Photovoltaic power station lightning protection device and maintenance method thereof

Abstract

The invention relates to a photovoltaic power station lightning protection device and a maintenance method thereof, and relates to the field of photovoltaic equipment.The photovoltaic power station lightning protection device comprises a metal shell, the metal shell is filled with fly ash filler and provided with a grounding terminal used for being connected with a grounding grid, and a high-voltage bushing is arranged between the metal shell and a power bus. Two shunt diodes are symmetrically arranged on the high-voltage bushing, the shunt diodes are connected with lamp holders, metal halide lamp bulbs are arranged on the lamp holders, and capacitors are arranged between the lamp holders and the metal shell. According to the invention, the working stability of the photovoltaic power station is improved, and the protection effect of the overvoltage protection equipment is improved.

Description

Technical Field

[0001] The present invention relates to the field of photovoltaic equipment, and in particular to a photovoltaic power station lightning protection device and a maintenance method thereof. Background Art

[0002] A photovoltaic power station refers to a power generation facility that uses solar panels to convert solar energy into electrical energy. It is mainly composed of solar panels, inverters, controllers and other auxiliary equipment.

[0003] In the prior art, the inverter of a photovoltaic power station has a weak ability to withstand lightning overvoltage, and generally requires additional overvoltage protection equipment. Overvoltage protection mainly uses three methods: SPD, gas discharge tube and transient voltage suppression diode. The SPD contains a zinc oxide lightning arrester. After the zinc oxide lightning arrester is activated, it cannot form a negative resistance characteristic and has a high residual voltage. Therefore, it cannot protect photovoltaic equipment that mainly consists of power electronic components. The main problem of the gas discharge tube is that it has no arc extinguishing ability. Once the breakdown discharge cannot restore the insulation, it will cause the equipment to trip. The transient voltage suppression diode has the defect of small current carrying capacity, generally only a few kiloamperes of current carrying capacity, which cannot meet the lightning protection current carrying capacity requirements of the busbar of large photovoltaic equipment.

[0004] When a photovoltaic power station encounters a severe lightning strike, using SPD, gas discharge tube or transient voltage suppression diode to protect the photovoltaic inverter can easily lead to inverter damage, which in turn causes the photovoltaic power station to stop working. Summary of the invention

[0005] In order to improve the working stability of a photovoltaic power station and enhance the protective effect of an overvoltage protection device, the present invention provides a photovoltaic power station lightning protection device and a maintenance method thereof.

[0006] In a first aspect, the present invention provides a photovoltaic power station lightning protection device, which adopts the following technical solution: A photovoltaic power station lightning protection device comprises a metal shell, wherein the interior of the metal shell is filled with fly ash filler, a grounding terminal for connecting to a grounding grid is arranged on the metal shell, a high-voltage bushing is arranged between the metal shell and a power bus, two shunt diodes are symmetrically arranged on the high-voltage bushing, the shunt diodes are both connected to lamp holders, the lamp holders are both provided with metal halide bulbs, and capacitors are arranged between the lamp holders and the metal shell.

[0007] By adopting the above technical scheme, overvoltage protection with extremely large current carrying capacity can be achieved for sensitive equipment such as photovoltaic inverters. It cleverly utilizes the large current carrying capacity of the gas discharge of the metal halide lamp bulb to solve the shortcoming of the small current carrying capacity of the transient voltage diode in the prior art. Since the present invention is a gas breakdown protection, the residual pressure after the overvoltage intrusion wave breaks through the gas is very low, which overcomes the defect of high residual pressure of the zinc oxide lightning arrester in the prior art. The present invention further uses the voltage maintenance effect of the capacitor and combines the arc tube discharge effect of the metal halide lamp bulb with a starting resistor to achieve continuous maintenance of the electrode ion state inside the arc tube under the power frequency voltage. This can further improve the protection action speed to the nanosecond level, providing an overvoltage protection means that completely eliminates the defects of the prior art.

[0008] Optionally, the connection directions of the two shunt diodes and the high-voltage bushing are opposite.

[0009] By adopting the above technical scheme, it further divides the AC voltage into two half-waves through an isolation diode and supplies them to two bulbs respectively. In this way, due to the isolation effect of the diode, when the arc breaks through the gas in the arc tube of the bulb to achieve overvoltage protection, the power frequency follow-through can be interrupted for half a cycle (10 milliseconds) due to the isolation effect of the diode. This gives the gas in the arc tube of the metal halide lamp bulb enough time to deionize and extinguish the arc, so that the power frequency follow-through is cut off, which overcomes the deficiency of the prior art gas discharge tube that cannot cut off the power frequency follow-through and cause tripping.

[0010] In a second aspect, the present application provides a maintenance method for a photovoltaic power station lightning protection device, which adopts the following technical solution: A maintenance method for a photovoltaic power station lightning protection device is applied to any of the photovoltaic power station lightning protection devices mentioned above, comprising: Get the bus voltage of the power bus; When the bus voltage falls into a preset lightning strike interval, the bulb brightness of the metal halide lamp is obtained; Matching the overvoltage brightness from a preset overvoltage brightness matching model according to the bus voltage; Calculate the difference between the bulb brightness and the overvoltage brightness and define it as the brightness difference; When the brightness difference exceeds a preset brightness error range, the device number of the lightning protection device is obtained; Matching damage warning parameters from a preset damage warning parameter model according to the device number; According to the damage warning parameters, a preset warning device is controlled to issue a damage warning.

[0011] By adopting the above technical solution, the voltage of the power bus is monitored to determine whether the photovoltaic power station is struck by lightning, and the brightness of the metal halide lamp is detected when the photovoltaic power station is struck by lightning, so as to determine whether the metal halide lamp is working normally. Then, when the metal halide lamp is not working normally, it is discovered in time and the staff is notified through the warning device, thereby improving the stability of the use of the lightning protection device of the photovoltaic power station.

[0012] Optionally, a method for obtaining the brightness of a light bulb is further included, and the method for obtaining the brightness of a light bulb includes: When the bus voltage falls into a preset lightning strike interval, controlling a preset photographing device to obtain a reference image of the metal halide lamp; matching the shooting position from a preset shooting position recognition model according to the reference image; Matching a shooting itinerary from a preset shooting itinerary matching model according to the shooting position; Controlling a shooting device preset on the metal housing to obtain a brightness image according to the shooting stroke; The bulb brightness is matched from a preset bulb brightness matching model according to the brightness image.

[0013] By adopting the above technical solution, the actual position of the shooting device is first determined by the reference image, and a shooting itinerary is planned to go to a predetermined position to shoot the metal halide lamp bulb according to the actual position of the shooting device. Finally, the brightness of the metal halide lamp is identified from the image of the metal halide lamp bulb.

[0014] Optionally, the method for obtaining the brightness of the light bulb further includes: Matching the ambient brightness from a preset ambient brightness matching model according to the reference image; When the ambient brightness falls into the preset interference interval, the illumination direction is matched from the preset ambient light illumination direction recognition model according to the reference image; Matching a backlight position interval from a preset backlight position interval matching model according to the illumination direction; The shooting stroke is matched from a preset backlight stroke matching model according to the backlight position interval and the shooting position.

[0015] By adopting the above technical solution, external light can easily affect the detection result of the brightness of the metal halide lamp. When the intensity of external light is high, the shooting direction of the shooting device is adjusted according to the irradiation direction of the external light, thereby reducing the situation where external light directly enters the lens of the shooting device.

[0016] Optionally, the method for obtaining the brightness of the light bulb further includes: Get the light bulb image; Determine whether there are multiple lighted light bulbs in the light bulb image from a preset light bulb identification model according to the light bulb image; When there are multiple lighted light bulbs in the light bulb image, the light bulb positions are matched from a preset light bulb position recognition model according to the light bulb image; Matching the light bulb position from a preset light bulb position recognition model according to the brightness image; Match the light direction from a preset light direction recognition model according to the light bulb position and the light on position; Matching the backlight position interval from a preset backlight position interval matching model according to the light direction; The shooting stroke is matched from a preset backlight stroke matching model according to the backlight position interval and the shooting position.

[0017] By adopting the above technical scheme, the brightness of other metal halide lamps can easily affect the brightness detection result of the metal halide lamp being detected. The shooting direction of the shooting device is adjusted according to the relative positions of the metal halide lamp being detected and the lit metal halide lamp, thereby reducing the situation where the light of other metal halide lamps directly enters the lens of the shooting device.

[0018] Optionally, the method for acquiring the light bulb image includes: Matching a shooting angle from a preset shooting angle matching model according to the reference image; determining whether the light bulbs in the reference image overlap according to the shooting angle and the shooting position from a preset light bulb overlap determination model; Judging whether a light bulb in the reference image is missing from a preset light bulb missing judgment model according to a shooting angle and a shooting position; When the light bulbs in the reference image overlap and / or are missing, an image acquisition itinerary is matched from a preset image acquisition itinerary matching model according to the shooting angle and the shooting position; According to the image acquisition stroke, a preset shooting device is controlled to acquire the image of the light bulb.

[0019] By adopting the above technical solution, it is determined whether the reference image can fully display the working condition of the metal halide lamp on the lightning protection device of the photovoltaic power station, and when the metal halide lamp information on the reference image is missing, the position and shooting angle of the shooting device are adjusted to obtain a new image.

[0020] Optionally, a damage analysis method is also included, and the damage analysis method includes: When the brightness difference exceeds the preset brightness error range, the detection stroke is matched from the preset detection stroke matching model according to the position of the bulb; Controlling a preset shooting device to acquire a detection image set of the metal halide lamp in a surrounding manner according to the detection stroke; Determine whether the metal halide lamp is damaged based on the detection image set and the preset lamp appearance detection model; When the metal halide lamp bulb is damaged, the bulb damage parameters are matched from the preset bulb damage parameter matching model according to the device number; According to the bulb damage parameters, a preset warning device is controlled to issue a bulb damage warning.

[0021] By adopting the above technical solution, when the metal halide lamp does not work normally, images of the metal halide lamp bulb at various angles are obtained through the shooting device, and damage features are identified from the images. When the damage features are identified, it is determined that the bulb is damaged, and the staff is promptly reminded that the bulb is damaged and needs to be replaced with a new one.

[0022] Optionally, the damage analysis method further includes: When the metal halide bulb is not damaged, the bulb model is matched from the preset metal halide bulb database according to the detection image set; According to the bulb model, the interference interval is matched from the preset interference interval matching model, and the pressure value at the bottom of the lamp holder is obtained; When the pressure value is lower than the conflict interval, the bulb loosening parameter is matched from the preset bulb loosening parameter matching model according to the device number; Controlling a preset warning device to issue a bulb loosening warning according to the bulb loosening parameter; When the pressure value exceeds the conflict interval, the bulb over-tightening parameter is matched from the preset bulb over-tightening parameter matching model according to the device number; According to the bulb over-tightening parameter, a preset warning device is controlled to issue a bulb over-tightening warning.

[0023] By adopting the above technical scheme, the installation condition of the metal halide lamp on the lamp holder is judged according to the resistance pressure value between the metal halide lamp bulb and the lamp holder, so as to remind the staff to make timely adjustments when the metal halide lamp is not installed properly, resulting in poor contact between the metal halide lamp and the lamp holder, and then causing the metal halide lamp to fail to work normally.

[0024] Optionally, the damage analysis method further includes: When the pressure value is lower than the preset resistance interval, the bulb number is determined according to the bulb position; Retrieve pressure parameters from a preset pressure parameter database according to the device number and the bulb number; Determine whether a pressure jump occurs in the preset pressure jump recognition model according to the pressure parameter; When the pressure value jumps, the lamp holder cracking parameter is matched from the preset lamp holder cracking parameter matching model according to the device number; According to the cracking parameters of the lamp holder, a preset warning device is controlled to issue a lamp holder cracking warning.

[0025] By adopting the above technical solution, when the lamp holder cracks, it is easy to cause the metal halide lamp bulb screwed on the lamp holder to loosen. By retrieving the historical change records of the pressure value, it is identified whether there is a sudden decrease in the pressure value, and it is determined that the sudden decrease in the pressure value is the moment when the lamp holder cracks, so that the staff can be reminded in time when the pressure value suddenly decreases.

[0026] In summary, the present application includes at least one of the following beneficial technical effects: 1. It can realize overvoltage protection with extremely large current carrying capacity for sensitive equipment such as photovoltaic inverters. It cleverly utilizes the large current carrying capacity of the gas discharge of the metal halide lamp bulb to solve the shortcoming of the small current carrying capacity of the transient voltage diode in the prior art. Since the present invention is a gas breakdown protection, the residual pressure after the overvoltage intrusion wave breaks through the gas is very low, which overcomes the defect of high residual pressure of the zinc oxide lightning arrester in the prior art. The present invention further realizes the continuous maintenance of the electrode ion state inside the arc tube under the power frequency voltage through the voltage maintenance function of the capacitor and the arc tube discharge function of the starting resistor of the metal halide lamp bulb, which can further improve the protection action speed to the nanosecond level, providing an overvoltage protection means that completely eliminates the defects of the prior art; 2. Further, the AC voltage is divided into two half-waves through the isolation diode and supplied to the two bulbs respectively. In this way, due to the isolation effect of the diode, when the arc breaks through the gas in the arc tube of the bulb to achieve overvoltage protection, the power frequency continuous current can be interrupted for half a cycle (10 milliseconds) due to the isolation effect of the diode. This gives the gas in the arc tube of the metal halide lamp enough time to deionize and extinguish the arc, so that the power frequency continuous current is cut off, overcoming the deficiency of the gas discharge tube in the prior art that the power frequency continuous current cannot be cut off and causes tripping; 3. Monitor the voltage of the power bus to determine whether the photovoltaic power station is struck by lightning, and detect the brightness of the metal halide lamp when the photovoltaic power station is struck by lightning to determine whether the metal halide lamp is working properly. Then, when the metal halide lamp is not working properly, it will be discovered in time and the staff will be notified through the warning device, thereby improving the stability of the use of the photovoltaic power station lightning protection device. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a structural schematic diagram of a lightning protection device for a photovoltaic power station; Figure 2 It is a flow chart of a maintenance method of a lightning protection device for a photovoltaic power station; Figure 3 This is the process of obtaining the brightness of the bulb Figure 1 ; Figure 4 This is the process of obtaining the brightness of the bulb Figure 2 ; Figure 5 This is the process of obtaining the brightness of the bulb Figure 3 ; Figure 6 is a flow chart of a method for acquiring a light bulb image; Figure 7 The process of damage analysis method Figure 1 ; Figure 8 The process of damage analysis method Figure 2 ; Fig. 9 The process of damage analysis method Figure 3 .

[0028] The names of the parts indicated by the numerical labels in the above drawings are as follows: 1. Metal casing; 2. Grounding terminal; 3. High-voltage bushing; 4. Metal halide lamp bulb; 5. Lamp holder; 6. Shunt diode; 7. Capacitor; 8. Fly ash filler. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0030] The embodiment of the present application discloses a photovoltaic power station lightning protection device and a maintenance method thereof. The present invention is used to provide a protection technology with low residual pressure, automatic arc extinguishing, and strong current carrying capacity. In addition, the present invention cleverly utilizes obsolete metal halide bulbs and scrapped light source devices. When they are aged due to lightning strikes, they can be repaired by replacing new bulbs, and have very low use and maintenance costs.

[0031] Reference Figure 1A photovoltaic power station lightning protection device includes a metal shell 1, a grounding terminal 2, a high-voltage bushing 3, a metal halide lamp bulb 4, a lamp holder 5, a shunt diode 6, a capacitor 7, and a fly ash filler 8. The metal shell 1 is an airtight structure that can withstand an air pressure of 2MPa, and the fly ash filler 8 is filled inside the metal shell 1. The grounding terminal 2 is located below the metal shell 1 and connected to the grounding grid, the high-voltage bushing 3 is located at the top middle position of the metal shell 1, the outside of the high-voltage bushing 3 is connected to the protected power bus L, one side of the inside of the high-voltage bushing 3 is connected to the positive electrode of a shunt diode 6, and the other side of the inside of the high-voltage bushing 3 is connected to the negative electrode of another shunt diode 6. The two shunt diodes 6 have the same model and specification, and their rated instantaneous through current is not less than 10kA. The other two positive and negative electrodes of the two shunt diodes 6 are respectively connected to the screw-type conductive parts of the two lamp holders 5. The metal halide bulbs 4 are respectively installed on the two lamp holders 5. The two metal halide bulbs 4 have auxiliary electrodes and have the same specifications. The capacitors 7 are two metal film impact capacitors of the same specifications with a capacity of 1μF. One end of the two capacitors 7 is connected to the end conductive part of one of the lamp holders 5, and the other end is connected to the metal shell 1.

[0032] Reference Figure 2 Based on the same inventive concept, an embodiment of the present invention provides a maintenance method for a photovoltaic power station lightning protection device, comprising: Step 100: Obtain the bus voltage of the power bus.

[0033] The bus voltage refers to the real-time voltage value of the power bus. The bus voltage can be obtained through a voltmeter. The method for obtaining the bus voltage is selected by the staff based on actual conditions and will not be elaborated here.

[0034] Step 101: When the bus voltage falls into a preset lightning strike interval, the bulb brightness of the metal halide lamp is obtained.

[0035] The lightning strike interval refers to the instantaneous voltage surge interval of the power bus after the photovoltaic power station is struck by lightning. The lightning strike interval is selected by the staff according to the actual situation and will not be described in detail here. The bulb brightness refers to the light intensity value of the metal halide lamp. The method for obtaining the bulb brightness is selected by the staff according to the actual situation and will not be described in detail here.

[0036] Step 102: matching the overvoltage brightness from a preset overvoltage brightness matching model according to the bus voltage.

[0037] The overvoltage brightness matching model refers to a neural network model that has been trained with samples in advance. The overvoltage brightness matching model can predict the intensity of light emitted by the metal halide lamp based on the bus voltage. The overvoltage brightness is the light intensity result obtained by matching the overvoltage brightness matching model. The overvoltage brightness matching model is common knowledge to people in this field and will not be elaborated here.

[0038] Step 103: Calculate the difference between the bulb brightness and the overvoltage brightness, and define it as the brightness difference.

[0039] The brightness difference is the difference between the actual intensity value of the light from the metal halide lamp and the predicted intensity value.

[0040] Step 104: When the brightness difference exceeds a preset brightness error range, the device number of the lightning protection device is obtained.

[0041] The brightness error range refers to the range of light intensity values ​​within which the metal halide lamp is allowed to experience light intensity fluctuations. The brightness error range is selected by the staff based on actual conditions and will not be elaborated here.

[0042] The device number refers to the number set in advance to distinguish the lightning protection device of the photovoltaic power station. The device number corresponds one-to-one to the lightning protection device of the photovoltaic power station. The method for obtaining the device number is selected by the staff according to the actual situation and will not be elaborated here.

[0043] Step 105: Match the damage warning parameter from the preset damage warning parameter model according to the device number.

[0044] The warning device refers to a device used to notify the staff. The warning device is selected by the staff according to the actual situation and will not be described in detail here. The damage warning parameter model refers to a neural network model that has been trained with samples in advance. The damage warning parameter model can match the damage warning parameters that warn the staff of the failure of the photovoltaic power station lightning protection device through the warning device. The damage warning parameter model is common knowledge among people in this field and will not be described in detail here.

[0045] Step 106: Control a preset warning device to issue a damage warning according to the damage warning parameter.

[0046] The working condition of the lightning protection device can be judged by detecting the brightness of the metal halide lamp when the power bus is over-voltage, so that when the brightness of the metal halide lamp is abnormal, it can be judged that the lightning protection device has a fault, and then the staff can be reminded in time through the warning device.

[0047] Reference Figure 3 , the methods for obtaining the brightness of the bulb include: Step 200: When the bus voltage falls into a preset lightning strike interval, a preset shooting device is controlled to obtain a reference image of the metal halide lamp.

[0048] The photographing device refers to a device used to obtain an image of the metal halide lamp bulb to determine the working status of the lightning protection device. The photographing device can be moved by a track preset on the metal shell of the lightning protection device. The photographing device is selected by the staff according to the actual situation and will not be described in detail here. The reference image refers to the image taken by the photographing device at the current position. The method for obtaining the reference image is selected by the staff according to the actual situation and will not be described in detail here.

[0049] Step 201: Matching a shooting position from a preset shooting position recognition model according to a reference image.

[0050] The shooting position recognition model refers to a neural network model that has been trained with samples in advance. The shooting position recognition model can determine the shooting position of the shooting device on the lightning protection device based on the part of the lightning protection device displayed in the reference image. The shooting position recognition model is common knowledge among people in this field and will not be elaborated here.

[0051] Step 202: matching a shooting itinerary from a preset shooting itinerary matching model according to the shooting position.

[0052] The shooting stroke matching model refers to a neural network model that has been trained with samples in advance. The shooting stroke matching model can match the shooting stroke of the shooting device from the current shooting position to the preset fixed position for obtaining the image of the metal halide bulb, wherein the shooting device is always facing the metal halide bulb. The fixed position refers to a position set in advance for obtaining the complete image of the metal halide bulb. A metal halide bulb may be provided with multiple fixed positions, and the shooting stroke matching model is used to select the fixed position closest to the shooting position. When there are multiple metal halide bulbs, the fixed position of each metal halide bulb is selected first, and then the shooting stroke matching model is used to match the shooting stroke that passes through the fixed positions of different metal halide bulbs in sequence. The shooting stroke matching model is common knowledge known to personnel in this field and will not be elaborated here.

[0053] Step 203: controlling a shooting device preset on the metal housing to acquire a brightness image according to the shooting stroke.

[0054] The brightness image refers to the image obtained when the shooting device reaches a fixed position during the shooting journey. The method for obtaining the brightness image is selected by the staff according to the actual situation and will not be elaborated here.

[0055] Step 204: Match the bulb brightness from a preset bulb brightness matching model according to the brightness image.

[0056] The bulb brightness matching model refers to a neural network model that has been trained with samples in advance. The bulb brightness matching model can identify the light intensity value emitted by the metal halide bulb in the brightness image and define the light intensity value as the bulb brightness. The bulb brightness matching model is common knowledge among people in this field and will not be elaborated here.

[0057] Reference Figure 4 , the method for obtaining the brightness of the bulb also includes: Step 205: Match the ambient brightness from a preset ambient brightness matching model according to the reference image.

[0058] The ambient brightness matching model refers to a neural network model that has been trained with samples in advance. The ambient brightness matching model can identify the light intensity of the external light in the environment where the metal halide lamp bulb shown in the reference image is located, and define the light intensity as the ambient brightness, where the external light can be sunlight or other lighting equipment. The ambient brightness matching model is common knowledge among people in this field and will not be elaborated here.

[0059] Step 206: When the ambient brightness falls into a preset interference interval, the illumination direction is matched from a preset ambient light illumination direction recognition model according to the reference image.

[0060] The interference interval refers to the light intensity value interval where external light can easily affect the light intensity test result of the metal halide lamp. The interference interval is selected by the staff according to the actual situation and will not be described here. If the ambient brightness falls into the interference interval, the detected bulb brightness may be inaccurate and it is necessary to reduce the influence of external light and obtain it again.

[0061] The ambient light direction recognition model refers to a neural network model that has been trained with samples in advance. The ambient light direction recognition model can recognize the main radiation direction of the external light, where the main radiation direction refers to the direction of the part of the light with the highest light intensity value in the external light. The ambient light direction recognition model is common knowledge among people in this field and will not be elaborated here.

[0062] Step 207: matching a backlight position interval from a preset backlight position interval matching model according to the illumination direction.

[0063] The backlight position interval matching model refers to a neural network model that has been trained with samples in advance. The backlight position interval matching model can match the backlight position interval in the track of the metal shell that is located in the direction facing the external light, wherein the direction facing the light refers to the light-facing part of the two parts divided by a dividing line centered on the metal halide lamp and oriented in the direction perpendicular to the irradiation direction. The backlight position interval matching model is common knowledge to those skilled in the art and will not be elaborated here.

[0064] Step 208: matching a shooting stroke from a preset backlight stroke matching model according to the backlight position interval and the shooting position.

[0065] The backlight stroke matching model refers to a neural network model that has been trained with samples in advance. The backlight stroke matching model can select a fixed position that falls within the backlight position interval, and then select a fixed position that is closest to the shooting position from the selected fixed positions, thereby matching the backlight position interval from the shooting position to the fixed position. The backlight stroke matching model is common knowledge among people in this field and will not be elaborated here.

[0066] When the external light is strong, the shooting device is controlled to reach the light-facing side of the metal halide lamp so that the shooting direction of the shooting device is close to the irradiation direction of the external light, thereby reducing the situation where the external light directly enters the shooting device and causes errors in the brightness detection results of the metal halide lamp.

[0067] Reference Figure 5 , the method for obtaining the brightness of the bulb also includes: Step 209: Acquire the light bulb image.

[0068] The bulb image refers to a picture used to show the working status of all metal halide bulbs on the lightning protection device. The bulb image is obtained by a shooting device. The method for obtaining the bulb image is selected by the staff according to the actual situation and will not be elaborated here.

[0069] Step 210: Determine whether there are multiple lighted light bulbs in the light bulb image from a preset light-on recognition model according to the light bulb image.

[0070] The lighting recognition model refers to a neural network model that has been trained with samples in advance. The lighting recognition model can identify the number of metal halide bulbs that are glowing in the bulb image. The lighting recognition model is common knowledge among people in this field and will not be described in detail here.

[0071] Step 211: when there are multiple lighted light bulbs in the light bulb image, the light positions are matched from a preset light position recognition model according to the light bulb image.

[0072] The lighting position recognition model refers to a neural network model that has been trained with samples in advance. The lighting position recognition model can match the lighting position of each light bulb that is glowing from the light bulb image. The lighting position recognition model is common knowledge among people in this field and will not be elaborated here.

[0073] Step 212: Match the light bulb position from a preset light bulb position recognition model according to the brightness image.

[0074] The bulb position recognition model refers to a neural network model that has been trained with samples in advance. The bulb position recognition model can recognize the bulb position of the metal halide lamp being detected from the brightness image. The bulb position recognition model is common knowledge to those skilled in the art and will not be elaborated on here.

[0075] Step 213: Match the light direction from a preset light direction recognition model according to the light bulb position and the light-on position.

[0076] The light direction recognition model refers to a neural network model that has been trained with samples in advance. The light direction recognition model can determine the light direction when the light emitted from the lighting position is directed toward the light bulb position based on the relative positions of the light bulb position and the lighting position. The light direction recognition model is common knowledge among people in this field and will not be elaborated here.

[0077] Step 214: Match the backlight position interval from the preset backlight position interval matching model according to the light direction.

[0078] The backlight position interval matching model refers to a neural network model that has been trained with samples in advance. The backlight position interval matching model is located in the light-facing direction in the track of the metal shell. The light-facing direction refers to the light-facing part of the two parts divided by a dividing line centered on the metal halide lamp and oriented perpendicular to the light direction. Different light-facing parts are matched according to different light directions, and the overlapping parts are selected as the backlight position interval. The backlight position interval matching model is common knowledge among people in this field and will not be elaborated here.

[0079] Step 215: Matching a shooting stroke from a preset backlight stroke matching model according to the backlight position interval and the shooting position.

[0080] When there are at least two metal halide bulbs lighting on the lightning protection device, the shooting device is controlled to reach the light-facing side of the metal halide bulbs so that the shooting direction of the shooting device is close to the direction of the light, thereby reducing the situation where the light emitted by other metal halide bulbs directly enters the shooting device and causes errors in the brightness detection results of the metal halide bulb being detected.

[0081] Reference Figure 6 , the method for obtaining the light bulb image includes: Step 300: Match the shooting angle from a preset shooting angle matching model according to the reference image.

[0082] The shooting angle matching model refers to a neural network model that has been trained with samples in advance. The shooting angle matching model can determine the shooting angle of the shooting device from the reference image. The shooting angle matching model is common knowledge among people in this field and will not be elaborated here.

[0083] Step 301: judging whether light bulbs in a reference image overlap according to a shooting angle and a shooting position from a preset light bulb overlap judgment model.

[0084] The bulb overlap judgment model refers to a neural network model that has been trained with samples in advance. The bulb overlap judgment model can judge whether there are overlapping parts of the metal halide bulbs in the reference image. The bulb overlap judgment model is common knowledge among people in this field and will not be described in detail here.

[0085] Step 302: judging whether the light bulb in the reference image is missing from a preset light bulb missing judgment model according to the shooting angle and shooting position.

[0086] The bulb missing judgment model refers to a neural network model that has been trained with samples in advance. The bulb missing judgment model can judge whether the number of metal halide bulbs in the reference image is missing. The bulb missing judgment model is common knowledge among those skilled in the art and will not be elaborated here.

[0087] Step 303: When the light bulbs in the reference image overlap and / or are missing, an image acquisition stroke is matched from a preset image acquisition stroke matching model according to the shooting angle and the shooting position.

[0088] The image acquisition stroke matching model refers to a neural network model that has been trained with samples in advance. The image acquisition stroke matching model can match the image acquisition stroke that enables the shooting device to be adjusted from the shooting position and shooting angle to acquire images of the metal halide lamps on all lightning protection devices. The image acquisition stroke matching model is common knowledge among people in this field and will not be elaborated here.

[0089] Step 304: controlling a preset photographing device to acquire the image of the light bulb according to the image acquisition process.

[0090] When the metal halide bulbs on the reference image overlap or are missing, the working status of all the metal halide bulbs on the lightning protection device cannot be determined through the reference image. In this case, the position and direction of the shooting device are adjusted to obtain a complete image.

[0091] Reference Figure 7 , damage analysis methods include: Step 400: When the brightness difference exceeds a preset brightness error range, a detection stroke is matched from a preset detection stroke matching model according to the position of the light bulb.

[0092] The detection stroke matching model refers to a neural network model that has been trained with samples in advance. The detection stroke matching model can match the detection stroke that controls the camera to surround the metal halide lamp. The detection stroke matching model is common knowledge among those skilled in the art and will not be elaborated on here.

[0093] Step 401: Control a preset shooting device to acquire a detection image set of a metal halide lamp according to a detection stroke.

[0094] The detection image set is a set of images formed when the camera captures images at a fixed period when it circles the metal halide lamp according to the detection stroke.

[0095] Step 402: Determine whether the metal halide lamp is damaged from a preset lamp appearance detection model according to the detection image set.

[0096] The light bulb appearance detection model refers to a neural network model that has been trained with samples in advance. The light bulb appearance detection model can determine whether there are damage features on the metal halide bulb in the detection image set. The damage features include breakage, cracks and black burnt marks. When there are damage features on the metal halide bulb, it is determined that the metal halide bulb is damaged. The light bulb appearance detection model is common knowledge among people in this field and will not be elaborated here.

[0097] Step 403: When the metal halide lamp bulb is damaged, the bulb damage parameters are matched from a preset bulb damage parameter matching model according to the device number.

[0098] The bulb damage parameter matching model refers to a neural network model that has been trained with samples in advance. The bulb damage parameter matching model can match the bulb damage parameters that notify the staff of the damage of the metal halide bulb through the warning device. The bulb damage parameter matching model is common knowledge among people in this field and will not be elaborated here.

[0099] Step 404: Control a preset warning device to issue a bulb damage warning according to the bulb damage parameter.

[0100] When it is determined that the metal halide lamp bulb is damaged, the staff is notified through the warning device, thereby facilitating the staff to perform maintenance in a timely manner.

[0101] Reference Figure 8 , damage analysis methods also include: Step 405: When the metal halide bulb is not damaged, the bulb model is matched from a preset metal halide bulb database according to the detection image set.

[0102] The metal halide bulb database refers to a pre-set database that records the appearance of various models of metal halide bulbs, and the bulb model is the model of the metal halide bulb matched from the metal halide bulb database by detecting the image set.

[0103] Step 406: Match the interference interval from the preset interference interval matching model according to the bulb model, and obtain the pressure value at the bottom of the lamp holder.

[0104] The conflict interval matching model refers to a neural network model that has been trained with samples in advance. The conflict interval matching model can match the interval of the pressure value at the bottom of the lamp holder when the metal halide lamp of the bulb model is stably connected to the lamp holder, and define the interval as the conflict interval. The conflict interval matching model is common knowledge among people in this field and will not be elaborated here.

[0105] The pressure value refers to the pressure on the bottom of the groove on the lamp holder for installing the metal halide lamp bulb. The pressure value can be obtained by a pressure sensor set at the bottom of the lamp holder groove. The method for obtaining the pressure value is selected by the staff according to the actual situation and will not be elaborated here.

[0106] Step 407: When the pressure value is lower than the conflict interval, the bulb loosening parameter is matched from a preset bulb loosening parameter matching model according to the device number.

[0107] The pressure value lower than the conflict interval indicates that the metal halide lamp bulb and the lamp holder do not fully conflict with each other. At this time, the metal halide lamp is prone to poor contact with the lamp holder, resulting in the metal halide lamp not being able to work properly. The bulb loose parameter matching model refers to a neural network model trained with samples in advance. The bulb loose parameter matching model can match the bulb loose parameters that notify the staff of the metal halide bulb looseness through the warning device. The bulb loose parameter matching model is common knowledge among people in this field and will not be described here.

[0108] Step 408: Control a preset warning device to issue a bulb loose warning according to the bulb loose parameter.

[0109] When it is determined that the metal halide bulb is loose, the staff is notified through the warning device, thereby facilitating the staff to perform maintenance in a timely manner.

[0110] Step 409: When the pressure value exceeds the conflict interval, the bulb over-tightening parameter is matched from a preset bulb over-tightening parameter matching model according to the device number.

[0111] The pressure value exceeding the conflict interval indicates that the metal halide lamp is too tightened in the lamp holder. At this time, the metal halide lamp is prone to poor contact with the lamp holder, resulting in the metal halide lamp not being able to work properly. The bulb over-tightening parameter matching model refers to a neural network model trained with samples in advance. The bulb over-tightening parameter matching model can match the bulb over-tightening parameters that notify the staff of the metal halide lamp bulb being over-tightened through the warning device. The bulb over-tightening parameter matching model is common knowledge among people in this field and will not be described here.

[0112] Step 410: Control a preset warning device to issue a bulb over-tightening warning according to the bulb over-tightening parameter.

[0113] When it is determined that the metal halide bulb is over-tightened, the staff is notified through the warning device, thereby facilitating the staff to perform maintenance in a timely manner.

[0114] The cause of the failure of the lightning protection device can be determined in advance based on the pressure value, which makes it convenient for the staff to conduct targeted problem troubleshooting and further facilitates the maintenance of the lightning protection device.

[0115] Reference Fig. 9 , damage analysis methods also include: Step 411: When the pressure value is lower than the preset conflict interval, the bulb number is determined according to the bulb position.

[0116] The bulb number refers to the pre-set number used to distinguish the metal halide bulbs on the lightning protection device. Each bulb number corresponds to a metal halide bulb one by one. The bulb position can be matched with the bulb number corresponding to the bulb position from the pre-set bulb number database. The method of setting the bulb number is selected by the staff according to the actual situation and will not be elaborated here.

[0117] Step 412: Retrieve the pressure parameter from a preset pressure parameter database according to the device number and the bulb number.

[0118] The pressure parameter database refers to a pre-set database that records the historical records of pressure values ​​of different metal halide bulbs on each lightning protection device. The pressure parameter is the historical record of pressure values ​​corresponding to the metal halide bulb with the bulb number on the lightning protection device with the device number retrieved from the pressure parameter database.

[0119] Step 413: judging whether a pressure jump occurs in the preset pressure jump recognition model according to the pressure parameter.

[0120] The pressure jump recognition model refers to a neural network model that has been trained with samples in advance. The pressure jump recognition model can recognize whether there is a sudden jump in the pressure value to a lower level in the historical data of the pressure value. When no jump is detected, the data in the pressure parameter database is marked to reduce repeated detection during the next detection. The pressure jump recognition model is common knowledge among people in this field and will not be elaborated here.

[0121] Step 414: When the pressure value jumps, the lamp holder cracking parameter is matched from a preset lamp holder cracking parameter matching model according to the device number.

[0122] A sudden jump in the pressure value to a low level indicates that the lamp holder may be cracked. When the lamp holder is cracked, the metal halide bulb that was originally stably installed is likely to suddenly loosen, causing the pressure value to suddenly decrease. The lamp holder crack parameter matching model refers to a neural network model that has been trained with samples in advance. The lamp holder crack parameter matching model can match the lamp holder crack parameters that notify the staff of the lamp holder crack through the warning device. The lamp holder crack parameter matching model is common knowledge among people in this field and will not be described in detail here.

[0123] Step 415: Control a preset warning device to issue a lamp holder crack warning according to the lamp holder crack parameter.

[0124] When it is determined that the lamp holder is cracked, the staff is notified through the warning device, thereby facilitating the staff to perform maintenance in a timely manner.

[0125] The changes in pressure values ​​can be identified based on the historical records of pressure values, thereby detecting abnormal jumps, which makes it easier for staff to conduct targeted problem troubleshooting and facilitate the maintenance of lightning protection devices.

[0126] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A photovoltaic power station lightning protection device, characterized in that: The invention comprises a metal shell (1), wherein the metal shell (1) is filled with fly ash filler (8), a grounding terminal (2) for connecting to a grounding grid is arranged on the metal shell (1), a high-voltage bushing (3) is arranged between the metal shell (1) and a power bus, two shunt diodes (6) are symmetrically arranged on the high-voltage bushing (3), the shunt diodes (6) are connected to lamp holders (5), the lamp holders (5) are provided with metal halide bulbs (4), and capacitors (7) are arranged between the lamp holders (5) and the metal shell (1).

2. A photovoltaic power station lightning protection device according to claim 1, characterized in that: The connection directions of the two shunt diodes (6) and the high-voltage bushing (3) are opposite.

3. A maintenance method for a photovoltaic power station lightning protection device, applied to a photovoltaic power station lightning protection device according to any one of claims 1 to 2, characterized in that: include: Get the bus voltage of the power bus; When the bus voltage falls into a preset lightning strike interval, the bulb brightness of the metal halide lamp is obtained; Matching the overvoltage brightness from a preset overvoltage brightness matching model according to the bus voltage; Calculate the difference between the bulb brightness and the overvoltage brightness and define it as the brightness difference; When the brightness difference exceeds a preset brightness error range, the device number of the lightning protection device is obtained; Matching damage warning parameters from a preset damage warning parameter model according to the device number; According to the damage warning parameters, a preset warning device is controlled to issue a damage warning.

4. A method for maintaining a photovoltaic power station lightning protection device according to claim 3, characterized in that: Also included is a method for obtaining the brightness of a light bulb, the method for obtaining the brightness of a light bulb comprising: When the bus voltage falls into a preset lightning strike interval, controlling a preset photographing device to obtain a reference image of the metal halide lamp; matching the shooting position from a preset shooting position recognition model according to the reference image; Matching a shooting itinerary from a preset shooting itinerary matching model according to the shooting position; Controlling a shooting device preset on the metal housing to obtain a brightness image according to the shooting stroke; The bulb brightness is matched from a preset bulb brightness matching model according to the brightness image.

5. A method for maintaining a photovoltaic power station lightning protection device according to claim 4, characterized in that: The method for obtaining the brightness of the light bulb further includes: Matching the ambient brightness from a preset ambient brightness matching model according to the reference image; When the ambient brightness falls into the preset interference interval, the illumination direction is matched from the preset ambient light illumination direction recognition model according to the reference image; Matching a backlight position interval from a preset backlight position interval matching model according to the illumination direction; The shooting stroke is matched from a preset backlight stroke matching model according to the backlight position interval and the shooting position.

6. A method for maintaining a photovoltaic power station lightning protection device according to claim 5, characterized in that: The method for obtaining the brightness of the light bulb further includes: Get the light bulb image; Determine whether there are multiple lighted light bulbs in the light bulb image from a preset light bulb identification model according to the light bulb image; When there are multiple lighted light bulbs in the light bulb image, the light bulb positions are matched from a preset light bulb position recognition model according to the light bulb image; Matching the light bulb position from a preset light bulb position recognition model according to the brightness image; Match the light direction from a preset light direction recognition model according to the light bulb position and the light on position; Matching the backlight position interval from a preset backlight position interval matching model according to the light direction; The shooting stroke is matched from a preset backlight stroke matching model according to the backlight position interval and the shooting position.

7. A method for maintaining a photovoltaic power station lightning protection device according to claim 6, characterized in that: The method for acquiring the light bulb image comprises: Matching a shooting angle from a preset shooting angle matching model according to the reference image; determining whether the light bulbs in the reference image overlap according to the shooting angle and the shooting position from a preset light bulb overlap determination model; Judging whether a light bulb in the reference image is missing from a preset light bulb missing judgment model according to a shooting angle and a shooting position; When the light bulbs in the reference image overlap and / or are missing, an image acquisition itinerary is matched from a preset image acquisition itinerary matching model according to the shooting angle and the shooting position; The preset shooting device is controlled to acquire the image of the light bulb according to the image acquisition stroke.

8. A method for maintaining a photovoltaic power station lightning protection device according to claim 6, characterized in that: Also included is a damage analysis method, the damage analysis method comprising: When the brightness difference exceeds the preset brightness error range, the detection stroke is matched from the preset detection stroke matching model according to the position of the bulb; Controlling a preset shooting device to acquire a detection image set of the metal halide lamp in a surrounding manner according to the detection stroke; Determine whether the metal halide lamp is damaged based on the detection image set from the preset lamp appearance detection model; When the metal halide lamp bulb is damaged, the bulb damage parameters are matched from the preset bulb damage parameter matching model according to the device number; According to the bulb damage parameters, a preset warning device is controlled to issue a bulb damage warning.

9. A method for maintaining a photovoltaic power station lightning protection device according to claim 8, characterized in that: The damage analysis method further comprises: When the metal halide bulb is not damaged, the bulb model is matched from the preset metal halide bulb database according to the detection image set; According to the bulb model, the interference interval is matched from the preset interference interval matching model, and the pressure value at the bottom of the lamp holder is obtained; When the pressure value is lower than the conflict interval, the bulb loosening parameter is matched from the preset bulb loosening parameter matching model according to the device number; Controlling a preset warning device to issue a bulb loosening warning according to the bulb loosening parameter; When the pressure value exceeds the conflict interval, the bulb over-tightening parameter is matched from the preset bulb over-tightening parameter matching model according to the device number; According to the bulb over-tightening parameter, a preset warning device is controlled to issue a bulb over-tightening warning.

10. A method for maintaining a photovoltaic power station lightning protection device according to claim 9, characterized in that: The damage analysis method further comprises: When the pressure value is lower than the preset resistance interval, the bulb number is determined according to the bulb position; Retrieve pressure parameters from a preset pressure parameter database according to the device number and the bulb number; Determine whether a pressure jump occurs in the preset pressure jump recognition model according to the pressure parameter; When the pressure value jumps, the lamp holder cracking parameter is matched from the preset lamp holder cracking parameter matching model according to the device number; According to the cracking parameters of the lamp holder, a preset warning device is controlled to issue a lamp holder cracking warning.

Images