An AI-based substation fire warning method and system

Abstract

The application provides a substation fire warning method and system based on AI; wherein the method comprises: receiving first image data of a first monitoring area of a substation, processing the first image data to determine a first target monitoring area; inputting second image data of the first target monitoring area into a flame AI recognition model, the flame AI recognition model outputs a flame recognition result; and determining whether to output a fire warning according to the flame recognition result. The scheme of the application is targeted to monitor the areas where the fire may occur, can reduce the data processing load, and improve the flame recognition efficiency; and the application also uses AI technology to identify the flame, the artificial participation workload is low, and the accuracy of the flame recognition is higher.

Description

Technical Field

[0001] The present invention relates to the field of fire warning technology, and in particular to an AI-based substation fire warning method, system, electronic equipment, and computer storage medium. Background Art

[0002] Substations are the hub of the power system. A fire can disrupt the entire power grid, severely compromising power supply reliability. Therefore, effective fire prevention measures in substations are crucial to ensuring the safe and stable operation of the power grid. Conventional fire monitoring uses smoke sensors to detect smoke concentrations. However, substation electrical equipment is mostly outdoors, making it difficult for smoke concentrations to reach the smoke sensor's trigger threshold. Consequently, smoke sensors can only respond to relatively severe fires. This clearly falls short of substation fire monitoring expectations and urgently requires improvement. Summary of the Invention

[0003] In order to at least solve the technical problems existing in the above-mentioned background technology, the present invention provides an AI-based substation fire warning method, system, electronic device and computer storage medium.

[0004] A first aspect of the present invention provides an AI-based substation fire early warning method, comprising the following steps:

[0005] receiving first image data of a first monitoring area of ​​a substation, and processing the first image data to determine a first target monitoring area;

[0006] Inputting the second image data of the first target monitoring area into a flame AI recognition model, wherein the flame AI recognition model outputs a flame recognition result;

[0007] Determine whether to output a fire warning based on the flame recognition result.

[0008] Furthermore, the processing of the first image data to determine a first target monitoring area includes:

[0009] performing foreground extraction on the first image data to obtain a number of candidate objects;

[0010] Comparing the candidate objects with pre-stored objects, and determining target objects based on the comparison results;

[0011] The area where each target object is located is determined as the first target monitoring area.

[0012] Furthermore, the processing of the first image data to determine the first target monitoring area further includes:

[0013] acquire third image data of a second monitoring area of the transformer substation, and process the third image data to determine a second target monitoring area; wherein the second monitoring area is adjacent to the first monitoring area;

[0014] screen a plurality of fourth target monitoring areas from the third image data; wherein the fourth target monitoring area is a region not located in the first monitoring area but with a distance less than or equal to a first threshold value from the first monitoring area;

[0015] determine a fifth target monitoring area in the first monitoring area according to the fourth target monitoring area, and take the fifth target monitoring area as a part of the first target monitoring area.

[0016] Further, the determination of the fifth target monitoring area in the first monitoring area according to the fourth target monitoring area comprises: determining a second position in the first image data according to a first position of the fourth target monitoring area in the third image data.

[0017] determine a first equivalent value of the first monitoring area according to a first feature in the first image data, determine a second equivalent value of the second monitoring area according to a second feature in the third image data, and determine a third equivalent value according to the first equivalent value and the second equivalent value.

[0018] determine the fifth target monitoring area in the first image data according to the second position and the third equivalent value.

[0019] Further, the flame AI recognition model is constructed by any one of back propagation, Boltzmann machine, convolutional neural network, Hopfield network, multilayer perceptron, radial basis function network, restricted Boltzmann machine, regression neural network, self-organizing map, and spiking neural network.

[0020] Further, the determination of whether to output the fire warning according to the flame recognition result comprises:

[0021] If the flame recognition result is that the flame exists, calculate an area ratio of a connected domain of the flame in the second image data, and if the area ratio is greater than or equal to a second threshold value, determine to output the fire warning;

[0022] If the area ratio is less than the second threshold value, extract personnel data according to the first image data, and determine whether to output the fire warning according to the extraction result.

[0023] Further, the determination of whether to output the fire warning according to the extraction result comprises:

[0024] If the personnel data is not extracted, it is determined to output a fire warning;

[0025] If the personnel data is extracted, a matching degree of the personnel data and target personnel data is calculated, if the matching degree meets a preset condition, a fire warning is not output, otherwise, a fire warning is output.

[0026] The second aspect of the present application provides an AI-based substation fire warning system, comprising an acquisition module, a processing module and a storage module; the processing module is connected with the acquisition module and the storage module;

[0027] The storage module is used for storing executable computer program codes;

[0028] The acquisition module is used for acquiring image data and transmitting the image data to the processing module;

[0029] The processing module is used for executing the method according to any one of the preceding aspects by calling the executable computer program codes in the storage module.

[0030] The third aspect of the present application provides an electronic device, comprising a memory storing executable program codes, and a processor coupled with the memory; the processor calls the executable program codes stored in the memory to execute the method according to any one of the preceding aspects.

[0031] The fourth aspect of the present application provides a computer storage medium, which stores a computer program; when the computer program is run by a processor, the method according to any one of the preceding aspects is executed.

[0032] According to the scheme of the present application, the whole image of the monitored area is not subjected to flame recognition, but the area where fire is likely to occur is monitored, so that the data processing load is reduced and the flame recognition efficiency is improved; in addition, the AI technology is used for flame recognition, compared with the traditional flame recognition method, the flame AI recognition model of the present application only needs to be pre-trained using image data labeled by humans, the workload of human participation is low, and with the increase of the amount of flame image data in various scenes, the accuracy of flame recognition can be continuously improved. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0034] Figure 1 is a flow diagram of an AI-based substation fire warning method disclosed by an embodiment of the application;

[0035] Figure 2 is a structural diagram of an AI-based substation fire warning system disclosed by an embodiment of the application;

[0036] Figure 3 is a structural diagram of an electronic device disclosed by an embodiment of the application. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0038] The terms used in the embodiments of the present application are merely for the purpose of describing particular embodiments and are not intended to limit the present application. The singular forms "a", "said" and "the" used in the embodiments of the present application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. "Multiple" generally includes at least two.

[0039] It should be understood that the term "and / or" used herein is merely to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone. In addition, the character " / " herein generally represents an "or" relationship between the front and rear associated objects.

[0040] It should be understood that although the terms first, second, third, etc. may be used in the embodiments of the present application to describe …, these … should not be limited to these terms. These terms are only used to distinguish … from each other. For example, without departing from the scope of the embodiments of the present application, the first … can also be called the second …, and similarly, the second … can also be called the first ….

[0041] Depending on the context, the word "if" as used herein can be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting". Similarly, depending on the context, the phrase "if it is determined" or "if (a stated condition or event) is detected" can be interpreted as "when it is determined" or "in response to determining" or "when (a stated condition or event) is detected" or "in response to detecting (a stated condition or event)".

[0042] It is also important to note that the terms "comprises" and / or "comprising," or other variations such as "includes," "including" or "contains," "containing," etc., shall not be construed as the use of an exclusive or exhaustive listing of the elements described. That is, use of such terms is intended to be inclusive of the elements recited and of additional elements not specifically recited. Further, it is to be understood that the use of the term "or" is inclusive of the conjunctive and disjunctive use of the term.

[0043] Preferred embodiments of the present application will be described in detail below with reference to the attached drawings.

[0044] Embodiment One

[0045] Please refer to Figure 1 , Figure 1 is a flowchart of an AI-based substation fire warning method disclosed by an embodiment of the present application. As shown in Figure 1 , the AI-based substation fire warning method of the embodiment of the present application includes the following steps:

[0046] receiving first image data of a first monitoring area of a substation, processing the first image data to determine a first target monitoring area;

[0047] inputting second image data of the first target monitoring area into a flame AI recognition model, the flame AI recognition model outputting a flame recognition result;

[0048] determining whether to output a fire warning according to the flame recognition result.

[0049] In the embodiment of the present application, as described in the background, smoke sensors are not suitable for substation monitoring scenes located outdoors, so the present application uses image recognition technology to monitor and warn of fires in substations. Specifically, a plurality of first target monitoring areas are determined from first image data of a first monitoring area, and a pre-set flame AI recognition model is used to analyze whether there is a flame in the second image data corresponding to the first target monitoring area, and a fire warning decision is made accordingly. The scheme of the present application does not perform flame recognition on the entire image of the monitoring area, but monitors the areas where fires are likely to occur, which can reduce the data processing load and improve the flame recognition efficiency. In addition, the present application uses AI technology to recognize flames. Compared with traditional flame recognition methods, the flame AI recognition model of the present application only needs to be pre-trained using manually annotated image data, which reduces the amount of human work, and as the amount of flame image data in various scenes increases, the accuracy of flame recognition can also be continuously improved.

[0050] Further, the processing of the first image data to determine the first target monitoring area comprises:

[0051] foreground extraction is performed on the first image data to obtain a plurality of candidate objects;

[0052] a comparison calculation is performed between the plurality of candidate objects and a pre-stored object, and a plurality of target objects are determined according to a comparison calculation result;

[0053] an area where each of the target objects is located is determined as the first target monitoring area.

[0054] In the embodiment of the present application, fire is unlikely to occur in some areas of the substation, such as cement ground, isolation pillars, etc., therefore, the present application screens the plurality of target objects that are likely to cause fire in the monitoring area through the above processing, and determines the area where the target objects are located as the first target monitoring area. In this way, the data processing amount can be reduced, and the fire identification efficiency can be improved.

[0055] Further, the processing of the first image data to determine the first target monitoring area further comprises:

[0056] third image data of a second monitoring area of the substation is obtained, and the third image data is processed to determine a second target monitoring area; wherein the second monitoring area is adjacent to the first monitoring area;

[0057] a plurality of fourth target monitoring areas are screened from the third image data; wherein the fourth target monitoring area is an area that is not located in the first monitoring area but has a distance less than or equal to a first threshold value from the first monitoring area;

[0058] a fifth target monitoring area is determined in the first monitoring area according to the fourth target monitoring area, and the fifth target monitoring area is taken as a part of the first target monitoring area.

[0059] In the embodiment of the present application, the monitoring camera of the transformer substation has a fixed angle type and an angle adjustable type, so that different first monitoring areas and second monitoring areas corresponding to different monitoring angles are generated. In view of the actual situation, the present application further screens a second target monitoring area (i.e. a fourth target monitoring area) close to the current first monitoring area in the prior second monitoring area, determines a fifth target monitoring area located in the first monitoring area according to the fourth target monitoring area, and supplements the fifth target monitoring area into the original first target detection area. Because the fourth target monitoring area is relatively close to the fifth target monitoring area, the fire condition is easy to be detected in the fifth target monitoring area. In this way, when the first monitoring area is monitored, the fire condition in the original second monitoring area can also be monitored, and under the premise of reducing the image data processing load, the omission of fire identification is also avoided as much as possible.

[0060] It should be noted that the first monitoring area and the second monitoring area may overlap or may not overlap at all, so the present application sets the condition of "not located in the first monitoring area" when determining the fourth target monitoring area, which can further reduce the number of first target detection areas, thereby reducing the image data processing load. In addition, the first threshold value can be determined based on the rotation angle of the camera. The greater the rotation angle, the smaller the first threshold value, that is, the area closer to the side edge of the third image data (corresponding to the first image data) is taken as the fourth target monitoring area, otherwise, the greater the first threshold value, that is, the area farther away from the side edge of the third image data (corresponding to the first image data) is taken as the fourth target monitoring area.

[0061] Further, the fifth target monitoring area is determined in the first monitoring area according to the fourth target monitoring area, comprising: determining a second position in the first image data according to a first position of the fourth target monitoring area in the third image data.

[0062] The first equivalent value of the first monitoring area is determined according to the first feature in the first image data, the second equivalent value of the second monitoring area is determined according to the second feature in the third image data, and the third equivalent value is determined according to the first equivalent value and the second equivalent value.

[0063] The fifth target monitoring area is determined in the first image data according to the second position and the third equivalent value.

[0064] In the embodiment of the present application, after switching the monitoring area, the depth of field and the camera focal length may change, which may cause the proportions of objects in the third image data of the second monitoring area and the first image data of the first monitoring area to be inconsistent. Accordingly, the fifth target monitoring area cannot be determined in the first image data according to the size of the fourth target monitoring area in the third image data. To solve this problem, the present application determines the equivalent values of the features in the first image data and the second image data, respectively, and then calculates a third equivalent value reflecting the difference between the two images through fusion calculation. Then, the fifth target monitoring area can be determined according to the third equivalent value and the previously determined second position.

[0065] The first feature and the second feature can be the same or different. For example, when the first image data and the second image data overlap, the same feature is preferred. The third equivalent value can be determined according to the proportional relationship between the first equivalent value and the second equivalent value, or the proportional relationship of the two relative to a reference value, without limitation.

[0066] Further, the flame AI recognition model is constructed by any one of back propagation, Boltzmann machine, convolutional neural network, Hopfield network, multilayer perceptron, radial basis function network, restricted Boltzmann machine, regression neural network, self-organizing map, and spiking neural network.

[0067] In the embodiment of the present application, the present application can be constructed based on any of the above AI algorithms, and the above AI algorithms also include improvements and variants of the AI algorithms.

[0068] Further, the determination of whether to output a fire warning according to the flame recognition result comprises:

[0069] If the flame recognition result is that a flame exists, the area ratio of the connected domain of the flame in the second image data is calculated. If the area ratio is greater than or equal to a second threshold value, it is determined to output a fire warning.

[0070] If the area ratio is less than the second threshold value, personnel data is extracted according to the first image data, and it is determined whether to output a fire warning according to the extraction result.

[0071] In the embodiment of the present application, the flame AI recognition model can not only recognize large-area flames and smoke, but also recognize electric arcs, which are obviously smaller in area than flames and smoke. Therefore, the difference in area ratio can be used for distinguishing and deciding whether to output a fire warning. In addition, the appearance of electric arcs may be caused by normal operation of personnel or abnormal situations. Therefore, the present application further detects personnel data to further decide whether to issue a warning for small-area electric arcs.

[0072] Further, the determining whether to output a fire warning according to the extraction result comprises:

[0073] If the personnel data is not extracted, it is determined to output a fire warning.

[0074] If the personnel data is extracted, the matching degree of the personnel data and the target personnel data is calculated. If the matching degree meets the preset condition, no fire warning is output, otherwise, a fire warning is output.

[0075] In the embodiment of the present application, the present application analyzes whether there is personnel in the monitoring area and whether the personnel is performing a specified operation. If the conditions are met, it means that the appearance of electric arcs is caused by normal operation and no warning should be output, thereby improving the reliability of the warning. For example, when the relevant personnel of a substation controls the closure of the electrical contact, electric arcs will be generated. At this time, no warning is needed.

[0076] The matching degree and the aforementioned comparison calculation can be obtained by calculating the similarity, and the similarity can be calculated by using any suitable calculation formula in the prior art, which is not limited.

[0077] Embodiment two

[0078] Please refer to Figure 2 , Figure 2 is a structural schematic diagram of a substation fire warning system based on AI disclosed by the embodiment of the present application. As shown in Figure 2 The substation fire warning system based on AI of the embodiment of the present application comprises an acquisition module (101), a processing module (102) and a storage module (103). The processing module (102) is connected with the acquisition module (101) and the storage module (103).

[0079] The storage module (103) is used for storing executable computer program codes.

[0080] The acquisition module (101) is used for acquiring image data and transmitting the image data to the processing module (102).

[0081] The processing module (102) is configured to execute the method according to any one of the preceding embodiments by invoking the executable computer program code in the storage module (103).

[0082] The specific function of the AI-based substation fire warning system in this embodiment refers to the above-mentioned embodiment one, since the system in this embodiment adopts all the technical solutions of the above-mentioned embodiments, at least has all the beneficial effects brought by the technical solutions of the above-mentioned embodiments, which will not be repeated here.

[0083] Embodiment three

[0084] Please refer to Figure 3 , Figure 3 The electronic device disclosed in the embodiment of the present application comprises a memory in which executable program codes are stored, a processor coupled with the memory, and the processor invokes the executable program codes stored in the memory to execute the method according to the embodiment one.

[0085] Embodiment four

[0086] The embodiment of the present application further discloses a computer storage medium, and the computer storage medium stores a computer program, and the computer program is executed by a processor to execute the method according to the embodiment one.

[0087] The computer storage medium of the embodiment of the present application can adopt any combination of one or more computer readable media. The computer readable medium can be a computer readable signal medium or a computer readable storage medium. The computer readable storage medium may, for example, be, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system or device, or any combination of the above. More specific examples (non-exhaustive list) of the computer readable storage medium include: an electrical connection with one or more conductive wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system or device.

[0088] A computer readable signal medium can include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated signal can take any of a variety of forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium can be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate or transport program code there within for use by or in connection with an instruction execution system, apparatus, or device.

[0089] Program code embodied on a computer readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire line, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0090] Computer program code for carrying out operations for aspects of the present application can be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C++ or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0091] It is to be understood that the above description is intended to be illustrative, and not restrictive. Many other embodiments will be apparent to those of skill in the art upon reviewing the above description. The scope of the application should, therefore, be determined not with reference to the above description, but instead with reference to the appended claims, along with their full scope of equivalents.

Claims

1. An AI-based substation fire early warning method, characterized in that: The process includes the following steps: receiving first image data of a first monitoring area of ​​a substation, and processing the first image data to determine a first target monitoring area; Inputting the second image data of the first target monitoring area into the flame AI recognition model, the flame AI recognition model outputs a flame recognition result; Determining whether to output a fire warning based on the flame recognition result; The processing of the first image data to determine a first target monitoring area further includes: obtaining third image data of a second monitoring area of ​​the substation, and processing the third image data to determine a second target monitoring area; wherein the second monitoring area is adjacent to the first monitoring area; and wherein the first monitoring area and the second monitoring area correspond to different monitoring angles; Filtering a plurality of fourth target monitoring areas from the third image data; wherein the fourth target monitoring area is a second target detection area that is not located in the first monitoring area but is at a distance from the first monitoring area that is less than or equal to a first threshold; Determining a fifth target monitoring area in the first monitoring area according to the fourth target monitoring area, and taking the fifth target monitoring area as a part of the first target detection area; processing the first image data to determine the first target monitoring area includes: Performing foreground extraction on the first image data to obtain several candidate objects; Comparing the candidate objects with the pre-stored objects, and determining the target objects based on the comparison results; Determining the area where each target object is located as the first target monitoring area; determining a fifth target monitoring area in the first monitoring area according to the fourth target monitoring area, including: determining a second position in the first image data according to the first position of the fourth target monitoring area in the third image data; Determine a first equivalent value of the first monitoring area based on a first feature in the first image data, determine a second equivalent value of the second monitoring area based on a second feature in the third image data, and determine a third equivalent value based on the first equivalent value and the second equivalent value; The fifth target monitoring area is determined in the first image data according to the second position and the third equivalent value.

2. The AI-based substation fire warning method according to claim 1 is characterized in that :The flame AI recognition model is constructed by any one of back propagation, Boltzmann machine, convolutional neural network, Hopfield network, multi-layer perceptron, radial basis function network, restricted Boltzmann machine, recurrent neural network, self-organizing map, and spiking neural network.

3. The AI-based substation fire warning method according to claim 2 is characterized in that : Determining whether to output a fire warning based on the flame recognition result, including: If the flame recognition result indicates that a flame exists, then calculating the area ratio of the connected domain of the flame in the second image data, and if the area ratio is greater than or equal to a second threshold, determining to output a fire warning; If the area ratio is less than a second threshold, personnel data is extracted based on the first image data, and whether to output a fire warning is determined based on the extraction result.

4. The AI-based substation fire warning method according to claim 3 is characterized in that The step of determining whether to output a fire warning based on the extraction result includes: If the personnel data is not extracted, it is determined to output a fire warning; If the personnel data is extracted, the matching degree between the personnel data and the target personnel data is calculated. If the matching degree meets the preset conditions, no fire warning is output; otherwise, a fire warning is output.

5. An AI-based substation fire warning system, comprising an acquisition module, a processing module, and a storage module; the processing module is connected to the acquisition module and the storage module; The storage module is used to store executable computer program code; The acquisition module is used to acquire image data and transmit it to the processing module; It is characterized by : The processing module is used to execute the method according to any one of claims 1 to 4 by calling the executable computer program code in the storage module.

6. An electronic device comprising: a memory storing executable program code; a processor coupled to the memory; characterized in that : The processor calls the executable program code stored in the memory to execute the method according to any one of claims 1 to 4.

7. A computer storage medium having a computer program stored thereon, characterized in that :When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is executed.

Images