.e., the relationship between the transient thermodynamic chaos index of the current period and the preset threshold value of the transient thermodynamic chaos index is determined). If the transient thermodynamic chaos index of the current period is greater than the preset threshold value of the transient thermodynamic chaos index, it is confirmed that a high-impedance grounding arc anomaly has occurred in the distribution network. That is, it is determined that the transient nonlinear residual current in the current period originates from a high-impedance grounding arc with safety risks rather than a compliant power electronic load. The main control chip immediately sends a high-level hardware interrupt trip command to the optocoupler isolation drive circuit of the circuit breaker in the target area through the general-purpose input / output pin to cut off the physical feeder switch to achieve hardware isolation of the high-impedance grounding anomaly.
[0098] If the transient thermodynamic chaos index of the current cycle is less than or equal to the preset transient thermodynamic chaos index threshold, it is determined that the current nonlinear residual current does not meet the diagnostic conditions for high impedance grounding arc, and no tripping command is issued and monitoring continues.
[0099] If the integral characteristic value of the nonlinear residual current in the current cycle is less than or equal to the preset threshold value of the integral characteristic value of the nonlinear residual current, the chaotic decision branch will not be triggered, the trip command will not be issued, and monitoring will continue, thus forming a closed-loop control data flow.
[0100] The preset nonlinear residual current integral characteristic value threshold is obtained by collecting baseline data under fault-free conditions of the distribution network and calibrating it based on the extreme values of the historical steady-state background noise distribution characteristics of the line. In this embodiment, its empirical value is specifically set to 5% of the square integral characteristic of the steady-state distributed capacitance reactive inrush current over the entire cycle. The preset transient thermodynamic chaos index threshold is calibrated and obtained based on the maximum waveform decorrelation limit of a typical power electronic load under the worst voltage fluctuation conditions of the power grid. Its empirical value range is limited to 0.15 to 0.25. In this embodiment, the preset transient thermodynamic chaos index threshold is set to 0.2 to ensure that the system has an effective tolerance range and anti-maloperation capability for compliant modern electronic equipment, while maintaining high sensitivity to concealed arcs.
[0101] This completes the anomaly diagnosis of the power distribution network.
[0102] In summary, in this embodiment of the invention, the equivalent linear conductance and equivalent distributed capacitance of the current cycle are first obtained, and then the transient nonlinear residual current at each moment within the current cycle is obtained. This effectively eliminates the interference from high-amplitude linear capacitive charging and discharging currents and linear ohmic leakage currents, isolating the nonlinear physical response mainly caused by high-voltage arc phase transitions or nonlinear loads. This solves the problem that conventional methods based on amplitude detection are covered by capacitive charging and discharging currents. Then, the waveform cross-correlation coefficient between the current cycle and the previous cycle is obtained. Using the waveform cross-correlation coefficient, the transient thermodynamic chaos index of the current cycle is obtained, thus representing the microscopic waveform distortion as macroscopic thermodynamic chaos. The system can accurately identify nonlinear currents caused by chaotic evolution of plasma thermodynamics, thus effectively distinguishing between compliant steady-state harmonics of charging piles and high-impedance grounding arcs with safety risks. Finally, it obtains the integral characteristic value of the nonlinear residual current in the current period. Using the integral characteristic value of the nonlinear residual current in the current period and the transient thermodynamic chaos index in the current period, a dual physical boundary judgment mechanism is constructed. This avoids increasing the computing load on the underlying controller by diagnosing and judging the chaotic characteristics of all small-level random fluctuations, and also prevents the risk of false trip alarms caused by measurement noise or environmental electromagnetic interference. This significantly improves the accuracy, safety and reliability of abnormal operation status diagnosis of the distribution network.
[0103] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.
Claims
1. A method for intelligent diagnosis of abnormal operating conditions in a power distribution network, characterized in that, The intelligent diagnostic method for abnormal operating states of a power distribution network includes: The transient zero-sequence voltage and transient zero-sequence current of the distribution network at each moment are obtained, and the preset time period up to the current moment is recorded as the current period. Based on the changing characteristics of transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, the equivalent linear conductance and equivalent distributed capacitance of the current cycle are obtained. Based on the equivalent linear conductance and equivalent distributed capacitance of the current cycle, as well as the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, the transient nonlinear residual current at each moment in the current cycle is obtained. Obtain the transient nonlinear residual current at each moment in the previous cycle of the current cycle. Based on the correlation characteristics between the transient nonlinear residual current at each moment in the current cycle and the transient nonlinear residual current at each moment in the previous cycle of the current cycle, obtain the waveform cross-correlation coefficient between the current cycle and the previous cycle of the current cycle. Use the waveform cross-correlation coefficient to obtain the transient thermodynamic chaos index of the current cycle. Based on the transient nonlinear residual current at each moment in the current cycle, the integral characteristic value of the nonlinear residual current in the current cycle is obtained. Using the integral characteristic value of the nonlinear residual current in the current cycle and the transient thermodynamic chaos index in the current cycle, the distribution network anomaly diagnosis is performed.
2. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 1, characterized in that, The step of obtaining the equivalent linear conductance and equivalent distributed capacitance for the current period based on the changing characteristics of the transient zero-sequence voltage and transient zero-sequence current at each moment within the current period includes: Based on the transient zero-sequence voltage and transient zero-sequence current at each moment in the current cycle, obtain the equivalent linear conductance of the current cycle; For any moment within the current period, the difference between the transient zero-sequence voltage at any moment and the transient zero-sequence voltage at the previous moment is obtained to obtain the transient zero-sequence voltage difference. The interval time between any moment and the previous moment is obtained, and the ratio of the transient zero-sequence voltage difference to the interval time is obtained to obtain the derivative of the transient zero-sequence voltage at any moment. Obtain the derivative of the transient zero-sequence voltage at each moment in the current cycle. Based on the derivatives of the transient zero-sequence current and the transient zero-sequence voltage at each moment in the current cycle, obtain the equivalent distributed capacitance of the current cycle.
3. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 2, characterized in that, The step of obtaining the equivalent linear conductance for the current period based on the transient zero-sequence voltage and transient zero-sequence current at each moment within the current period includes: The formula for calculating the equivalent linear conductance of the current period is: ； in, The equivalent linear conductance for the current period, For the current period, the first The transient zero-sequence current at time t. For the current period, the first The transient zero-sequence voltage at time t is the current time, and T is the preset time period.
4. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 2, characterized in that, The step of obtaining the equivalent distributed capacitance of the current period based on the derivatives of the transient zero-sequence current and the transient zero-sequence voltage at each moment within the current period includes: The formula for calculating the equivalent distributed capacitance of the current period is: ； in, The equivalent distributed capacitance for the current period. For the current period, the first The transient zero-sequence current at time t. For the current period, the first The transient zero-sequence voltage at time t. For the current period, the first The derivative of the transient zero-sequence voltage at time t, where t is the current time and T is the preset time period.
5. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 2, characterized in that, The step of obtaining the transient nonlinear residual current at each moment in the current period based on the equivalent linear conductance and equivalent distributed capacitance of the current period, and the transient zero-sequence voltage and transient zero-sequence current at each moment in the current period, includes: For any moment within the current cycle, the product of the equivalent linear conductance of the current cycle and the transient zero-sequence voltage at any moment is obtained to obtain the purely resistive leakage current at any moment. The product of the equivalent distributed capacitance of the current period and the derivative of the transient zero-sequence voltage at any given moment is obtained to get the purely capacitive charging and discharging current at any given moment. The difference between the transient zero-sequence current at any given time and the purely resistive leakage current at any given time is obtained to obtain a first difference. The difference between the first difference and the purely capacitive charging and discharging current at any given time is obtained to obtain the transient nonlinear residual current at any given time.
6. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 1, characterized in that, The step of obtaining the waveform cross-correlation coefficient between the current period and the previous period based on the correlation characteristics between the transient nonlinear residual current at each moment in the current period and the transient nonlinear residual current at each moment in the previous period includes: The formula for calculating the cross-correlation coefficient between the current period and the previous period is: ； in, This represents the cross-correlation coefficient between the current period and the previous period. For the current period, the first The transient nonlinear residual current at time t. The first in the previous cycle of the current cycle The transient nonlinear residual current at time t is the current time and T is the preset time period.
7. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 1, characterized in that, The step of obtaining the transient thermodynamic chaos index of the current period using the cross-correlation coefficient of the waveform includes: The difference between the constant 1 and the cross-correlation coefficient of the waveform is obtained to obtain the transient thermodynamic chaos index of the current period.
8. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 1, characterized in that, The step of obtaining the integral characteristic value of the nonlinear residual current for the current period based on the transient nonlinear residual current at each moment within the current period includes: Based on the current period, the square of the transient nonlinear residual current at each moment within the current period is integrated to obtain the integral characteristic value of the nonlinear residual current in the current period.
9. The intelligent diagnostic method for abnormal operating states of a power distribution network according to claim 1, characterized in that, The method of using the integral characteristic value of the nonlinear residual current in the current period and the transient thermodynamic chaos index in the current period for distribution network anomaly diagnosis includes: If the integral characteristic value of the nonlinear residual current in the current period is greater than the preset threshold value of the integral characteristic value of the nonlinear residual current, and the transient thermodynamic chaos index in the current period is greater than the preset threshold value of the transient thermodynamic chaos index, then the distribution network is confirmed to have a high-impedance grounding arc anomaly.