A low-voltage direct-current arc hidden danger identification method based on information fusion

Abstract

The application discloses a low-voltage direct-current arc hidden danger identification method based on information fusion, first, a mathematical model of an arc is established, and a direct-current microgrid arc fault test simulation platform is built; an arc fault is simulated, fault branch voltage and non-fault branch voltage are collected, time domain characteristics of the voltage are extracted, and FFT analysis is performed on the voltage, so that a fundamental component and a proportion of components in each characteristic frequency band are obtained; characteristics such as voltage offset, voltage change rate, harmonic distortion rate and oscillation frequency are analyzed, an arc fault voltage characteristic matrix is obtained, and arc characteristics under different fault positions are researched; whether an arc fault occurs in the system is judged according to the direct-current voltage characteristics, and fault positioning is performed according to a fault voltage harmonic distortion rate. The arc fault identification method analyzes arc fault types from fault branch voltage and non-fault branch voltage characteristics, and precise positioning is realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of research on arc fault identification of power electronic power systems, and particularly relates to a low-voltage DC arc hidden danger identification method based on information fusion. BACKGROUND

[0002] Arc fault is a common hidden danger problem of DC microgrid system and has attracted extensive attention. Arc fault of DC bus is often caused by device aging, wire damage and other factors. However, as an important part of the connection between power output and power utilization system, the fault of DC bus will lead to equipment damage and even system off-grid, so it is of great significance to maintain the stability of low-voltage DC system to propose a low-voltage DC arc hidden danger identification method based on information fusion.

[0003] At present, the research on arc fault is divided into series fault and parallel fault according to the fault position. When parallel arc fault occurs, the system current increases, and the DC overcurrent protection can accurately detect the fault. When series arc fault occurs in the system, the arc voltage increases and the current decreases, so it is necessary to use information fusion technology to collect the waveforms of voltage, current and power of fault branch and non-fault branch changing with time, extract the fault voltage, current and power time domain characteristic criteria to distinguish normal working state and fault arc state, realize accurate identification of fault arc, and reduce or even eliminate the harm of fault arc, thereby providing a strong guarantee for the safe and stable operation of power system. SUMMARY

[0004] The purpose of the present application is to provide a low-voltage DC arc hidden danger identification method based on information fusion, which analyzes the characteristics of voltage amplitude and harmonic distortion rate, obtains the arc fault voltage characteristic matrix, studies the arc characteristics under different fault positions, analyzes the arc fault type from the fault branch voltage and non-fault branch voltage characteristics, and realizes accurate positioning.

[0005] The present application provides a low-voltage DC arc hidden danger identification method based on information fusion, which specifically comprises the following steps:

[0006] Step one, based on the arc mathematical model, a DC microgrid fault arc test simulation platform is built;

[0007] Step two, simulate the fault arc on the DC microgrid fault arc test simulation platform, collect the waveforms of voltage, current and power of fault branch and non-fault branch changing with time, analyze the time domain characteristics of voltage, current and power of fault branch and non-fault branch before fault, extract the voltage time domain characteristics after fault and perform FFT analysis on them, and obtain the fundamental component and the proportion of each characteristic component;

[0008] Step three, analyze the fault voltage amplitude and fault voltage harmonic distortion rate, obtain the DC voltage characteristic matrix, study the arc characteristics under different fault arc positions based on the DC voltage characteristic matrix, the DC voltage characteristic matrix includes the voltage offset, voltage change rate, fault voltage harmonic distortion rate, voltage oscillation frequency, fundamental component amplitude of the fault branch and non-fault branch;

[0009] Step four, according to the DC voltage characteristic matrix, judge whether the system has a fault arc, and according to the fault voltage harmonic distortion rate, the DC voltage characteristic matrix is composed of the voltage offset, voltage change rate and oscillation frequency before and after the fault arc of each measurement point.

[0010] The present application has the following beneficial effects:

[0011] (1) The present application obtains the fundamental component and the proportion of each characteristic frequency band component by collecting the fault branch voltage and the non-fault branch voltage; analyzes the voltage amplitude and harmonic distortion rate and other characteristics to obtain the arc fault voltage characteristic matrix for identifying arc fault hazards.

[0012] (2) The present application discloses a low-voltage DC arc hazard identification method based on information fusion, analyzes the voltage, current and power time domain characteristics before the fault, measures the arc fault time domain characteristics by the load power change, voltage change rate and voltage offset, and realizes accurate positioning of the arc fault. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 It is a topology diagram of a DC microgrid multi-machine parallel system;

[0014] Figure 2 It is a DC voltage simulation waveform under the condition that a 30V series fault arc occurs at point S1;

[0015] Figure 3 It is a DC voltage simulation waveform under the condition that a 30V series fault arc occurs at point S2;

[0016] Figure 4 It is a DC voltage waveform spectrum diagram under the condition that a 30V series fault arc occurs at point S2;

[0017] Figure 5 It is a DC voltage simulation waveform under the condition that a 20V series fault arc occurs at point S2;

[0018] Figure 6 It is a DC voltage simulation waveform under the condition that a 20V series fault arc occurs at point S2 when the L line parameter is changed;

[0019] Figure 7 It is a DC voltage simulation waveform under the condition that a 20V series fault arc occurs at point S3;

[0020] Figure 8 DC voltage simulation waveform under 30V series fault arc for point S4. DETAILED DESCRIPTION

[0021] The embodiments of the present application will be described in detail below with reference to the drawings. The embodiments described are only a part of the embodiments of the present application, and all other embodiments obtained by those skilled in the art without creative labor on the basis of the embodiments of the present application are within the scope of protection of the present application.

[0022] The present application aims to provide a low-voltage DC arc hidden danger identification method based on information fusion to solve the problem that the circuit breaker cannot act due to the decrease of current caused by series arc fault. Figure 1 The DC microgrid multi-machine parallel system topology provided by the embodiment of the present application is shown, and the figure includes a plurality of series arc fault points. The research on arc fault identification at home and abroad mainly analyzes the current and voltage characteristics, and it is difficult to identify and locate the series arc fault based on a single characteristic. The present application acquires the fundamental wave component and the proportion of each characteristic component by collecting the voltage of the fault branch and the voltage of the non-fault branch; analyzes the voltage amplitude and harmonic distortion rate and other characteristics to obtain the arc fault voltage characteristic matrix for identifying the fault arc hidden danger.

[0023] A low-voltage DC arc hidden danger identification method based on information fusion, specifically comprising the following steps:

[0024] Step one, based on the arc mathematical model, a DC microgrid fault arc test simulation platform is built;

[0025] Step two, simulate the fault arc based on the DC microgrid fault arc test simulation platform built in step one, collect the waveforms of the voltage, current and power of the fault branch and the non-fault branch changing with time, analyze the time domain characteristics of the voltage, current and power of the fault branch and the non-fault branch before the fault, extract the time domain characteristics of the voltage after the fault and perform fast Fourier transform (FFT) analysis to obtain the fundamental wave component and the proportion of each characteristic component; the characteristic component refers to the component under the oscillation frequency;

[0026] Step three, simulate the fault arc at different positions according to step two, analyze the voltage amplitude and fault voltage harmonic distortion rate and other characteristics after the fault to obtain the DC voltage characteristic matrix, and study the arc characteristics under different fault arc positions based on the DC voltage characteristic matrix;

[0027] Step four, determine whether the system has a fault arc according to the DC voltage characteristic matrix obtained in step three, and locate the fault according to the fault voltage harmonic distortion rate.

[0028] In step one, the arc mathematical model is:

[0029] (1)

[0030] where g represents the conductance per unit length of the arc, τ is the arc time constant, u represents the arc voltage, U c is the voltage gradient of the arc.

[0031] A DC microgrid fault arc test simulation platform based on the cassie arc model is established, which includes an inverter, a bus box, a DC / DC converter, wherein the fault arc is located at the front end of the inverter, the front end of the bus box and the front end of the DC / DC converter.

[0032] In step two, the voltage time domain characteristics after the fault are extracted and fast Fourier transform (FFT) analysis is performed to obtain the fundamental component and the proportion of each characteristic component, including:

[0033] The fault voltage time domain characteristics are measured by the output power change, voltage change rate and voltage offset after the fault. The voltage offset represents the difference between the voltage at any time after the fault and the steady-state voltage value before the fault; the voltage change rate represents the ratio of the voltage change amount at any time to the time, that is, the slope of the voltage change curve at any time.

[0034] Output power change:

[0035] (2)

[0036] where, represents the power change, represents the output voltage before the fault, represents the output current before the fault, represents the output voltage after the fault, represents the output current after the fault.

[0037] For fault arcs at different positions, the dynamic characteristics of the system voltage and current before and after the fault are also different, therefore, the load power change can also be identified as:

[0038] (3)

[0039] where, represents the output voltage offset, δ represents the output current offset.

[0040] Voltage change rate:

[0041] (4)

[0042] The voltage offset and the voltage change rate , if the maximum voltage offset is greater than its threshold, Or the rate of voltage change is greater than its threshold. If so, it can be determined that a system malfunction has occurred. It is the voltage offset threshold. It is the voltage change rate threshold.

[0043] Extract the time-domain characteristics of the voltage before the fault and perform FFT analysis on it to obtain its fundamental component and the proportion of each characteristic frequency band component, and determine the magnitude of the fault arc voltage.

[0044] In step three, characteristics such as voltage amplitude and fault voltage harmonic distortion rate are analyzed to obtain the DC voltage characteristic matrix. Based on the DC voltage characteristic matrix, the fault voltage harmonic distortion rate is expressed as follows:

[0045] (5)

[0046] In the formula, Indicates DC voltage The harmonic distortion rate, DC voltage The First harmonic components, DC voltage The fundamental component, where N is the DC voltage. The order of.

[0047] The DC voltage characteristic matrix includes the voltage offset, voltage change rate, voltage harmonic distortion rate, voltage oscillation frequency, and fundamental component amplitude of the faulty and non-faulty branches.

[0048] In step four, the system is assessed for fault arcing based on DC voltage characteristics, and fault location is determined based on the fault voltage harmonic distortion rate. The specific details are as follows:

[0049] a. Obtain the DC voltage characteristic matrix;

[0050] b. Identify fault arcs based on the DC voltage characteristic matrix;

[0051] DC voltage characteristics refer to the effective value of DC voltage V dcRMS1,2,3 :

[0052] (6)

[0053] In the formula, This indicates the output voltage across the capacitor. Indicates one oscillation period. Represents the steady-state DC voltage component. This represents the DC transient voltage component.

[0054] The fault arc recognition according to the DC voltage feature matrix comprises:

[0055] The energy accumulation of the DC transient component is calculated, and the DC voltage harmonic distortion rate is combined for system oscillation recognition:

[0056] (7)

[0057] In the formula, t0 represents the starting time of the arc occurrence, represents the energy accumulation of the DC transient component.

[0058] c. According to the fault arc recognition result, the voltage offset, the voltage change rate, the fault voltage harmonic distortion rate, and the oscillation frequency of the fault branch and the non-fault branch voltage after the fault are fused to locate the fault arc;

[0059] d. The fault arc locating result is used to verify the low-voltage DC arc fault recognition method based on information fusion. The verification process comprises the following steps: a DC microgrid multi-machine parallel system simulation model is built in matlab, different intensity fault arcs at different positions are simulated in the DC microgrid multi-machine parallel system simulation model, the output voltage waveforms of the fault branch and the non-fault branch are collected, the time-frequency analysis method is used for signal processing of the fault voltage, the output voltage feature parameters are extracted, and data analysis is performed.

[0060] Figure 2 The DC voltage simulation waveform under the condition of 30V series fault arc at S1 is given; from Figure 2 It can be seen that the system output DC voltage only fluctuates 6-7V when the fault arc occurs at S1, and it returns to steady state after about 0.1s, and the system has good stability.

[0061] Figure 3 The DC voltage simulation waveform under the condition of 30V series fault arc at S2 is given, wherein L line1 =1.5mH, L line2 =1mH, L line3 =1.5mH, and other parameters are the same; from Figure 3 It can be seen that the system output DC voltage occurs 134Hz super-synchronous oscillation when the fault arc occurs at S2, the oscillation frequencies of the branches are the same and there is a short time delay, V dc2 is delayed by about 2.4ms than V dc1 , V dc3 is delayed by about 4.1ms than V dc1 , and the fault branch port output voltage V dc1 is distorted, and its frequency spectrum is shown in Figure 4 It can be seen from Figure 4 that different L lineThe total harmonic distortion (THD) of the DC output voltage of the non-faulty branches is similar in magnitude, but V dc2 Compared to V dc3 The oscillation deviations are different, L line The larger the value, the greater the DC voltage deviation.

[0062] To investigate the effect of fault arc strength on DC voltage oscillation characteristics, Figure 5 The simulated DC voltage waveform under a 20V series fault arc at point S2 is given. It can be seen from the figure that the voltage oscillation deviation is relatively small. Figure 3 The frequency of the arc decreases but remains almost unchanged. The fault arc occurs at 0.5s and the oscillation occurs about 0.07s later, indicating a certain delay. Therefore, the strength of the arc can be judged by the delay time and voltage offset.

[0063] To investigate the influence of system parameters on DC voltage oscillation characteristics, Figure 6 L is given line1 =L line2 =L line3 Simulated DC voltage waveform under a 20V series fault arc at S2 when C1=C2=C3=1mF = 1mH, compared with Figure 5 and Figure 6 It can be observed that reducing L line1 and L line3 Afterwards, V dc1 and V dc2 The voltage oscillation offset has increased, and the oscillation frequency of the output voltage in each branch has increased, and V dc2 The increase was even more significant. Additionally, V dc1 With V dc2 and V dc3 There is still a certain time lag, but V dc2 and V dc3 Under this operating condition, they oscillate at the same frequency.

[0064] Figure 7 The simulated DC voltage waveform under a 20V series fault arc at point S3 is given, where L line1 =L line2 =L line3 =1mH, where C1=C2=1mF, C3=2mF. and Figure 5 In comparison, the harmonic distortion rate of the output voltage at the faulty branch port is significantly reduced at this time, unlike V. dc1 and V dc2 Increasing capacitor C3 will increase voltage V dc3 The oscillation amplitude decreased significantly.

[0065] Figure 8The simulation waveforms of DC voltage under 30 V series arc fault at S4 are given. It can be seen from the figure that series arc only makes voltage offset about 2.2 V, and it returns to steady state value after about 0.04 s, the system has good dynamic characteristics. By comparing the DC voltage offset, voltage change rate, oscillation frequency and voltage lag characteristics of each branch when arc fault occurs at different positions, the system arc fault recognition and positioning can be realized.

Claims

1. A low-voltage direct-current arc hazard identification method based on information fusion, characterized in that, The method comprises the following steps: Step one, based on the arc mathematical model, a DC microgrid fault arc test simulation platform is built; Step two, based on the DC microgrid fault arc test simulation platform built in step one, simulate the fault arc, collect the voltage, current and power waveforms of the fault branch and the non-fault branch, analyze the voltage, current and power time domain characteristics of the fault branch and the non-fault branch, extract the voltage time domain characteristics after the fault and perform FFT analysis, and obtain the fundamental component and the proportion of each characteristic component; Step three, according to the fault arc at different positions simulated in step two, analyze the fault voltage amplitude and the fault voltage harmonic distortion rate, obtain the DC voltage characteristic matrix, and study the arc characteristics under different fault arc positions based on the DC voltage characteristic matrix, the DC voltage characteristic matrix includes the voltage offset, voltage change rate, fault voltage harmonic distortion rate, voltage oscillation frequency and fundamental component amplitude of the fault branch and the non-fault branch; Step four, according to the DC voltage characteristic matrix obtained in step three, judge whether the system has a fault arc, and perform fault positioning according to the fault voltage harmonic distortion rate. 2.The low-voltage DC arc hazard identification method based on information fusion according to claim 1, characterized in that, In the step one, The arc mathematical model is: (1) where g represents the conductance per unit length of the arc, τ is the time constant of the arc, u represents the arc voltage, U c is the voltage gradient of the arc; A DC microgrid fault arc test simulation platform based on the arc mathematical model is established, and the fault arc is located at the front end of the inverter, the front end of the bus box and the front end of the DC / DC converter.

3. The low-voltage DC arc hazard identification method based on information fusion according to claim 1, characterized in that, In the step two, simulate the arc fault, collect the voltage, current and power waveforms of the fault branch and the non-fault branch, analyze the voltage, current and power time domain characteristics, and measure the fault arc time domain characteristics by the output power change, voltage change rate and voltage offset after the fault; The output power change is: (2) wherein represents the power variation amount, represents the output voltage before the failure, represents the output current before the failure, represents the output voltage after the failure, represents the output current after the failure; For fault arcs at different positions, the output power change is identified as: (3) where δ represents the output voltage offset, δ represents the output current offset.

4. The low-voltage DC arc hazard identification method based on information fusion according to claim 1, characterized in that, In the step three, analyze the voltage amplitude and the fault voltage harmonic distortion rate, obtain the DC voltage characteristic matrix, and study the arc characteristics under different fault positions based on the DC voltage characteristic matrix; The fault voltage harmonic distortion rate is: (4) In the formula, THD_V dc represents the total harmonic distortion rate of the direct current voltage V dc , V dcn is the nth harmonic component of the direct current voltage , V dc0 is the fundamental component of the direct current voltage , and N is the order of the direct current voltage .

5. The low-voltage DC arc hazard identification method based on information fusion according to claim 1, characterized in that, In the step four, according to the DC voltage characteristic matrix, judge whether the system has a fault arc, and perform fault positioning according to the fault voltage harmonic distortion rate, including: a. Obtain the DC voltage characteristics; b. Identify the fault arc according to the DC voltage characteristic matrix; c. Fuse the voltage offset, voltage change rate, fault voltage harmonic distortion rate and oscillation frequency of the voltage of the fault branch and the non-fault branch after the fault, and position the fault arc; d. Verify the proposed low-voltage DC arc hidden danger identification method based on information fusion, the verification process includes building a DC microgrid multi-machine parallel system simulation model in matlab, simulating fault arcs of different strengths at different positions in the DC microgrid multi-machine parallel system simulation model, collecting the output voltage waveforms of the fault branch and the non-fault branch, using time-frequency analysis method to process the fault voltage, extracting the output voltage characteristic parameters, and performing data analysis.

Images