Fault arc acquisition device and arc starting identification model construction method

Abstract

The invention provides a fault arc acquisition device and an arc starting identification model construction method. The device comprises a processor used for controlling a driving module to move a copper rod and / or a carbon rod so as to adjust the distance between the copper rod and the carbon rod to be gradually increased or gradually decreased; the current sampling module is used for collecting current data between the copper rod and the carbon rod in the moving process of the copper rod and / or the carbon rod and transmitting the current data to the processor; the processor is also used for identifying an arcing point in the current data based on a pre-trained arcing identification model, controlling the driving module to stop the movement of the copper rod and / or the carbon rod after the arcing point is identified, and storing fault arc data in an acquisition period starting from the arcing point; and when the acquisition period is finished, judging whether the fault arc is extinguished or not in the acquisition period. According to the method, the fault arc experiment acquisition efficiency can be improved, invalid data can be reduced, and the data analysis difficulty is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of fault arc technology, and in particular to a fault arc acquisition device and a method for constructing an arc starting identification model. Background Art

[0002] With the rapid development of society and industry, energy has gradually become the most precious resource for mankind. With the depletion of fossil energy and the resulting environmental pollution and other problems are also plaguing mankind. Based on this, people pay more and more attention to various renewable energy sources, such as tidal energy, wind energy, solar energy, etc. Among them, solar energy is a clean energy, and it is almost endless in a limited time, so it has extraordinary potential. Photovoltaic power generation has become an important supplementary source of electricity for current life and production due to its environmental protection, high efficiency and renewability. In areas where large-scale and long-distance transmission lines are difficult to build, photovoltaic power generation has become an important means to solve the power supply problems in these areas with its unique advantages. We can expect that distributed power sources will be widely used in the power market in the future, partially replacing existing energy sources. Compared with some conventional power generation methods, photovoltaic power generation is safer and more reliable, and it is pollution-free, easy to obtain, easy to install, short in construction time, and controllable in size. With the problems of increasing energy pressure and pollution caused by energy use brought about by urbanization in my country, photovoltaic power generation will play a greater role. Even when the efficiency of photovoltaic power generation is improved to a high enough level, it is just around the corner for this technology to replace traditional coal power generation.

[0003] However, with the rapid development of the photovoltaic industry, the frequency of fires in photovoltaic systems has also increased, seriously endangering the normal operation of photovoltaic systems. Some research results show that the direct cause of fires in photovoltaic systems is caused by DC fault arcs. DC arcs do not have characteristics such as zero crossing and periodicity, and are characterized by high temperature and short duration. At the same time, the photovoltaic system has a complex structure, which makes it very difficult to detect DC fault arcs in photovoltaic systems. Therefore, the research on DC fault arc detection in photovoltaic systems has important application value.

[0004] The United States first proposed the UL1699B standard, which is mainly used for arc fault circuit interrupters (AFCI). The standard specifies product categories and relevant test specifications for fault arcs. The application and practice of AFCI abroad have shown that the device can effectively improve the efficiency of fault arc prevention and reduce the risk of fire. The UL1699B standard stipulates the recognition time of photovoltaic fault arcs, requiring the arc to be interrupted within 2.5 seconds under the condition of limited energy. The existing photovoltaic experimental platform generally uses copper rods and carbon rods to simulate arc generators. By controlling the spacing between copper rods and carbon rods, it can break through the air to achieve the effect of "arcing". When simulating DC fault arcs in the experiment, after the arc is generated, it is necessary to collect current data in time before the arc is extinguished. At present, most experimental equipment manually adjusts the spacing between copper rods and carbon rods to generate arcs, continuously collects current, and determines arc data through screening after collection, but this method requires a lot of data to be processed and is difficult to analyze. Moreover, since the voltage of the photovoltaic DC circuit is small and the current is large, if you want to generate an arc, the spacing conditions between the copper rod and the carbon rod of the arc generator are particularly strict. Once an arc is generated, the arc burning time is often short, which causes great difficulties for the experiment. Summary of the invention

[0005] The present invention provides a fault arc acquisition device and a method for constructing an arc identification model to solve the problems of low acquisition efficiency and much invalid data in a fault arc experiment of a DC side circuit of a photovoltaic system.

[0006] In a first aspect, the present invention provides a fault arc collection device, comprising: a photovoltaic simulation module, a fault arc generation module, a current sampling module, a processor and a driving module; the photovoltaic simulation module comprises a photovoltaic panel, the fault arc generation module comprises a copper rod and a carbon rod, and the copper rod and the carbon rod are respectively connected to two ends of the photovoltaic panel;

[0007] A processor, used for moving the copper rod and / or the carbon rod by controlling the driving module to adjust the distance between the copper rod and the carbon rod to gradually increase or gradually decrease;

[0008] A current sampling module is used to collect current data between the copper rod and the carbon rod during the movement of the copper rod and / or the carbon rod, and transmit the current data to the processor;

[0009] The processor is also used to identify the arc starting point in the current data based on a pre-trained arc starting identification model. After identifying the arc starting point, the control driving module stops the movement of the copper rod and / or the carbon rod, and stores the fault arc data within a collection cycle starting from the arc starting point; at the end of the collection cycle, it is determined whether the fault arc is extinguished within the collection cycle. If so, the control driving module continues to move the copper rod and / or the carbon rod. If not, the control driving module moves the copper rod and the carbon rod to completely disconnect or completely contact, and then readjusts the distance between the copper rod and the carbon rod to gradually increase or gradually decrease.

[0010] In a possible implementation, the driving module includes a stepping motor and a motor driving unit;

[0011] The output end of the processor is connected to the input end of the motor drive unit;

[0012] The output end of the motor drive unit is connected to the control end of the stepper motor.

[0013] In a possible implementation, the current sampling module includes a sampling resistor, a filter amplifier, and an analog-to-digital converter;

[0014] One end of the sampling resistor is connected to a copper rod or a carbon rod, and the other end is connected to the input end of the filter amplifier;

[0015] The output end of the filter amplifier is connected to the input end of the analog-to-digital converter;

[0016] The output end of the analog-to-digital converter is connected to the processor.

[0017] In a possible implementation, the photovoltaic simulation module further includes an MPPT controller and a battery, the photovoltaic panel is connected to the MPPT controller, and the MPPT controller is connected to the battery.

[0018] In a second aspect, the present invention provides a method for constructing an arc identification model, comprising: obtaining current data in a fault arc generation module collected in a preset fault arc collection device, wherein the current data includes normal current data and arc current data;

[0019] Comparing the normal current data with the arc current data, and determining the characteristics of the arc current data that are different from the normal current data;

[0020] Determine a fault arc evaluation value based on the characteristics; the fault arc evaluation value represents the possibility of a fault arc occurring in the current;

[0021] Machine learning training is performed based on the fault arc evaluation value to generate an arc starting identification model, wherein the arc starting identification model is used to be deployed on a processor in a fault arc acquisition device as claimed in any one of claims 1 to 4 to identify the arc starting point in the current data collected by the current sampling module in the fault arc acquisition device.

[0022] In a possible implementation, the current sampling module includes a sampling resistor, a filter amplifier, and an analog-to-digital converter, and the fault arc evaluation value is expressed as:

[0023]

[0024] In the formula, C represents the fault arc evaluation value, D m represents the chaotic characteristics of the current, Δv represents the change in the voltage of the sampling resistor, P is the current arcing threshold power, P L is the arc starting current value.

[0025] In a possible implementation, the voltage change Δv of the sampling resistor is expressed as:

[0026] Δv=|V(t+Δt)-V(t)|

[0027] V(t)=V0+(V1+V0) / (t / τ)

[0028] Wherein, V(t) is the voltage value at time t, V0 is the initial voltage value of the sampling resistor, V1 is the final voltage value of the sampling resistor, and τ is the convex change time constant, where the convex change time constant τ characterizes the change speed of the convex change voltage in the arc starting circuit.

[0029] In one possible implementation, the chaotic characteristic D of the current m It is expressed as:

[0030]

[0031] Where T is the sampling time, t is the sampling moment, H is the system function, and the system function is solved by the sequence minimum optimization algorithm and the RBF kernel function binary classification algorithm. V0 is the initial voltage value of the sampling resistor during the acquisition period.

[0032] In a possible implementation, the current arcing threshold power P is expressed as:

[0033] P=I*U

[0034] Where I represents the arc starting current of the photovoltaic panel, and U represents the arc starting voltage of the photovoltaic panel.

[0035] In a possible implementation, the arc current data includes multiple types of arc current data.

[0036] The present invention provides a fault arc collection device and a method for constructing an arc identification model, the method comprising a photovoltaic simulation module, a fault arc generation module, a current sampling module, a processor and a driving module; the photovoltaic simulation module comprises a photovoltaic panel, the fault arc generation module comprises a copper rod and a carbon rod, the copper rod and the carbon rod are respectively connected to the two ends of the photovoltaic panel; the processor is used to move the copper rod and / or the carbon rod by controlling the driving module to adjust the distance between the copper rod and the carbon rod to gradually increase or gradually decrease; the current sampling module is used to collect current data between the copper rod and the carbon rod during the movement of the copper rod and / or the carbon rod, and to collect the current data between the copper rod and the carbon rod. The current data is transmitted to the processor; the processor is also used to identify the arc starting point in the current data based on the pre-trained arc starting recognition model. After identifying the arc starting point, the control driving module stops the movement of the copper rod and / or the carbon rod, and stores the fault arc data within a collection cycle starting from the arc starting point; at the end of the collection cycle, it is determined whether the fault arc is extinguished within the collection cycle. If so, the control driving module continues to move the copper rod and / or the carbon rod. If not, the control driving module moves the copper rod and / or the carbon rod to completely disconnect or completely contact, and then readjusts the distance between the copper rod and the carbon rod to gradually increase or gradually decrease. The present invention controls the distance between the copper rod and the carbon rod through the driving module, and samples the current data between the copper rod and the carbon rod in real time. Since the arc starting recognition model can identify the arc starting point in the current data, further, the fault arc data within the collection cycle can be obtained. Therefore, the present invention can improve the working efficiency of the fault arc collection device, reduce invalid data, and reduce the difficulty of data analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 It is a structural schematic diagram of a traditional fault arc collection device provided by an embodiment of the present invention;

[0039] Figure 2 It is a structural schematic diagram of a fault arc collection device provided by an embodiment of the present invention;

[0040] Figure 3 is an implementation flow chart of a processor provided by an embodiment of the present invention;

[0041] Figure 4 It is a flow chart of an implementation method of a method for constructing an arc starting identification model provided by an embodiment of the present invention;

[0042] Figure 5is a normal current data waveform diagram provided by an embodiment of the present invention;

[0043] FIG. 6A to FIG. 6D These are four arc current data waveform diagrams provided by embodiments of the present invention;

[0044] Figure 7 is a simulated output characteristic curve of a photovoltaic panel provided by an embodiment of the present invention;

[0045] Figure 8 is a schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0047] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0048] Figure 1 The structure diagram of the conventional arc fault acquisition device provided by the embodiment of the present invention is shown in FIG. Figure 1 As shown, the conventional fault arc acquisition device includes a photovoltaic panel, a sampling resistor, a fault arc generating device, a micrometer, a filter amplifier, an AD7606 analog-to-digital converter, an STM32F, an MPPT controller and a battery. The fault arc generating device includes a fixed carbon rod and a movable copper rod, and the copper rod and the carbon rod are respectively connected to the two ends of the photovoltaic panel.

[0049] The current between the copper rod and the carbon rod cannot be directly collected, so one end of the sampling resistor is connected to a fixed carbon rod to convert the current data between the copper rod and the carbon rod into the voltage data of the sampling resistor. The traditional fault arc collection device requires manual movement of the micrometer to control the distance between the copper rod and the carbon rod, collects the voltage data of the sampling resistor and sends it to the STM32F microcontroller, which processes the collected voltage data to filter out the arc data.

[0050] The inventors found that in this type of fault arc collection device, since the voltage of the photovoltaic DC side circuit is relatively small and the current is relatively large, the spacing conditions between the copper rod and the carbon rod in the fault arc generation module are particularly strict when an arc occurs. If they are too close, the arc cannot be ignited, and if they are too far, the circuit will be directly disconnected. Moreover, when an arc is generated, the arcing time is often short. At the same time, manually adjusting the distance between the copper rod and the carbon rod through a micrometer cannot accurately control the arcing node, and the current sampling module will collect the current data when no arc is generated, which not only has a low arc sampling efficiency, but also has a large amount of data redundancy.

[0051] Based on this background technology, the present invention provides a fault arc collection device for a DC side circuit of a photovoltaic system. The core purpose of the present invention is to design a controllable fault arc generation device, which uses a machine instead of a manual method to accurately adjust the distance between the copper rod and the carbon rod, collect the current data between the copper rod and the carbon rod in real time, and identify the current data through a processor, determine the arc starting point, and then collect the arc data. The device is specifically as follows: Figure 2 shown.

[0052] Figure 2 The fault arc collection device shown includes: a photovoltaic simulation module 201, a fault arc generation module 202, a current sampling module 203, a processor 204 and a driving module 205.

[0053] The photovoltaic simulation module 201 includes a photovoltaic panel 2011 , and the fault arc generation module 202 includes a copper rod 2021 and a carbon rod 2022 , and the copper rod 2021 and the carbon rod 2022 are respectively connected to two ends of the photovoltaic panel 2011 .

[0054] The processor 204 is used to move the copper rod 2021 and / or the carbon rod 2022 by controlling the driving module 205 to adjust the distance between the copper rod 2021 and the carbon rod 2022 to gradually increase or gradually decrease;

[0055] The current sampling module 203 is used to collect the current data between the copper rod 2021 and the carbon rod 2022 during the movement of the copper rod 2021 and / or the carbon rod 2022, and transmit the current data to the processor 204;

[0056] The processor 204 is also used to identify the arc starting point in the current data based on a pre-trained arc starting identification model. After identifying the arc starting point, the control driving module 205 stops the movement of the copper rod 2021 and / or the carbon rod 2022, and stores the fault arc data within a collection cycle starting from the arc starting point; at the end of the collection cycle, it is determined whether the fault arc is extinguished within the collection cycle. If so, the control driving module 205 continues to move the copper rod 2021 and / or the carbon rod 2022. If not, the control driving module 205 moves the copper rod 2021 and / or the carbon rod 2022 to completely disconnect or completely contact, and then readjusts the distance between the copper rod 2021 and the carbon rod 2022 to gradually increase or gradually decrease.

[0057] In a possible implementation, one acquisition cycle may be set to 2.5 seconds.

[0058] In a possible implementation, the driving module 205 includes a stepping motor 2051 and a motor driving unit 2052;

[0059] The output end of the processor 204 is connected to the input end of the motor drive unit 2052;

[0060] The output end of the motor drive unit 2052 is connected to the control end of the stepper motor 2051;

[0061] The motor driving unit 2052 controls the movement of the stepping motor 2051 through a PWM signal.

[0062] In one possible implementation, the processor 204 is also used to identify the arc starting point in the current data through a window function based on a pre-trained arc starting identification model. After identifying the arc starting point, the drive module 205 is controlled to stop the movement of the copper rod 2021 and / or the carbon rod 2022, and store the fault arc data within a collection cycle starting from the arc starting point.

[0063] In a possible implementation, the photovoltaic simulation module 201 further includes a solar energy controller (Maximum Power Tracking, MPPT) controller 2012 and a battery 2013 , wherein the photovoltaic panel 2011 is connected to the MPPT controller 2012 , and the MPPT controller 2012 is connected to the battery 2013 .

[0064] In a possible implementation, one of the copper rod 2021 and the carbon rod 2022 of the fault arc generating module 202 may be fixed during the movement. For example, the position of the carbon rod 2022 is fixed, and the distance between the two is adjusted by moving the copper rod 2021.

[0065] In a possible implementation, the current sampling module 203 includes a sampling resistor 2031, a filter amplifier 2032 and an analog-to-digital converter 2033, wherein one end of the sampling resistor 2031 is connected to a copper rod or a carbon rod, and the other end is connected to the input end of the filter amplifier 2032, the output end 2032 of the filter amplifier is connected to the input end of the analog-to-digital converter 2033, and the output end of the analog-to-digital converter 2033 is connected to the processor 204.

[0066] The current between the copper rod 2021 and the carbon rod 2022 cannot be directly collected, so one end of the sampling resistor 2031 is connected to one of the copper rod 2021 and the carbon rod 2022, and the current data between the copper rod 2021 and the carbon rod 2022 is converted into voltage data of the sampling resistor 2031. Therefore, in this embodiment, the current data between the copper rod 2021 and the carbon rod 2022 is collected by connecting the sampling resistor 2031.

[0067] Exemplarily, the sampling resistor 2031 may be a manganese-copper sampling resistor.

[0068] Exemplarily, the analog-to-digital converter 2033 may select an AD7606 analog-to-digital acquisition system.

[0069] In a possible implementation, the processor 204 includes a single-chip microcomputer 2041 and an electronic device 2042. Exemplarily, the single-chip microcomputer 2041 may be a STM32F single-chip microcomputer.

[0070] In order to further explain the working process of a fault arc acquisition device, Figure 3 The working process of the processor 204 in the device is described in detail, see Figure 3 .

[0071] The current sampling module 203 acquires the current data between the copper rod 2021 and the carbon rod 2022, sets a window function in the processor 204, and performs segmented recognition on the acquired current data based on the arc starting recognition model. If the arc starting point is recognized, the stepper motor 2051 is stopped from rotating and the current data within the window function is output. If the arc starting point is not recognized, the stepper motor 2051 continues to rotate to control the movement of the copper rod 2021 and / or the carbon rod 2022.

[0072] The device generates current data by controlling the distance between the moving copper rod 2021 and the carbon rod 2022 through the stepper motor 2051, and identifies the current data based on a pre-trained arc identification model. The device can realize continuous screening and collection of fault arc data.

[0073] The following is an embodiment of the method of the present invention, and the pre-trained arc identification model will be described in detail below. Figure 4 .

[0074] Figure 4 This is a flow chart of a method for constructing an arc starting identification model provided by an embodiment of the present invention, which is described in detail as follows:

[0075] In step 401, current data in a fault arc generating module collected by a preset fault arc collecting device is obtained.

[0076] In a possible implementation, the preset fault arc collection device can select Figure 1 The traditional fault arc collection device is shown in the structural schematic diagram.

[0077] In this embodiment, the current data includes normal current data and arc current data.

[0078] The arc is a transient spark phenomenon caused by the electric field being too strong, the gas undergoing electrical collapse and continuously forming plasma, which causes the current to pass through the insulating medium (such as air) under normal conditions. The embodiment of the present invention obtains normal current data and four types of arc current data, see Figure 5 , FIG. 6A to FIG. 6D .

[0079] Figure 5 The normal current data waveform diagram provided by the embodiment of the present invention, the horizontal axis is time, the vertical axis is current value, it can be found that Fig. 6A The waveform of the medium current fluctuates smoothly within a certain range.

[0080] Fig. 6A This is one of the four arc current data waveforms provided by the embodiment of the present invention. The horizontal axis is time and the vertical axis is current value. In the fault arc acquisition device, when the circuit is connected (the copper rod and the carbon rod are in contact), the distance between the copper rod and the carbon rod is gradually increased. After the arc is generated, the arc light continues until the end of the acquisition. In this case, the current has only a sudden change at the beginning of the arc, that is, the collected Fig. 6A The waveform shown.

[0081] Figure 6B This is one of the four arc current data waveforms provided by the embodiment of the present invention. The horizontal axis is time and the vertical axis is current value. In the fault arc acquisition device, when the circuit is connected (the copper rod and the carbon rod are in contact), the distance between the copper rod and the carbon rod is gradually increased. After the arc is generated, the arc ends within the acquisition time. At this time, the distance between the copper rod and the carbon rod becomes larger, the arc is extinguished, and the circuit is broken. In this case, the current will return to zero after a period of time, that is, the collected Figure 6B The waveform shown.

[0082] Figure 6CThis is one of the four arc current data waveforms provided by the embodiment of the present invention. The horizontal axis is time and the vertical axis is current value. In the fault arc acquisition device, when the circuit is disconnected (the copper rod and the carbon rod are in contact), the distance between the copper rod and the carbon rod is gradually reduced, and the current is from zero to arcing. After the arc is generated, the arc light continues until the end of the acquisition, that is, the acquired Figure 6C The waveform shown.

[0083] Fig.6D This is one of the four arc current data waveforms provided in the embodiment of the present invention. The horizontal axis is time and the vertical axis is current value. In the fault arc acquisition device, when the circuit is disconnected (the copper rod and the carbon rod are in contact), the distance between the copper rod and the carbon rod is gradually reduced. The current will first jump from zero to the arc current, and then return to zero after the arc is extinguished. That is, the collected Fig.6D The waveform shown.

[0084] Exemplarily, the acquisition time of the normal current data and the four arc current data may be selected to be 2.5 seconds.

[0085] In step 402, the normal current data and the arc current data are compared to determine the features that distinguish the arc current data from the normal current data.

[0086] By comparing the normal current data and arc current data obtained in step 401, it can be found that when an arc is generated, the current will undergo a large change in a short period of time, so the slope information is highly correlated with the arc.

[0087] In a possible implementation, the characteristics of arc current data that are different from normal current data include: According to this variation law and the output characteristic curve of the photovoltaic panel itself in the fault arc acquisition device, the voltage variation Δv of the sampling resistor in the fault arc acquisition device is obtained as follows:

[0088] Δv=|V(t+Δt)-V(t)|

[0089] V(t)=V0+(V1+V0) / (t / τ)

[0090] Wherein, V(t) is the voltage value at time t, V0 is the initial voltage value of the sampling resistor, V1 is the final voltage value of the sampling resistor, and τ is the convex change time constant, where the convex change time constant τ characterizes the change speed of the convex change voltage in the arc starting circuit.

[0091] Figure 7 This is a simulated output characteristic curve of a photovoltaic panel provided by an embodiment of the present invention. In the figure, A1 is the minimum arc starting point, B1 is the maximum arc starting voltage value, and C1 is the cut-off voltage value. Points 2, 3, and 4 are curves under ordinary controller testing, and points 1, 5, and 6 are characteristic curves under MPPT controller.

[0092] In a possible implementation, the characteristics of arc current data that are different from normal current data include: when the current generates a fault arc, the sequence has a chaotic tendency, and the chaotic characteristics of the current D m It is expressed as:

[0093]

[0094] Where T is the sampling time, t is the sampling moment, H is the system function, and the system function is solved by the sequence minimum optimization algorithm and the RBF kernel function binary classification algorithm. V0 is the initial voltage value of the sampling resistor during the acquisition period.

[0095] In a possible implementation, the feature that distinguishes arc current data from normal current data also includes current arcing threshold power, combined with Figure 7 The arc starting line and the power generation voltage of the photovoltaic panel determine the current arc starting threshold power, and the current arc starting threshold power P is expressed as:

[0096] P=I*U

[0097] Where I represents the arc starting current of the photovoltaic panel, and U represents the arc starting voltage of the photovoltaic panel.

[0098] In step 403, an arc fault assessment value is determined based on the characteristics.

[0099] The arc fault assessment value represents the possibility of an arc fault occurring in the current.

[0100] The three features that distinguish the arc current data from the normal current data in step 403 are integrated to obtain a fault arc evaluation value.

[0101] In a possible implementation, the fault arc evaluation value is expressed as:

[0102]

[0103] Where C represents the fault arc assessment value, D m represents the chaotic characteristics of the current, Δv represents the change in the voltage of the sampling resistor, P is the current arcing threshold power, P L is the arc starting current value.

[0104] In step 404, machine learning training is performed based on the fault arc evaluation value to generate an arc starting identification model. The arc starting identification model is used to be deployed on a processor in a fault arc acquisition device as claimed in any one of claims 1 to 4 to identify an arc starting point in current data collected by a current sampling module in the fault arc acquisition device.

[0105] Based on the normal current data and the arc current data, the fault arc evaluation value in step 403 is obtained, and machine learning training is performed on a number of fault arc evaluation values ​​to generate an arc starting recognition model.

[0106] In a possible implementation, before performing machine learning training on the fault arc evaluation value, data preprocessing, such as data cleaning, may also be included.

[0107] In one possible implementation, a support vector machine (SVM) algorithm with better binary classification effect can be selected for machine learning training.

[0108] In summary, the embodiments of the present invention provide a fault arc collection device and method for the DC side circuit of a photovoltaic system. In terms of hardware, the control mode of the stepper motor 2051 is set to change the distance between the copper rod 2021 and the carbon rod 2022. After the current data is collected, it is promptly judged by the arc starting recognition model, thereby completing the automatic screening of the arc data, reducing invalid data, and reducing the difficulty of data analysis. In terms of software, the distinguishing features of normal current data and arc current data are obtained by comparing the two, and then combined with machine learning training, an arc starting recognition model is generated, which effectively improves the recognition accuracy of the fault arc.

[0109] The following is an explanation through a specific implementation example.

[0110] In photovoltaic DC circuits, due to the low voltage and high current, the generation of arcs has very strict requirements on the spacing between the copper rod and the carbon rod in the arc generating device. If the spacing is too small, the arc will not be generated; if the spacing is too large, the circuit will be directly disconnected. In addition, once the arc is formed, its duration is usually very short. Traditional fault arc collection devices, such as Figure 1 As shown in the figure, it is necessary to manually move the micrometer to control the distance between the copper rod and the carbon rod, collect the voltage data of the sampling resistor and send it to the STM32F microcontroller, which processes the collected voltage data to filter out the arc data. However, it is difficult to accurately control the generation of arcs by manually adjusting the distance between the carbon rods, and when collecting data, the collection system often collects data when no arc is generated, which leads to low collection efficiency and a large amount of unnecessary data redundancy.

[0111] In order to solve the above problems, the present invention provides a fault arc collection device, which uses a single-chip computer 2041 to control a stepper motor 2051 to accurately adjust the distance between the copper rod 2021 and the carbon rod 2022, obtains the voltage data collected by the current sampling module 203 through a feedback branch, and collects arc data after determining the arc starting point. The device can improve experimental efficiency, reduce invalid data, and reduce analysis difficulty.

[0112] To implement the above technical solution, the present invention has made improvements in both the software part and the hardware part.

[0113] In terms of software design, by collecting normal current data and arc current data, the data features with a high correlation with arc generation are studied, and machine learning training is performed. The trained arc identification model is then re-implanted into the microcontroller 2041 to identify the fault arc data in the current data collected in the current sampling module 203 and determine the arcing node.

[0114] In terms of hardware design, a suitable feedback node is selected, the control mode of the driving module 205 is set, and after the current feedback information of the current sampling module 203 is collected, the arc starting recognition model in the single chip microcomputer 2041 is used for identification, and different PWM waveforms are generated according to the identification results to control the operation mode of the stepping motor 2051. Specifically, Figure 2 As shown, Figure 2 This is a structural schematic diagram of a fault arc acquisition device provided by the present invention after hardware improvement. Figure 2 It can be seen that the circuit related to current collection is Figure 1 Compared with the above, the sampling resistor 2031 end is still used as the collection point, and the collection module is connected in parallel to the sampling resistor 2031 in the DC side circuit of the photovoltaic system. After the processor 203 receives the feedback of the current information, a PWM output channel is brought out at the microcontroller 2051 end to control the movement of the stepper motor 2051.

[0115] The specific implementation process is as follows:

[0116] First pass Figure 1 The conventional fault arc acquisition device shown in FIG. obtains normal current data and arc current data. The waveform of the normal current data is as follows: Figure 5 As shown, the waveform of the arc current data is as follows Figures 6A to 6D shown.

[0117] Then, the normal current data and arc current data are compared to determine the three characteristics of arc current data that are different from normal current data, namely, the voltage change Δv of the sampling resistor, the chaotic characteristics D of the current, and the arc current data. m And the current arcing threshold power P, combining the above three characteristics, the present invention proposes a fault arc evaluation value, the fault arc evaluation value is expressed as:

[0118]

[0119] Where C represents the fault arc assessment value, D m represents the chaotic characteristics of the current, Δv represents the change in the voltage of the sampling resistor, P is the current arcing threshold power, P L is the arc starting current value.

[0120] Finally, machine learning training is performed based on the fault arc evaluation value to generate an arc starting recognition model, which is then implanted into the single chip computer 2041 as a basis for identifying the arc starting point.

[0121] After the generated arc identification model is implanted into the single chip microcomputer 2041, the complete process of using the fault arc acquisition device to obtain and identify the arc is as follows:

[0122] First, the processor 204 sets the initial state of the copper rod 2021 and the carbon rod 2022 by controlling the driving module 205. The initial state can be complete contact or complete separation. In this embodiment, the initial state of the copper rod 2021 and the carbon rod 2022 is complete separation. At the same time, in this embodiment, the carbon rod 2022 is set to be fixed, and the distance between the two is changed by moving the copper rod 2021.

[0123] Then the motor drive unit 2052 controls the stepper motor 2051 to move the copper rod 2021 to adjust the distance between the copper rod 2021 and the carbon rod 2022 to gradually decrease. During the movement of the copper rod 2021, the current sampling module 203 collects the current data between the copper rod 2021 and the carbon rod 2022 and transmits the current data to the processor 204.

[0124] The single chip microcomputer 2041 in the processor detects the current information acquired in the current sampling module 203 through a continuously moving window function, wherein the size of the window function is the size of the acquisition period, for example, it can be set to 2.5 seconds.

[0125] When the microcontroller 2041 detects FIG. 6A to FIG. 6D When any waveform in the waveform is detected, the motor driving unit 2052 controls the stepping motor 2051 to stop rotating, and transmits the collected data to the processor 204, thus completing the collection of arc data.

[0126] If the fault arc is extinguished during the collection cycle, the motor drive unit 2052 controls the stepper motor 2051 to continue moving the copper rod 2021 in the same direction to adjust the distance between the copper rod 2021 and the carbon rod 2022 to continue to decrease, and the same current data collection and identification steps are performed again.

[0127] If the fault arc is not extinguished during the collection cycle, the motor drive unit 2052 controls the stepper motor 2051 to move the copper rod 2021 and the carbon rod 2022 to fully contact each other, and then readjusts the distance between the copper rod 2021 and the carbon rod 2022 to gradually increase, and then performs the same current data collection and identification steps, and repeats this operation to achieve automatic arcing.

[0128] The present invention sets a series of operation modes in controlling the movement of the stepper motor 2051 in combination with the arc identification model, and finally realizes that the fault arc collection device can complete the screening and collection of arc current data, and after the arc collection is completed, the copper rod 2021 can realize the origin relocation function in the next stage to complete the automatic screening and continuous collection of fault arcs.

[0129] The embodiments of the present invention have the following advantages and beneficial effects:

[0130] 1. The embodiment of the present invention consists of two parts: a fault arc collection device and an arc starting identification model. In the identification of the arc starting point in the arc starting identification model, the characteristics of the arc current that distinguish it from the normal current are integrated, and a new feature calculation method is proposed. This feature has a higher correlation with the fault current and is put into the machine learning algorithm to effectively improve the recognition accuracy.

[0131] 2. In controlling the movement of the stepper motor 2051, a series of operating modes are set in combination with the arc identification model, and finally the fault arc collection device is enabled to complete the screening and collection of arc current data. In addition, after the arc collection is completed, the copper rod 2021 can realize the origin repositioning function in the next stage to complete the automatic screening and continuous collection of the fault arc.

[0132] In summary, the present invention proposes a controllable automatic fault arc collection device and constructs a method for arc starting identification model, and realizes automatic identification and collection of fault arc data through improvements in both hardware and software.

[0133] Figure 8 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 8 As shown, the electronic device 8 of this embodiment includes: a processor 80, a memory 81, and a computer program 82 stored in the memory 81 and executable on the processor 80. When the processor 80 executes the computer program 82, the steps in the above-mentioned various arc fault acquisition devices and arc identification model construction method embodiments are implemented, for example Figure 4 Alternatively, when the processor 80 executes the computer program 82, the functions of each module in the above-mentioned device embodiments are realized, for example, Figure 2 Functions of modules 201 to 205 are shown.

[0134] The terminal electronic device 8 may be a computing device such as a desktop computer and a notebook. The terminal electronic device 8 may include, but is not limited to, a processor 80 and a memory 81. Those skilled in the art will understand that Figure 8It is only an example of the terminal electronic device 8 and does not constitute a limitation of the terminal electronic device 8. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal electronic device may also include input and output devices, network access devices, buses, etc.

[0135] The processor 80 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0136] The memory 81 may be an internal storage unit of the terminal electronic device 8, such as a hard disk or memory of the terminal electronic device 8. The memory 81 may also be an external storage device of the terminal electronic device 8, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the terminal electronic device 8. Further, the memory 81 may also include both an internal storage unit of the terminal electronic device 8 and an external storage device. The memory 81 is used to store the computer program and other programs and data required by the terminal electronic device. The memory 81 may also be used to temporarily store data that has been output or is to be output.

[0137] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0138] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0139] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0140] In the embodiments provided by the present invention, it should be understood that the disclosed devices / electronic devices and methods can be implemented in other ways. For example, the device / electronic device embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0141] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0142] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0143] If the integrated module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various fault arc acquisition devices and arc identification model construction method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practices in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practices, computer-readable media does not include electrical carrier signals and telecommunication signals.

[0144] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A fault arc collection device, characterized in that: include: Photovoltaic simulation module, fault arc generation module, current sampling module, processor and drive module; The photovoltaic simulation module includes a photovoltaic panel, and the fault arc generation module includes a copper rod and a carbon rod, wherein the copper rod and the carbon rod are respectively connected to two ends of the photovoltaic panel; The processor is used to control the driving module to move the copper rod and / or the carbon rod so as to adjust the distance between the copper rod and the carbon rod to gradually increase or gradually decrease; The current sampling module is used to collect current data between the copper rod and the carbon rod during the movement of the copper rod and / or the carbon rod, and transmit the current data to the processor; The processor is further used to identify the arc starting point in the current data based on a pre-trained arc starting identification model. After identifying the arc starting point, the processor controls the driving module to stop the movement of the copper rod and / or the carbon rod, and stores the fault arc data within a collection cycle starting from the arc starting point; at the end of the collection cycle, it is determined whether the fault arc is extinguished within the collection cycle. If so, the driving module is controlled to continue moving the copper rod and / or the carbon rod. If not, the driving module is controlled to move the copper rod and the carbon rod so that they are completely disconnected or completely contacted, and then the distance between the copper rod and the carbon rod is readjusted to gradually increase or gradually decrease.

2. The arc fault collection device according to claim 1, characterized in that: The driving module includes a stepping motor and a motor driving unit; The output end of the processor is connected to the input end of the motor drive unit; The output end of the motor driving unit is connected to the control end of the stepping motor.

3. The arc fault collection device according to claim 1, characterized in that: The current sampling module includes a sampling resistor, a filter amplifier and an analog-to-digital converter; One end of the sampling resistor is connected to the copper rod or the carbon rod, and the other end is connected to the input end of the filter amplifier; The output end of the filter amplifier is connected to the input end of the analog-to-digital converter; The output end of the analog-to-digital converter is connected to the processor.

4. The arc fault collection device according to claim 1, characterized in that: The photovoltaic simulation module also includes an MPPT controller and a battery; The photovoltaic panel is connected to the MPPT controller; The MPPT controller is connected to the battery.

5. A method for constructing an arc starting identification model, characterized in that: include: Acquire current data in the fault arc generation module collected by a preset fault arc collection device, wherein the current data includes normal current data and arc current data; comparing the normal current data with the arc current data to determine a feature of the arc current data that is different from the normal current data; Determine a fault arc evaluation value based on the feature; the fault arc evaluation value represents the possibility of a fault arc occurring in the current; Machine learning training is performed based on the fault arc evaluation value to generate an arc starting identification model, wherein the arc starting identification model is used to be deployed on a processor in the fault arc acquisition device as described in any one of claims 1 to 4 to identify the arc starting point in the current data collected by the current sampling module in the fault arc acquisition device.

6. The method for constructing an arc starting identification model according to claim 5, characterized in that: The current sampling module includes the sampling resistor, the filter amplifier and the analog-to-digital converter, and the fault arc evaluation value is expressed as: Where, C represents the fault arc assessment value, D m represents the chaotic characteristics of the current, Δv represents the change in the voltage of the sampling resistor, P is the current arcing threshold power, P L is the arc starting current value.

7. The method for constructing an arc starting identification model according to claim 6, characterized in that: The voltage change Δv of the sampling resistor is expressed as: Δv=|V(t+Δt)-V(t)| V(t)=V0+(V1+V0) / (t / τ) Wherein, V(t) is the voltage value at time t, V0 is the initial voltage value of the sampling resistor in the acquisition period, V1 is the final voltage value of the sampling resistor in the acquisition period, and τ is the convex change time constant, wherein the convex change time constant τ characterizes the change speed of the convex change voltage in the arc starting circuit.

8. The method for constructing an arc starting identification model according to claim 6, characterized in that: The chaotic characteristics of the current D m It is expressed as: Wherein, T is the sampling time, t is the sampling moment, H is the system function, the system function is solved by using the sequence minimum optimization algorithm and the RBF kernel function binary classification algorithm, and V0 is the initial voltage value of the sampling resistor during the acquisition period.

9. The method for constructing an arc starting identification model according to claim 5, characterized in that: The current arcing threshold power P is expressed as: P=I*U Wherein, I represents the arc starting current of the photovoltaic panel, and U represents the arc starting voltage of the photovoltaic panel.

10. The method for constructing an arc starting identification model according to claim 5, characterized in that: The arc current data includes a plurality of arc current data.

Images