Power optimization method and system based on edge calculation, terminal and storage medium

Abstract

The invention relates to a power optimization method and system based on edge calculation, a terminal and a storage medium, and relates to the field of power data analysis, and the method comprises the steps: obtaining regional power data; matching equipment power from a preset equipment power matching model according to the regional power data; judging whether the equipment excessively consumes power or not from a preset power consumption judgment model according to the equipment power; when the equipment excessively consumes power, identifying a power supply partition from a preset power supply partition identification model according to the regional power data; and controlling a preset power distribution device to stop supplying power to the equipment according to the power supply zone. The application has the effects of improving the stability of power distribution and reducing the influence of fault equipment on the power system.

Description

Technical Field

[0001] The present invention relates to the field of power data analysis, and in particular to a power optimization method, system, terminal and storage medium based on edge computing. Background Art

[0002] Edge computing refers to the technology of data processing and storage at the edge of the power system network.

[0003] In the existing technology, the power in the power grid is generally monitored and allocated by the power grid management center. The power grid management center generally monitors and analyzes power data such as current, voltage, and load in the power grid to measure the situation in the power grid. When there is a lot of power data, the analysis pressure of the power grid management center is relatively large. Generally, the analysis pressure of the power grid management center is shared by adding edge computing equipment, so as to reasonably allocate power.

[0004] When an electrical device fails, it is easy to cause a sharp increase in the power consumption of the electrical equipment, resulting in an increase in the load on the power system and a waste of electricity. Summary of the invention

[0005] In order to improve the stability of power allocation and reduce the impact of faulty equipment on the power system, the present invention provides a power optimization method, system, terminal and storage medium based on edge computing.

[0006] In a first aspect, the present invention provides a power optimization method based on edge computing, which adopts the following technical solution:

[0007] A power optimization method based on edge computing, comprising:

[0008] Obtain regional power data;

[0009] Matching device power from a preset device power matching model based on regional power data;

[0010] Determining whether the device consumes excessive power from a preset power consumption judgment model according to the power of the device;

[0011] When the equipment consumes excessive power, the power supply partition is identified from a preset power supply partition identification model according to regional power data;

[0012] The preset power distribution device is controlled according to the power supply partition to stop supplying power to the equipment.

[0013] By adopting the above technical solution, regional power data is detected through edge computing devices and compared with historical data, so that when abnormal power consumption occurs in the power equipment, the power supply to the power equipment is disconnected, thereby reducing the increase in power system load caused by power equipment failure, which in turn leads to abnormal power allocation and improves the stability of power allocation.

[0014] Optionally, also include:

[0015] When a device consumes excessive power, the device number is matched from the device number matching model based on regional power data;

[0016] According to the device number, a device position corresponding to the device number is matched from a preset device position database;

[0017] Match the drone position from the preset drone position matching model according to the device position;

[0018] According to the position of the drone, the drone number is matched from the preset drone number matching model;

[0019] Matching the exploration route from the preset exploration route matching model according to the drone position and the equipment position;

[0020] According to the drone number, the preset drone is controlled to fly to the device corresponding to the device number along the exploration route, and the device image of the device is obtained;

[0021] Determine whether the device has a fault based on the device image and the preset fault judgment model;

[0022] When a device fails, the preset power distribution device is controlled according to the power supply partition to stop supplying power to the device.

[0023] By adopting the above technical solution, when abnormal power consumption occurs in power equipment, images of the abnormal power equipment are obtained through drones, so that it can be determined whether the power equipment has a fault by combining the image with power data, and the power supply to the power equipment is disconnected when a fault occurs, thereby ensuring the stability of power allocation.

[0024] Optionally, a power allocation method is also included, and the power allocation method includes:

[0025] Get total power data;

[0026] Matching the power demand from a preset power demand model according to the total power data;

[0027] When the demand power is not greater than the preset supply power, the difference between the preset supply power and the demand power is calculated and defined as the idle power;

[0028] Controlling the preset power distribution device to distribute power according to the power demand;

[0029] When the power demand is greater than the preset power supply, the difference between the power demand and the preset power supply is calculated and defined as the power difference;

[0030] Matching the adjusted power from a preset adjusted power matching model according to the power difference and the idle power;

[0031] The power distribution device distributes the power according to the preset power distribution device of the adjusted power control.

[0032] By adopting the above technical solution, the power demand of each line is estimated through historical data, so that the power distribution device is controlled to distribute power according to demand, thereby reducing power waste or power shortage and improving the stability of power allocation.

[0033] Optionally, it also includes a power output control method, the power output control method comprising:

[0034] When the idle power falls into the preset shortage interval, the power generation is matched from the preset power generation matching model according to the idle power;

[0035] Controlling the preset power generation device to generate electricity according to the power generation power;

[0036] When the idle power falls into the preset surplus range, the preset power generation device is controlled to stop generating power;

[0037] When the idle power falls into a preset reference interval, a preset power generation device is controlled to generate power according to the preset reference power.

[0038] By adopting the above technical solution, the working state of the power generation device is controlled according to the amount of idle power, so that the power of the power generation device is increased to supplement the power when the power is insufficient, and the power device is stopped to wait for the power to be consumed when the power is sufficient, thereby improving the stability of power allocation.

[0039] Optionally, a wire fault troubleshooting method is also included, and the wire fault troubleshooting method includes:

[0040] Judging whether there is a current jump from a preset current jump judgment model according to regional power data;

[0041] When there is a current jump, an inspection route is matched from a preset line inspection model based on regional power data and a preset grid laying route;

[0042] Control the preset drone to fly along the line according to the inspection route and obtain the wire images of the wires;

[0043] Judging whether there is a damage feature from a preset damage feature recognition model according to the wire image;

[0044] When there are damage features, identifying the damage location from a preset damage location identification model based on the damage features;

[0045] According to the damaged position, a preset warning device is controlled to warn the staff, and according to the damaged position, a preset power distribution device is controlled to stop supplying power to the damaged line.

[0046] By adopting the above technical solution, when the wires are damaged, it is easy to cause the current in the wires to jump. When the current jumps, the location of the damaged wires can be found by drones, so as to control the power distribution device to stop supplying power to the damaged wires, thereby reducing accidents caused by damaged wires.

[0047] Optionally, the wire troubleshooting method further includes:

[0048] When a damage feature exists, identifying whether a burning feature exists from a preset burning feature recognition model based on the wire image;

[0049] When there is a combustion feature, identifying the combustion area from a preset combustion area identification model based on the combustion feature;

[0050] Identify the wind direction from a preset wind direction identification model based on the wire image;

[0051] Matching the extinguishing stroke from a preset extinguishing stroke matching model according to the wind direction and the burning area;

[0052] According to the extinguishing stroke, the preset UAV is controlled to fly to the windward side of the burning area, and the extinguishing device preset on the UAV is controlled to extinguish the flame on the wire.

[0053] By adopting the above technical solution, electric sparks are easily generated when electric wires are damaged, and electric sparks are easy to ignite combustible materials in the wires and the surrounding environment. UAVs are used to detect whether there is a fire at the damaged part and extinguish the flame in time when a fire occurs, thereby reducing the situation of fire caused by circuit damage.

[0054] Optionally, a damage repair method is also included, and the damage repair method includes:

[0055] When there is a damage feature, judging whether there is a foreign object at the damaged part of the wire from a preset foreign object recognition model based on the damage feature;

[0056] When there is foreign matter at the damaged part of the wire, a cleaning stroke is matched from a preset cleaning stroke matching model based on the damage characteristics;

[0057] According to the cleaning schedule, the preset drone is controlled to fly to the damaged part of the wire, and the blowing device preset on the drone is controlled to blow away the foreign matter on the wire;

[0058] Based on the state that there is no foreign matter at the damaged part of the wire, a repair stroke is matched from a preset repair stroke matching model based on the damage characteristics;

[0059] According to the repairing itinerary, the preset UAV is controlled to fly to the damaged part of the electric wire, and the repairing device preset on the UAV is controlled to fill the preset repairing material into the damaged part of the electric wire.

[0060] By adopting the above technical solution, when the wire is damaged, the repair material is injected into the damaged part of the wire through the repair device on the drone for preliminary repair, thereby reducing the situation where the wire leaks from the damaged part and causes accidents such as electric shock.

[0061] In the second aspect, the present application provides a power optimization system based on edge computing, which adopts the following technical solutions:

[0062] A power optimization system based on edge computing, comprising:

[0063] An acquisition module, used to acquire regional power data, equipment images, total power data, and wire images;

[0064] A memory, used to store a program of any one of the control methods of the power optimization method based on edge computing;

[0065] The program in the processor memory can be loaded and executed by the processor to implement any of the above-mentioned control methods of the power optimization method based on edge computing.

[0066] In a third aspect, the present application provides a computer storage medium capable of storing corresponding programs, which is convenient for implementing a power optimization method based on edge computing, and adopts the following technical solutions:

[0067] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any of the above-mentioned edge computing-based power optimization methods.

[0068] In a fourth aspect, the present application provides an intelligent terminal, which adopts the following technical solution:

[0069] A smart terminal includes a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and execute any of the above-mentioned edge computing-based power optimization methods.

[0070] By adopting the above technical solution, regional power data is detected through edge computing devices and compared with historical data, so that when abnormal power consumption occurs in the power equipment, the power supply to the power equipment is disconnected, thereby reducing the increase in power system load caused by power equipment failure, which in turn leads to abnormal power allocation and improves the stability of power allocation.

[0071] In summary, the present application includes at least one of the following beneficial technical effects:

[0072] 1. Use edge computing devices to detect regional power data and compare it with historical data. When abnormal power consumption occurs in power equipment, the power supply to the power equipment is disconnected, thereby reducing the increase in power system load caused by power equipment failure, which in turn leads to abnormal power allocation and improves the stability of power allocation;

[0073] 2. When abnormal power consumption occurs in power equipment, the image of the abnormal power equipment is obtained through the drone, and then the image is combined with the power data to determine whether the power equipment is faulty. When a fault occurs, the power supply to the power equipment is disconnected, thereby ensuring the stability of power allocation;

[0074] 3. Use historical data to estimate the power demand of each line, so as to control the power distribution device to distribute power according to demand, thereby reducing power waste or power shortage and improving the stability of power allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 This is the process of the power optimization method based on edge computing Figure 1 ;

[0076] Figure 2 This is the process of the power optimization method based on edge computing Figure 2 ;

[0077] Figure 3 It is a flow chart of the power dispatching method;

[0078] Figure 4 is a flow chart of a power output control method;

[0079] Figure 5 Is the process of wire troubleshooting method Figure 1 ;

[0080] Figure 6 Is the process of wire troubleshooting method Figure 2 ;

[0081] Figure 7 It is a flowchart of the damage repair method. DETAILED DESCRIPTION

[0082] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the following Figure 1 -Attached Figure 7 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.

[0083] The embodiments of the present invention disclose an edge computing-based power optimization method, system, terminal and storage medium. The present invention detects the power grid status of a region through edge computing equipment, so as to promptly handle the abnormalities when the power grid is abnormal, reduce the situation where the abnormalities are not handled in time due to the abnormalities detected by the power grid management center, and improve the stability of power allocation.

[0084] Reference Figure 1 ,The power optimization method based on edge computing includes the following steps:

[0085] Step 100: Acquire regional power data.

[0086] Regional power data refers to the real-time power data of the area where the edge computing device is located, which is obtained through the edge computing device. The method for obtaining regional power data is selected by the staff according to the actual situation and will not be elaborated here.

[0087] Step 101: Match device power from a preset device power matching model according to regional power data.

[0088] The device power matching model refers to a neural network model obtained in advance through sample training. The device power matching model can analyze the device power consumed by each power device in the area where the edge computing device is located from the regional power data.

[0089] Step 102: Determine whether the device consumes excessive power from a preset power consumption determination model according to the device power.

[0090] The power consumption judgment model refers to a neural network model obtained in advance through sample training. The power consumption judgment model can judge whether the power consumption rate of the power equipment is too large according to the power of the equipment, that is, whether the power equipment consumes excessive power.

[0091] Step 103: When the device consumes excessive power, the power supply partition is identified from a preset power supply partition identification model according to the regional power data.

[0092] Excessive power consumption by equipment indicates that the power equipment may fail. Power supply partition refers to the power supply area divided in advance. The power grid systems of different power supply partitions do not affect each other. The stability of the power grid system is improved through power supply partition. The power supply partition is selected by the staff according to the actual situation, which will not be described here. The power supply partition identification model refers to the neural network model obtained through sample training in advance. The power supply partition identification model can match the power supply partition where the power equipment that may fail is located from the regional power data.

[0093] Step 104: Control the preset power distribution device to stop supplying power to the equipment according to the power supply partition.

[0094] The power distribution device refers to a device used to distribute power. The power distribution device is selected by the staff according to the actual situation and will not be described in detail here. The power distribution device is used to close the circuit that supplies power to the power supply partition where the power equipment that may fail is located, thereby reducing the impact of the failed power equipment on the power grid.

[0095] Reference Figure 2 , the power optimization method based on edge computing also includes the following steps:

[0096] Step 105: When the device consumes excessive power, the device number is matched from the device number matching model according to the regional power data.

[0097] The equipment number refers to a number set in advance to distinguish power equipment. Each equipment number corresponds to a power equipment one by one. The equipment number is selected by the staff according to the actual situation, which will not be described here. The equipment number matching model refers to a neural network model obtained through sample training in advance. The equipment number matching model can match the equipment number of the power equipment that may fail.

[0098] Step 106: Match the device location corresponding to the device number from a preset device location database according to the device number.

[0099] The device location database refers to a pre-set database that records device numbers and their corresponding device locations.

[0100] Step 107: Match the drone position from a preset drone position matching model according to the device position.

[0101] The drone position matching model refers to a neural network model obtained through sample training in advance. The drone position matching model can match the drone position of the drone closest to the device position.

[0102] Step 108: Match the drone number from a preset drone number matching model according to the drone position.

[0103] The drone number refers to a number set in advance to distinguish drones. Each drone number corresponds to a drone one-to-one. The drone number is selected by the staff according to the actual situation, which will not be described here. The drone number matching model refers to a neural network model obtained through sample training in advance. The drone number matching model can match the drone number of the drone located at the drone position. When there are multiple drones at the drone position, the drone number of a drone is randomly selected.

[0104] Step 109: Match the exploration route from the preset exploration route matching model according to the drone position and the device position.

[0105] The exploration route matching model refers to a neural network model obtained through sample training in advance. The exploration route matching model can match the exploration route of the drone from the drone position to the device position.

[0106] Step 110: According to the drone number, a preset drone is controlled to fly to the device corresponding to the device number along the exploration route, and a device image of the device is obtained.

[0107] Drones refer to equipment used to determine the actual situation of the power grid system. Drones are selected by staff based on actual conditions and are not described in detail here. Equipment images refer to pictures of power equipment corresponding to equipment numbers. Equipment images can be obtained through cameras installed on drones. The method for obtaining equipment images is selected by staff based on actual conditions and is not described in detail here.

[0108] Step 111: Determine whether a device failure occurs based on the device image from a preset fault judgment model.

[0109] The fault judgment model refers to a neural network model obtained in advance through sample training. The fault judgment model can combine the working status of the power equipment and the power data displayed in the equipment image to determine whether the power equipment has a fault.

[0110] Step 112: When a device fails, a preset power distribution device is controlled according to the power supply partition to stop supplying power to the device.

[0111] If a device fails, the power distribution device is controlled to stop supplying power to the failed power device. If the device does not fail, the power distribution device is controlled to keep supplying power to the power device, thereby reducing the possibility of misjudgment of equipment failures through drones.

[0112] Reference Figure 3 , the power allocation method includes the following steps:

[0113] Step 200: Obtain total power data.

[0114] The total power data refers to the overall power data of the power grid obtained by the power grid management center. The method for obtaining the total power data is selected by the staff according to the actual situation and will not be elaborated here.

[0115] Step 201: matching the power demand from a preset power demand model according to the total power data.

[0116] The power demand model refers to a neural network model obtained in advance through sample training. The power demand model can match the power demand from the historical total power data based on the actual total power data.

[0117] Step 202: When the required power is not greater than the preset supplied power, the difference between the preset supplied power and the required power is calculated and defined as the idle power.

[0118] The supply power refers to the power supply of the pre-set power grid. The supply power is selected by the staff according to the actual situation, which will not be elaborated here. The demand power is not greater than the supply power, which means that the power supply of the power grid is sufficient at this time. The idle power is the surplus part of the supply power.

[0119] Step 203: Control the preset power distribution device to distribute power according to the power demand.

[0120] The power distribution device adjusts the power supplied by the power grid to the required power, thereby reserving idle power for use by other circuits.

[0121] Step 204: When the required power is greater than the preset supplied power, the difference between the required power and the preset supplied power is calculated and defined as the power difference.

[0122] The demand for power is greater than the supply power, which means that the power supplied by the power grid is insufficient. The idle power needs to be diverted to the power supply circuit to meet the demand for power. The power difference is the shortfall in the supply power.

[0123] Step 205: Match the adjusted power from a preset adjusted power matching model according to the power difference and the idle power.

[0124] The adjustment power matching model refers to a neural network model obtained in advance through sample training. The adjustment power matching model can match the adjustment power that meets the demand power through idle power. When the idle power is greater than the power difference, the adjustment power is the demand power. When the idle power is not greater than the power difference, the adjustment power is the sum of the supply power and the idle power.

[0125] Step 206: Distribute power according to the power distribution device preset by adjusting the power control.

[0126] The power distribution device is controlled to distribute power according to the power demand, thereby optimizing the power supply structure of the power grid and reducing the situation of insufficient power supply or excessive power supply.

[0127] Reference Figure 4 , the power output control method comprises the following steps:

[0128] Step 300: When the idle power falls into a preset shortage interval, the power generation power is matched from a preset power generation power matching model according to the idle power.

[0129] The shortage interval refers to the pre-set idle power shortage power value interval, which is selected by the staff according to the actual situation and will not be described here. The power generation matching model refers to the neural network model obtained through sample training in advance, and the power generation matching model can match the power generation required to increase the idle power generator.

[0130] Step 301: Control a preset power generation device to generate power according to the generated power.

[0131] The power generation device refers to a device used to supply electricity. The power generation device is selected by the staff according to the actual situation and will not be described in detail here. By increasing the power generation capacity of the power generation device to increase idle power, the situation where the idle power is insufficient and the power demand cannot be met is reduced.

[0132] Step 302: When the idle power falls into a preset surplus range, the preset power generation device is controlled to stop generating power.

[0133] The surplus interval refers to the pre-set power value interval with excessive idle power. The surplus interval is selected by the staff according to the actual situation and will not be described here. When the idle power falls into the surplus interval, it means that a large amount of power is idle, that is, the power supply is too much, and the power generation device is controlled to stop generating power to consume the existing idle power.

[0134] Step 303: When the idle power falls into a preset reference interval, the preset power generation device is controlled to generate power according to the preset reference power.

[0135] The benchmark interval refers to a pre-set range of power values ​​with sufficient idle power, and the benchmark power refers to the pre-set working power of the power generation device used to keep the idle power within the benchmark interval. The benchmark interval and benchmark power are selected by the staff based on actual conditions and will not be elaborated here.

[0136] Reference Figure 5 , the wire troubleshooting method includes the following steps:

[0137] Step 400: Determine whether there is a current jump from a preset current jump determination model according to regional power data.

[0138] The current jump judgment model refers to a neural network model obtained through sample training in advance. The current jump judgment model can identify whether there is a current jump in the regional power data.

[0139] Step 401: When there is a current jump, an inspection route is matched from a preset line inspection model based on regional power data and a preset grid laying route.

[0140] The existence of current jumps indicates that the power supply line may be damaged and cause leakage. The power grid laying route refers to the laying route of the power grid in the area where the edge computing device is located. The power grid laying route is selected by the staff according to the actual situation and will not be described here. The line inspection model refers to the neural network model obtained through sample training in advance. The line inspection model can match the inspection route for inspecting the power supply line of the power grid through drones.

[0141] Step 402: Control a preset UAV to fly along the line according to the inspection route, and obtain wire images of the wires.

[0142] The wire image refers to a picture of the wire, which can be obtained through a camera installed on a drone. The method of obtaining the wire image is selected by the staff according to the actual situation, and will not be described here. The wire is inspected by a drone to determine the location of the damaged wire.

[0143] Step 403: Determine whether there is a damage feature from a preset damage feature recognition model based on the wire image.

[0144] The damaged feature recognition model refers to a neural network model obtained through sample training in advance. The damaged feature recognition model can determine whether the wire in the wire image has damaged features.

[0145] Step 404: When damage features exist, the damage location is identified from a preset damage location identification model based on the damage features.

[0146] The damage location recognition model refers to a neural network model obtained through sample training in advance. The damage location recognition model can recognize the damage location of the damage features.

[0147] Step 405: Control a preset warning device to warn the staff according to the damaged position, and control a preset power distribution device to stop supplying power to the damaged line according to the damaged position.

[0148] The warning device is a device used to warn the staff so that they can deal with the wire damage in time. The warning device is selected by the staff according to the actual situation and will not be described here. The warning device can be used to notify the staff in time so that they can deal with the damage, and the power distribution device can be used to disconnect the circuit that supplies power to the damaged location, thereby reducing the occurrence of electric shock accidents.

[0149] Reference Figure 6 , the wire troubleshooting method also includes the following steps:

[0150] Step 406: When a damage feature exists, whether a burning feature exists is identified from a preset burning feature recognition model based on the wire image.

[0151] The combustion feature recognition model refers to a neural network model obtained through sample training in advance. The combustion feature recognition model can recognize whether there are combustion features in the wire image.

[0152] Step 407: When the combustion feature exists, the combustion region is identified from a preset combustion region identification model based on the combustion feature.

[0153] The presence of burning features indicates that the wire is burning and the flame needs to be extinguished in time to reduce further damage to the wire. The burning area recognition model refers to a neural network model obtained through sample training in advance. The burning area recognition model can identify the burning area where the burning features are located.

[0154] Step 408: Identify the wind direction from a preset wind direction identification model based on the wire image.

[0155] The wind direction recognition model refers to a neural network model obtained through sample training in advance. The wind direction recognition model can identify the wind direction in the burning area from the power line image.

[0156] Step 409: Matching the extinguishing stroke from a preset extinguishing stroke matching model according to the wind direction and the combustion area.

[0157] The extinguishing device refers to a device used to spray fireproof materials to extinguish flames, where the fireproof materials are made of materials with poor electrical conductivity such as dry powder. The extinguishing device is selected by the staff according to the actual situation and will not be described in detail here. The extinguishing stroke matching model refers to a neural network model obtained through sample training in advance. The extinguishing stroke matching model can match the extinguishing stroke that uses the extinguishing device on the drone to extinguish the burning area and reduce the impact of the smoke generated by the burning on the drone.

[0158] Step 410: Control the preset UAV to fly to the windward side of the burning area according to the extinguishing stroke, and control the extinguishing device preset on the UAV to extinguish the flame on the wire.

[0159] Electric sparks are easily generated at damaged wires, causing them to burn. The drone is flown to the windward side of the burning area to reduce the smoke from the burning from directly hitting the drone and causing the drone to malfunction. The drone's flight attitude is adjusted through the extinguishing stroke to adjust the spray direction of the extinguishing device, and the flame area is covered with fireproof materials to extinguish the flame.

[0160] Reference Figure 7 , the damage repair method includes the following steps:

[0161] Step 500: When there is a damage feature, determine whether there is a foreign object at the damaged part of the wire based on the damage feature from a preset foreign object recognition model.

[0162] Foreign matter refers to objects such as plastic waste that cover the damaged features. The foreign matter recognition model refers to a neural network model obtained through sample training in advance. The foreign matter recognition model can determine whether there is a foreign matter at the damaged part of the wire.

[0163] Step 501: When there is foreign matter at the damaged part of the wire, a cleaning stroke is matched from a preset cleaning stroke matching model based on the damage feature.

[0164] The blowing device refers to a device installed on the drone that blows air to make foreign objects fly away from the damaged features. The blowing device is selected by the staff according to the actual situation and will not be described here. The cleaning stroke matching model refers to a neural network model obtained through sample training in advance. The cleaning stroke matching model can match the cleaning stroke that blows away foreign objects through the blowing device.

[0165] Step 502: Control a preset drone to fly to the damaged part of the wire according to the cleaning process, and control a blowing device preset on the drone to blow away foreign matter on the wire.

[0166] The posture of the drone is adjusted through the cleaning process, thereby adjusting the direction of the blowing device, and then the foreign matter is blown away by the blowing device.

[0167] Step 503: Based on the state that there is no foreign matter at the damaged part of the electric wire, a repair stroke is matched from a preset repair stroke matching model based on the damage feature.

[0168] The repair device refers to a device installed on the drone for preliminary repair of the damaged part. The repair device is selected by the staff according to the actual situation and will not be described in detail here. The repair stroke matching model refers to a neural network model obtained through sample training in advance. The repair stroke matching model can match the repair stroke of aiming the repair device at the damaged part of the wire and repairing the damaged part through the repair device.

[0169] Step 504: Control the preset UAV to fly to the damaged part of the wire according to the repairing itinerary, and control the repairing device preset on the UAV to fill the preset repairing material into the damaged part of the wire.

[0170] The repair material refers to the material stored in the repair device for filling the damaged part of the wire to repair the wire. The repair material is selected by the staff according to the actual situation and will not be described here. The attitude of the drone is adjusted through the repair process so that the repair device is aimed at the damaged part of the wire, and the repair material is injected onto the wire through the repair device.

[0171] Based on the same inventive concept, an embodiment of the present invention provides a power optimization system based on edge computing, including:

[0172] An acquisition module, used to acquire regional power data, equipment images, total power data, and wire images;

[0173] A memory, used to store a program of any one of the control methods of the power optimization method based on edge computing;

[0174] The program in the processor memory can be loaded and executed by the processor to implement any of the above-mentioned control methods of the power optimization method based on edge computing.

[0175] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0176] An embodiment of the present invention provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute an edge computing-based power optimization method.

[0177] Computer storage media include, for example, various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0178] Based on the same inventive concept, an embodiment of the present invention provides a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes an edge computing-based power optimization method.

[0179] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0180] 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 power optimization method based on edge computing, characterized in that: include: Obtain regional power data; Matching device power from a preset device power matching model based on regional power data; Determining whether the device consumes excessive power from a preset power consumption judgment model according to the power of the device; When the equipment consumes excessive power, the power supply partition is identified from a preset power supply partition identification model according to regional power data; The preset power distribution device is controlled according to the power supply partition to stop supplying power to the equipment.

2. The power optimization method based on edge computing according to claim 1, characterized in that: Also includes: When a device consumes excessive power, the device number is matched from the device number matching model based on regional power data; According to the device number, a device position corresponding to the device number is matched from a preset device position database; Match the drone position from the preset drone position matching model according to the device position; According to the position of the drone, the drone number is matched from the preset drone number matching model; Matching the exploration route from the preset exploration route matching model according to the drone position and the equipment position; According to the drone number, the preset drone is controlled to fly to the device corresponding to the device number along the exploration route, and the device image of the device is obtained; Determine whether the device has a fault based on the device image and the preset fault judgment model; When a device fails, the preset power distribution device is controlled according to the power supply partition to stop supplying power to the device.

3. The power optimization method based on edge computing according to claim 1, characterized in that: Also included is a power allocation method, the power allocation method comprising: Get total power data; Matching the power demand from a preset power demand model according to the total power data; When the demand power is not greater than the preset supply power, the difference between the preset supply power and the demand power is calculated and defined as the idle power; Controlling the preset power distribution device to distribute power according to the power demand; When the power demand is greater than the preset power supply, the difference between the power demand and the preset power supply is calculated and defined as the power difference; Matching the adjusted power from a preset adjusted power matching model according to the power difference and the idle power; The power distribution device distributes the power according to the preset power distribution device of the adjusted power control.

4. The power optimization method based on edge computing according to claim 3 is characterized in that: Also included is a power output control method, the power output control method comprising: When the idle power falls into the preset shortage interval, the power generation is matched from the preset power generation matching model according to the idle power; Controlling the preset power generation device to generate electricity according to the power generation power; When the idle power falls into the preset surplus range, the preset power generation device is controlled to stop generating power; When the idle power falls into a preset reference interval, a preset power generation device is controlled to generate power according to the preset reference power.

5. The power optimization method based on edge computing according to claim 1, characterized in that: Also included is a wire troubleshooting method, the wire troubleshooting method comprising: Judging whether there is a current jump from a preset current jump judgment model according to regional power data; When there is a current jump, an inspection route is matched from a preset line inspection model based on regional power data and a preset grid laying route; Control the preset drone to fly along the line according to the inspection route and obtain the wire images of the wires; Judging whether there is a damage feature from a preset damage feature recognition model according to the wire image; When there are damage features, identifying the damage location from a preset damage location identification model based on the damage features; According to the damaged position, a preset warning device is controlled to warn the staff, and according to the damaged position, a preset power distribution device is controlled to stop supplying power to the damaged line.

6. The power optimization method based on edge computing according to claim 5, characterized in that: The wire troubleshooting method further comprises: When a damage feature exists, identifying whether a burning feature exists from a preset burning feature recognition model based on the wire image; When there is a combustion feature, identifying the combustion area from a preset combustion area identification model based on the combustion feature; Identify the wind direction from a preset wind direction identification model based on the wire image; Matching the extinguishing stroke from a preset extinguishing stroke matching model according to the wind direction and the burning area; According to the extinguishing stroke, the preset UAV is controlled to fly to the windward side of the burning area, and the extinguishing device preset on the UAV is controlled to extinguish the flame on the wire.

7. The power optimization method based on edge computing according to claim 5, characterized in that: Also included is a damage repair method, the damage repair method comprising: When there is a damage feature, judging whether there is a foreign object at the damaged part of the wire from a preset foreign object recognition model based on the damage feature; When there is foreign matter at the damaged part of the wire, a cleaning stroke is matched from a preset cleaning stroke matching model based on the damage characteristics; According to the cleaning schedule, the preset drone is controlled to fly to the damaged part of the wire, and the blowing device preset on the drone is controlled to blow away the foreign matter on the wire; Based on the state that there is no foreign matter at the damaged part of the wire, a repair stroke is matched from a preset repair stroke matching model based on the damage characteristics; According to the repairing itinerary, the preset UAV is controlled to fly to the damaged part of the electric wire, and the repairing device preset on the UAV is controlled to fill the preset repairing material into the damaged part of the electric wire.

8. A power optimization system based on edge computing, characterized in that: include: An acquisition module, used to acquire regional power data, equipment images, total power data, and wire images; A memory for storing a program of a control method of a power optimization method based on edge computing according to any one of claims 1 to 7; The program in the memory can be loaded and executed by the processor to implement the control method of the power optimization method based on edge computing as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the power optimization method based on edge computing as described in any one of claims 1 to 7.

10. An intelligent terminal, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program that can be loaded by the processor and executes the power optimization method based on edge computing as described in any one of claims 1 to 7.

Images