Security inspection method, security inspection device, electronic device, storage medium and program

Abstract

The present disclosure relates to the field of computer technologies, and particularly to a safety inspection method, a safety inspection device, an electronic device, a storage medium, and a program. 【Solution means】The method includes obtaining actual environmental parameters of each first inspection area among a plurality of first inspection areas in a first period; obtaining a first environmental parameter prediction result of each first inspection area in a second period based on the actual environmental parameters of each first inspection area; obtaining a plurality of second inspection areas based on the area positions in the production workplaces of each first inspection area; obtaining a second environmental parameter prediction result of each second inspection area among the plurality of second inspection areas in the second period based on the first environmental parameter prediction result of each first inspection area; and generating safety warning information corresponding to a risk area when it is determined that a risk area exists in the plurality of second inspection areas based on the second environmental parameter prediction result of each second inspection area.

Description

Technical Field

[0001] The present disclosure relates to the field of computer technology, and particularly to a safety inspection method, a safety inspection device, an electronic device, a storage medium, and a program.

Background Art

[0002] As the most core part of the spinning industry, due to the complexity of its production process and special environment, the spinning process often involves many safety risks. These safety risks not only threaten production safety but also become an important constraint on the sound, stable, and sustainable development of the industry.

Summary of the Invention

Problems to be Solved by the Invention

[0003] Therefore, at present when the spinning industry is developing rapidly, how to realize an effective safety warning for the spinning process has become an urgent technical problem in the spinning industry.

Means for Solving the Problems

[0004] The present disclosure provides a safety inspection method, a safety inspection device, an electronic device, a storage medium, and a program to solve or alleviate one or more technical problems of the prior art.

[0005] According to a first aspect of the present disclosure, a safety inspection method is provided, and the method includes: obtaining the actual environmental parameters of each first inspection area among a plurality of first inspection areas in a first period, where the plurality of first inspection areas are located in a production workplace; obtaining a predicted result of the first environmental parameters of each first inspection area among the plurality of first inspection areas in a second period based on the actual environmental parameters of each first inspection area among the plurality of first inspection areas, where the second period is a future period of the first period; Based on the area positions of each first inspection area in a plurality of first inspection areas in the production workplace, perform an inspection area expansion operation on the production workplace to obtain a plurality of second inspection areas; Based on the first environmental parameter prediction results of each first inspection area in a plurality of first inspection areas, obtain the second environmental parameter prediction results of each second inspection area in the plurality of second inspection areas in the second period; Based on the second environmental parameter prediction results of each second inspection area in the plurality of second inspection areas, when it is determined that there is a risk area in the plurality of second inspection areas, generate safety warning information corresponding to the risk area, and transmit the safety warning information to a target terminal for broadcasting the safety warning information.

[0006] According to a second aspect of the present disclosure, a safety inspection device is provided, and the device includes: A parameter acquisition unit for obtaining the actual environmental parameters of each first inspection area in a plurality of first inspection areas in the first period, where the plurality of first inspection areas are located in a production workplace; A first prediction unit for obtaining the first environmental parameter prediction results of each first inspection area in the plurality of first inspection areas in the second period based on the actual environmental parameters of each first inspection area in the plurality of first inspection areas, where the second period is a future period of the first period; An area selection unit for performing an inspection area expansion operation on the production workplace based on the area positions of each first inspection area in the plurality of first inspection areas in the production workplace to obtain a plurality of second inspection areas; A second prediction unit for obtaining the second environmental parameter prediction results of each second inspection area in the plurality of second inspection areas in the second period based on the first environmental parameter prediction results of each first inspection area in the plurality of first inspection areas; An inspection result acquisition unit for generating safety warning information corresponding to the risk area when it is determined that there is a risk area in the plurality of second inspection areas based on the second environmental parameter prediction results of each second inspection area in the plurality of second inspection areas, and transmitting the safety warning information to a target terminal for broadcasting the safety warning information.

[0007] According to a third aspect of the present disclosure, an electronic device is provided, the device comprising: at least one processor; a memory communicatively connected to the at least one processor; Instructions executable by the at least one processor are stored in the memory, and when the instructions are executed by the at least one processor, any one of the methods in the embodiments of the present disclosure is executed.

[0008] According to a fourth aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions for causing a computer to execute any one of the methods in the embodiments of the present disclosure is provided.

[0009] According to a fifth aspect of the present disclosure, a program is provided, and when the program is executed by a processor, any one of the methods in the embodiments of the present disclosure is realized.

[0010] According to the solution of the present disclosure, when there is a risk area in the production workshop, by accurately inspecting the risk area and generating safety warning information corresponding to the risk area, an effective safety warning for the spinning process can be realized.

[0011] It should be understood that the content described herein is not intended to describe the key points or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. For other features of the present disclosure, understanding is promoted through the following specification.

Brief Description of the Drawings

[0012] In the accompanying drawings, unless otherwise specified, the same reference numerals indicate the same or similar components or elements throughout the plurality of accompanying drawings. These accompanying drawings are not necessarily drawn to scale. It should be understood that these drawings only show some embodiments provided by the present disclosure and should not be regarded as limiting the scope of the present disclosure.

[0013]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 3C

Figure 3D

Figure 4

Figure 5

Figure 6

Embodiments for Carrying Out the Invention

[0014] Hereinafter, the present disclosure will be described in more detail with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar elements. Also, in the accompanying drawings, although various aspects of the embodiments are shown, these accompanying drawings are not necessarily drawn to scale unless otherwise specified.

[0015] Furthermore, in order to better explain the present disclosure, many specific details are described in the following specific embodiments. Those skilled in the art should understand that the present disclosure can be implemented similarly even without some of these details. In some embodiments, methods, means, components, and circuits well known to those skilled in the art are not described in detail so that the gist of the present disclosure can be made clear.

[0016] As described above, as the most core part of the spinning industry, due to the complexity of its production process and special environment, the spinning process often involves many safety risks. For example, it often involves safety risks such as high temperature, harmful gases, and volatile substances of chemical fiber lubricants. These safety risks not only threaten production safety but also become important constraints for the sound, stable, and sustainable development of the industry. Therefore, at present when the spinning industry is developing rapidly, how to realize effective safety warnings for the spinning process has become an urgent technical issue in the spinning industry.

[0017] To realize effective safety warnings for the spinning process, the embodiments of the present disclosure provide a safety inspection method applied to an electronic device. The electronic device can represent a server or various forms of terminal devices, such as a computer (desktop computer, notebook computer, etc.), an IoT device, or other similar computing devices.

[0018] Also, in the embodiments of the present disclosure, the main task of the spinning process is to convert the most primitive spinning raw materials into winding package products that can be used by a loom through a series of processes. Moreover, the main types of winding package products can include one or more of Partially Oriented Yarns (POY), Fully Drawn Yarns (FDY), Draw Textured Yarns (DTY) (or low-elastic filaments), Polyester Staple Fiber (PSF), etc. For example, specific types of winding package products include Polyester Partially Oriented Yarns, Polyester Fully Drawn Yarns, Polyester Drawn Yarns, Polyester Draw Textured Yarns, etc.

[0019] FIG. 1 is a flowchart of a safety inspection method according to an embodiment of the present disclosure. Hereinafter, with reference to FIG. 1, the safety inspection method provided by the embodiment of the present disclosure will be described. Note that although a logical order is shown in the flowchart, it should be noted that in some cases, the steps shown or described may be executed in a different order.

[0020] In step S101, the actual environmental parameters of each first inspection area among a plurality of first inspection areas in the first period are obtained.

[0021] Here, the first period may be the current period having a first preset time length, and the first preset time length can be set according to application requirements, and the embodiments of the present disclosure are not limited thereto. The plurality of first inspection areas are located in a production workplace. Here, the production workplace may be a common process workplace such as a boiler (or a heat medium furnace), and the production workplace may also be an overlapping workplace, a spinning workplace, a winding workplace, a short fiber pre-spinning workplace, a short fiber post-spinning workplace, etc.

[0022] In one example, the production workplace is a common process workplace, and the plurality of first inspection areas can include adjacent areas of a plurality of boilers in the common process workplace. Here, the adjacent area can be used to represent a specified position in a circumferential range centered on a target object (for example, the first boiler, the second boiler, and the third boiler among the plurality of boilers). Here, the length of the radius of the circumferential range can be set according to application requirements, and the embodiments of the present disclosure are not limited thereto.

[0023] In another example, the production workplace is a co - working workplace. In the co - working workplace, pure terephthalic acid (PTA) and ethylene glycol (EG) are placed in a reaction environment around 200 °C to carry out an esterification reaction to produce polyethylene terephthalate (PET) with a low degree of polymerization. Then, the PET with a low degree of polymerization is condensed into a high polymer in a reaction environment around 280 °C. Here, the high polymer is transported to a spinning workplace and used as the raw material for melt spinning. Alternatively, the high polymer is conveyed to a granulator, and the granulator processes the high polymer to obtain polyester chips, which are used as the raw material for chip spinning. Based on this, in the embodiments of the present disclosure, the plurality of first inspection regions can include the adjacent region of the first reaction kettle for realizing the esterification reaction in the co - working workplace, the adjacent region of the second reaction kettle for realizing the polycondensation reaction, and the adjacent region of the heat medium pipeline for transporting the high polymer.

[0024] In yet another example, the production workplace is a spinning workplace. In the spinning workplace, operations such as raw material pressure increase, melt cooling, and static mixing are performed on the raw material for spinning conveyed to the spinning workplace. The raw material for spinning that has undergone the raw material pressure increase, melt cooling, and static mixing is conveyed to a spinning box for spinning to obtain a wound package product. Then, cooling treatment, sizing treatment, oiling treatment, winding treatment, and packaging treatment are performed on the wound package product to obtain a packaged wound package product. Based on this, in the embodiments of the present disclosure, the plurality of first inspection regions can include the adjacent region of the heat medium pipeline for transporting the high polymer in the spinning workplace and the adjacent regions of a plurality of spinning boxes.

[0025] In an embodiment of the present disclosure, by having a robot perform a patrol inspection or pre-installing environmental parameter sensors in each of a plurality of first inspection areas, the actual environmental parameters of each of the plurality of first inspection areas during a first period can be collected. Further, in an embodiment of the present disclosure, when the production workplace is a common process workplace, the actual environmental parameters can include at least one of temperature and harmful gas concentration. When the production workplace is a spinning workplace, the actual environmental parameters can include at least one of temperature, noise intensity, and the concentration of chemical fiber oil agent volatile substances. Here, the harmful gas concentration is used to represent the concentration of harmful gases such as sulfur dioxide, nitrogen oxides, escaped ammonia, and dust particles. The concentration of chemical fiber oil agent volatile substances is used to represent the concentration of chemical fiber oil agent volatile substances.

[0026] In step S102, based on the actual environmental parameters of each of the plurality of first inspection areas, a first environmental parameter prediction result for each of the plurality of first inspection areas during a second period is obtained.

[0027] Here, the second period is a future period of the first period. Specifically, the second period may be a future period having a second preset time length from the first period. Here, the second preset time length may be the same as the first preset time length, or may be a different time length from the first preset time length. Specifically, it can be set according to application requirements, and the embodiments of the present disclosure are not limited thereto.

[0028] In addition, in the embodiments of the present disclosure, based on the actual environmental parameters of each first inspection area among a plurality of first inspection areas by using a preset prediction model, the first environmental parameter prediction results of each first inspection area among the plurality of first inspection areas in the second period can be obtained. Here, the first environmental parameter prediction results represent the same data characteristics as the actual environmental parameters. For example, when the actual environmental parameters include temperature and harmful gas concentration, the first environmental parameter prediction results also include temperature and harmful gas concentration. Also, for example, when the actual environmental parameters include temperature, noise intensity, and chemical fiber oil agent volatile substance concentration, the first environmental parameter prediction results also include temperature, noise intensity, and chemical fiber oil agent volatile substance concentration.

[0029] In step S103, based on the area positions in the production workplaces of each first inspection area among the plurality of first inspection areas, an inspection area expansion operation for the production workplace is performed to obtain a plurality of second inspection areas.

[0030] Here, the plurality of second inspection areas can include the plurality of first inspection areas and M risk diffusion areas correlated with the plurality of first inspection areas, or the plurality of second inspection areas can include the plurality of first inspection areas and at least one risk diffusion integration area obtained based on the M risk diffusion areas. Here, M≥2 and M is an integer.

[0031] In step S104, based on the first environmental parameter prediction results of each first inspection area among the plurality of first inspection areas, the second environmental parameter prediction results of each second inspection area among the plurality of second inspection areas in the second period are obtained.

[0032] Here, the second environmental parameter prediction results represent the same data characteristics as the first environmental parameter prediction results. For example, when the first environmental parameter prediction results include temperature and harmful gas concentration, the second environmental parameter prediction results also include temperature and harmful gas concentration. Also, for example, when the first environmental parameter prediction results include temperature, noise intensity, and chemical fiber oil agent volatile substance concentration, the second environmental parameter prediction results also include temperature, noise intensity, and chemical fiber oil agent volatile substance concentration.

[0033] In step S105, when it is determined that there is a risk area in the plurality of second inspection areas based on the predicted results of the second environmental parameters of each second inspection area in the plurality of second inspection areas, safety warning information corresponding to the risk area is generated, and the safety warning information is transmitted to the target terminal.

[0034] Here, the safety warning information may be voice information, may be graphic information, or may be audio-video information generated by combining voice information and graphic information. The embodiments of the present disclosure are not limited thereto. The target terminal is used to broadcast the safety warning information.

[0035] According to the safety inspection method provided by the embodiments of the present disclosure, the actual environmental parameters of each first inspection area among a plurality of first inspection areas in the first period can be obtained, and based on the actual environmental parameters of each first inspection area among the plurality of first inspection areas, the prediction results of the first environmental parameters of each first inspection area among the plurality of first inspection areas in the second period can be obtained. After obtaining the prediction results of the first environmental parameters of each first inspection area among the plurality of first inspection areas in the second period, instead of directly determining whether there is a risk area in the production workplace based on the prediction results of the first environmental parameters, based on the area positions of each first inspection area in the production workplace among the plurality of first inspection areas, an inspection area expansion operation for the production workplace is executed to obtain a plurality of second inspection areas, and based on the prediction results of the first environmental parameters of each first inspection area among the plurality of first inspection areas, the prediction results of the second environmental parameters of each second inspection area among the plurality of second inspection areas in the second period are obtained, and based on the prediction results of the second environmental parameters of each second inspection area among the plurality of second inspection areas, it is determined whether there is a risk area in the plurality of second inspection areas. The final risk area is determined with the plurality of second inspection areas as the candidate dataset. Since the plurality of second inspection areas are obtained by executing an inspection area expansion operation for the production workplace based on the area positions of each first inspection area in the production workplace among the plurality of first inspection areas, it can be ensured that the candidate dataset has a high area coverage rate for the production workplace. In this way, when there is a risk area in the production workplace, the risk area can be accurately inspected, and by generating safety warning information corresponding to the risk area, an effective safety warning for the spinning process can be realized.

[0036] Also, in the embodiments of the present disclosure, the preset prediction model may be a trained time series model. Based on this, in some selectable embodiments, step S102 may include the following steps.

[0037] In step S102-1, each first inspection area among the plurality of first inspection areas is set as a first processing target area, and a first input sequence is constructed based on the actual environment parameters of the first processing target area.

[0038] In one example, the first input sequence can be constructed as follows.

[0039] (1) Analyze the actual environment parameters of the first processing target area to obtain a plurality of actual environment sub-parameters.

[0040] For example, when the production workplace is a common process workplace, by analyzing the actual environment parameters of the first processing target area, the plurality of obtained actual environment sub-parameters can include temperature and harmful gas concentration. Also, for example, when the production workplace is a spinning workplace, by analyzing the actual environment parameters of the first processing target area, the plurality of obtained actual environment sub-parameters can include temperature, noise intensity, and the concentration of chemical fiber oil agent volatile substances. (2) Set each actual environment sub-parameter among the plurality of actual environment sub-parameters as a first parameter sequence, and add parameter attribute information and production task information to the first parameter sequence to obtain the corresponding first input sequence of the first parameter sequence.

[0041] Here, when the production workplace is a common process workplace, the parameter attribute information is used to represent whether the actual environment sub-parameter represented by the first parameter sequence is temperature or harmful gas concentration, and the production task information can include the fuel component used for the boiler in the common process workplace, the temperature of the high-temperature steam to be created, etc. Also, when the production workplace is a spinning workplace, the parameter attribute information is used to represent whether the first parameter sequence is temperature, noise intensity, or the concentration of chemical fiber oil agent volatile substances, and the production task information can include the product type, product specifications, etc. of the spinning products that are the production targets in the production workplace.

[0042] In one example, the first input sequence can be denoted as (A, X11, X12 ··· X1n, B1, B2 ···). Here, A is parameter attribute information, (X11, X12 ··· X1n) is the first parameter sequence, and X11 is the actual environment sub-parameter value of the first processing target area at time T11 in the first period, X12 is the actual environment sub-parameter value of the first processing target area at time T12 in the first period, X1n is the actual environment sub-parameter value of the first processing target area at time T1n in the first period, and B1, B2, etc. are production task information of the production workshop. Here, n ≥ 20 and n is an integer.

[0043] In step S102-2, based on the actual environment parameters of the first processing target area and the preset input sequence, a second input sequence is constructed.

[0044] In one example, the second input sequence can be constructed as follows.

[0045] (1) Analyze the actual environment parameters of the first processing target area to obtain a plurality of actual environment sub-parameters.

[0046] For example, when the production workshop is a common process workshop, by analyzing the actual environment parameters of the first processing target area, the plurality of obtained actual environment sub-parameters can include temperature and harmful gas concentration. Also, for example, when the production workshop is a spinning workshop, by analyzing the actual environment parameters of the first processing target area, the plurality of obtained actual environment sub-parameters can include temperature, noise intensity, and chemical fiber oil agent volatile substance concentration.

[0047] (2) Take each actual environment sub-parameter among the plurality of actual environment sub-parameters as the second parameter sequence, and cut out at least a part of the end sequence from the second parameter sequence to serve as the additional target sequence.

[0048] In one example, the second parameter sequence can be denoted as (X11, X12 ··· X1n). Here, X11 is the actual environmental sub-parameter value of the first processing target area at time T11 in the first period, X12 is the actual environmental sub-parameter value of the first processing target area at time T12 in the first period, and X1n is the actual environmental sub-parameter value of the first processing target area at time T1n in the first period. At least a part of the end sequence cut from the second parameter sequence, that is, the sequence to be added, can be denoted as (X1n-9, X1n-8 ··· X1n). Here, X1n-9 is the actual environmental sub-parameter value of the first processing target area at time T1n-9 in the first period, X1n-8 is the actual environmental sub-parameter value of the first processing target area at time T1n-8 in the first period, and X1n is the actual environmental sub-parameter value of the first processing target area at time T1n in the first period.

[0049] (2) Add the parameter attribute information and the production task information to the sequence to be added to obtain an intermediate input sequence corresponding to the second parameter sequence.

[0050] As described above, when the production workplace is a common process workplace, the parameter attribute information is used to represent whether the actual environmental sub-parameters represented by the second parameter sequence are temperature or harmful gas concentration, and the production task information can include the fuel components used in the boiler in the common process workplace and the temperature of the high-temperature steam to be created. Also, when the production workplace is a spinning workplace, the parameter attribute information is used to represent whether the second parameter sequence is temperature, noise intensity, or the concentration of chemical fiber oil volatiles, and the production task information can include the product type and product specifications of the spinning products to be produced in the production workplace.

[0051] In one example, the intermediate input sequence can be denoted as (A, X1n-9, X1n-8 ··· X1n, B1, B2 ···). Here, A is parameter attribute information, (X1n-9, X1n-8 ··· X1n) is the sequence to be added, and X1n-9 is the actual environment sub-parameter value of the first processing target area at time T1n-9 in the first period, X1n-8 is the actual environment sub-parameter value of the first processing target area at time T1n-8 in the first period, X1n is the actual environment sub-parameter value of the first processing target area at time T1n in the first period, and B1, B2, etc. are production task information of the production workplace.

[0052] (3) Add the preset input sequence to the intermediate input sequence to obtain the second input sequence corresponding to the second parameter sequence.

[0053] Here, each numerical value in the preset input sequence can be set to the same initial value, the sequence length of the preset input sequence can be set according to the application requirements, and the embodiments of the present disclosure are not limited thereto.

[0054] In one example, the second input sequence can be denoted as (A, X1n-9, X1n-8 ··· X1n, B1, B2 ··· C1, C2 ··· Cm). Here, (A, X1n-9, X1n-8 ··· X1n, B1, B2 ···) is the intermediate input sequence corresponding to the second parameter sequence, and A is parameter attribute information. X1n-9 is the actual environment sub-parameter value of the first processing target area at time T1n-9 in the first period, X1n-8 is the actual environment sub-parameter value of the first processing target area at time T1n-8 in the first period, X1n is the actual environment sub-parameter value of the first processing target area at time T1n in the first period, B1 and B2, etc. are production task information of the production workplace, and (C1, C2 ··· Cm) is the preset input sequence. Here, m ≧ 20 and m is an integer.

[0055] In step S102-3, the first input sequence and the second input sequence are input into the trained time series model to obtain the overall output sequence of the time series model.

[0056] Here, the trained time series model may be an Informer model, an Autoregressive Integrated Moving Average Model, or the like.

[0057] In one example, the trained time series model can be obtained as follows.

[0058] Obtain the first actual environmental parameters of each target inspection area among a plurality of target inspection areas selected from the production workplace during the first training period. Each target inspection area among the plurality of target inspection areas is used as a model training area. Based on the actual environmental parameters of the model training area, construct a first training input sequence. Based on the first actual environmental parameters of the model training area and a preset training input sequence, construct a second training input sequence. Input the first input training sequence and the second input training sequence into the initial time series model to obtain the overall training output sequence of the initial time series model. Based on the overall training output sequence, obtain the environmental parameter prediction result of the model training area during the second training period. Based on the environmental parameter prediction result and the second actual environmental parameters of the model training area during the second training period, perform model parameter adjustment on the initial time series model. As described above, until the initial time series model meets the convergence condition, perform model parameter adjustment on the initial time series model multiple times to make the initial time series model a trained time series model.

[0059] Regarding the above method, it can be specifically understood by referring to steps S101 and S102 (including steps S102-1, S102-2, and S102-3), and will not be repeated here.

[0060] In addition, in the embodiments of the present disclosure, when executing step S102-3, it can be understood that it is necessary to input a first input sequence and a second input sequence having the same parameter attribute information into a trained time series model to obtain the overall output sequence of the time series model.

[0061] In one example, there are a first input sequence D1 and a first input sequence D2, as well as a second input sequence E1 and a second input sequence E2.

[0062] Here, the first input sequence D1 and the second input sequence E1 have the same parameter attribute information. Specifically, the parameter attribute information is used to represent that the actual environmental sub-parameters in the first input sequence D1 and the second input sequence E1 are temperature. The first input sequence D2 and the second input sequence E2 have the same parameter attribute information. Specifically, the parameter attribute information is used to represent that the actual environmental sub-parameters in the first input sequence D2 and the second input sequence E2 are harmful gas concentration.

[0063] Then, when executing step S102-3, input the first input sequence D1 and the second input sequence E1 into the trained time series model to obtain the overall output sequence of the time series model, and the overall output sequence can be correlated with the first temperature prediction value in the first environmental parameter prediction result of the first processing target area in the second period. Also, when executing step S102-3, input the first input sequence D2 and the second input sequence E2 into the trained time series model to obtain the overall output sequence of the time series model, and the overall output sequence can be correlated with the first harmful gas concentration prediction value in the first environmental parameter prediction result of the first processing target area in the second period.

[0064] In the embodiments of the present disclosure, through the above steps included in step S102, each first inspection area among a plurality of first inspection areas is used as a first processing target area, a first input sequence is constructed based on the actual environment parameters of the first processing target area, and a second input sequence is constructed based on the actual environment parameters of the first processing target area and a preset input sequence. The first input sequence and the second input sequence are input into a trained time series model to obtain an overall output sequence of the time series model. Based on the overall output sequence, a prediction result of the first environmental parameter of the first processing target area in the second period is obtained. Since the construction of both the first input sequence and the second input sequence depends on the actual environment parameters of the first processing target area, and the actual environment parameters of the first processing target area are used to reflect the real-time environment of the production workplace, positive guidance can be provided for the data processing of the time series model, thereby improving the accuracy of the prediction result of the first environmental parameter.

[0065] Referring to FIG. 2, in one example, the trained time series model includes an encoder and a decoder. Based on this, in this example, the overall output sequence of the time series model can be obtained as follows.

[0066] (1) Input the first input sequence into the encoder, and process the first input sequence using the first self-attention module and the distillation module in the encoder to obtain a first input feature mapping result.

[0067] Here, the encoder can include a plurality of serially connected feature encoder structures, and each feature encoder structure can include a serially connected first self-attention module and distillation module.

[0068] In addition, in the embodiments of the present disclosure, the first self-attention module uses a ProbSparse sparse self-attention mechanism to perform self-attention calculation on the input sequence to obtain an intermediate sequence, and inputs the intermediate sequence into a distillation module belonging to the same encoder structure as the first self-attention module. By using the distillation module to perform distillation processing on the intermediate sequence to reduce the complexity of the output sequence, the distilled intermediate sequence is used as the output sequence of this feature encoder structure.

[0069] In the embodiments of the present disclosure, it can be understood that among the plurality of serially connected feature encoder structures, the output sequence of the last feature encoder structure is the first input feature mapping result.

[0070] (2) Input the second input sequence into the second self-attention module in the encoder, and use the second self-attention module to process the second input sequence to obtain a second input feature mapping result.

[0071] Here, the second self-attention module uses a ProbSparse sparse self-attention mechanism and a mask mechanism to perform self-attention calculation on the second input sequence and is used to obtain the second input feature mapping result.

[0072] (3) Input the first input feature mapping result and the second input feature mapping result into the mutual attention module in the encoder, and use the mutual attention module to process the first input feature mapping result and the second input feature mapping result to obtain the overall output sequence of the time series model.

[0073] In step S102-4, based on the overall output sequence, obtain the first environmental parameter prediction result of the first processing target area in the second period.

[0074] In one specific example, from the overall output sequence, a sequence portion corresponding to a preset input sequence can be determined as the prediction result of the first environmental parameter of the first processing target area in the second period.

[0075] For example, when the overall output sequence correlates with the predicted value of the first temperature in the prediction result of the first environmental parameter of the first processing target area in the second period, the prediction result of the first environmental parameter is the predicted value of the first temperature. Also, for example, when the overall output sequence is related to the predicted value of the first harmful gas concentration in the prediction result of the first environmental parameter of the first processing target area in the second period, the prediction result of the first environmental parameter is the predicted value of the first harmful gas concentration.

[0076] In the above example, since the trained time-series model has a simple model structure, by using the trained time-series model, the prediction result of the first environmental parameter of the first processing target area in the second period can be obtained quickly, and the execution efficiency of the safety inspection method can be improved.

[0077] Furthermore, as described above, the plurality of second inspection areas can include a plurality of first inspection areas, M risk diffusion areas correlated with the plurality of first inspection areas, or at least one risk diffusion integration area obtained based on the plurality of first inspection areas and the M risk diffusion areas. Based on this, in some selectable embodiments, step S103 can include the following steps.

[0078] In step S103-1, each first inspection area among the plurality of first inspection areas is set as the first processing target area, and at least one risk diffusion area correlated with the first processing target area is determined from the production workplace to obtain M risk diffusion areas correlated with the plurality of first inspection areas.

[0079] In one example, at least one risk diffusion area correlated with the first processing target area can be determined as follows.

[0080] (1) Obtain the workplace structure model of the production workplace.

[0081] Here, the workplace structure model can be a three-dimensional software model constructed using SketchUp, Revit, 3dsMax, Cedreo, AutoCAD, etc. based on the workplace structure and internal composition of the production workplace. Here, the workplace structure can include the workplace shape, workplace dimensions, etc., and the internal composition can include the equipment arranged in the production workplace. Based on this, in this example, the production workplace structure model can have production workplace structure information for representing the production workplace structure of the production workplace and internal composition information for representing the internal composition of the production workplace.

[0082] Here, the equipment arranged in the production workplace can include production equipment, auxiliary equipment, etc.

[0083] For example, when the production workplace is a common process workplace, the production equipment can include a plurality of boilers, and the auxiliary equipment can include exhaust equipment, air conditioning equipment, etc. Or, for example, when the production workplace is a polymerization workplace, the production equipment can include a first reaction kettle for realizing an esterification reaction, a second reaction kettle for realizing a polycondensation reaction, and a heat medium pipeline for transporting a high polymer, and the auxiliary equipment can include exhaust equipment, air conditioning equipment, etc. Or, for example, when the production workplace is a spinning workplace, the production equipment can include a heat medium pipeline for transporting a high polymer and a plurality of spinning boxes, and the auxiliary equipment can include exhaust equipment, air conditioning equipment, etc.

[0084] (2) Based on the workplace structure information and the internal composition information, determine a candidate diffusible region correlated with the first processing target region from the production workplace.

[0085] Here, the candidate diffusible region correlated with the first processing target region may be all regions in the production workplace where safety impact factors (such as heat, harmful gases, noise, volatile substances of chemical fiber lubricants, etc.) can diffuse from the first processing target region. Further, in this example, the candidate diffusible region correlated with the processing target region may be all vacant regions in the production workplace.

[0086] Referring to FIGS. 3A, 3B, and 3C, the production workplace is a common process workplace 300, and the production equipment includes a first boiler 301, a second boiler 302, and a third boiler 303.

[0087] In this case, the plurality of first inspection areas can include the adjacent area of the first boiler 301, the adjacent area of the second boiler 302, and the adjacent area of the third boiler 303. When the adjacent area of the first boiler 301 is taken as the first processing target area F1, based on the workplace structure information and the internal configuration information, all areas within the production workplace 300 where the safety impact factor can diffuse from the first processing target area F1 (specifically, all vacant areas in the production workplace 300) can be determined as the candidate diffusible areas G1 correlated with the first processing target area F1 (for example, the hatched area shown in FIG. 3A). When the adjacent area of the second boiler 302 is taken as the first processing target area F2, based on the workplace structure information and the internal configuration information, all areas within the production workplace 300 where the safety impact factor can diffuse from the first processing area F2 (specifically, all vacant areas in the production workplace 300) can be determined as the candidate diffusible areas G2 correlated with the first processing target area F2 (for example, the hatched area shown in FIG. 3B). When the adjacent area of the third boiler 303 is taken as the first processing target area F3, based on the workplace structure information and the internal configuration information, all areas within the production workplace 300 where the safety impact factor can diffuse from the first processing target area F3 (specifically, all vacant areas in the production workplace 300) can be determined as the candidate diffusible areas G3 correlated with the first processing target area F3 (for example, the hatched area shown in FIG. 3C).

[0088] (3) Based on the internal configuration information and the first processing target area, at least one risk diffusion area correlated with the first processing target area is determined from among the candidate diffusible areas.

[0089] Here, at least one risk diffusion area correlated with the first processing target area may be an area among the candidate diffusible areas where the safety impact factor can diffuse from the first processing target area at the fastest speed and in the largest amount. Also, usually, at least one risk diffusion area correlated with the first processing target area is affected by at least the exhaust equipment installed in the production workplace.

[0090] Based on this, in one specific example, based on the internal configuration information and the first processing target area, an area in the candidate diffusible areas where the safety impact factor can diffuse from the first processing target area at the fastest speed and in the largest amount can be determined as at least one risk diffusion area correlated with the first processing target area.

[0091] Specifically, based on the internal configuration information, the shape, dimensions, and installation position of the exhaust equipment in the production workplace can be obtained. Based on the shape, dimensions, and installation position of the exhaust equipment, the exhaust area in the production workplace can be determined. Based on the exhaust area and the first processing target area, an area in the candidate diffusible areas where the safety impact factor can diffuse from the first processing target area at the fastest speed and in the largest amount is determined as at least one risk diffusion area correlated with the first processing target area.

[0092] More specifically, in the candidate diffusible area, a reference circumference centered on the first processing target area can be drawn, and a plurality of reference positions can be marked according to a preset circumferential distance on the reference circumference. Based on the internal configuration information, the shape, dimensions, and installation position of the exhaust equipment in the production workplace are obtained. Based on the shape, dimensions, and installation position of the exhaust equipment, the exhaust area in the production workplace is determined. From among the plurality of reference positions, the area position located between the exhaust area and the first processing target area is selected as at least one risk diffusion area correlated with the first processing target area. Here, the length of the radius of the reference circumference and the length of the preset circumferential distance can be set according to the application requirements, and the embodiments of the present disclosure are not limited thereto.

[0093] Referring to FIG. 3A, when the adjacent area of the first boiler 301 is taken as the first processing target area F1, there is a candidate diffusion possible area G1 correlated with the first processing target area F1. At this time, in the candidate diffusion possible area G1, a reference circumference centered on the first processing target area F1 can be drawn, and a plurality of reference positions can be marked according to a preset circumferential distance on the reference circumference (not fully illustrated in FIG. 3A). Based on the internal configuration information, the shape, dimensions, and installation position of the exhaust equipment 304 in the production workshop are obtained. Based on the shape, dimensions, and installation position of the exhaust equipment 304, the exhaust area Q in the production workshop is determined. From among the plurality of reference positions, the area position located between the exhaust area Q and the first processing target area F1 is selected as at least one risk diffusion area correlated with the first processing target area G1, that is, risk diffusion area H1, risk diffusion area H2, and risk diffusion area H3.

[0094] Referring to FIG. 3B, when the adjacent area of the second boiler 302 is taken as the first processing target area F2, there is a candidate diffusion possible area G2 correlated with the first processing target area F2. At this time, within the candidate diffusion possible area G2, a reference circumference centered on the first processing target area F2 can be drawn, and a plurality of reference positions can be marked according to a preset circumferential distance on the reference circumference (not fully illustrated in FIG. 3B). Based on the internal configuration information, the shape, dimensions, and installation position of the exhaust equipment 304 in the production workshop are obtained. Based on the shape, dimensions, and installation position of the exhaust equipment 304, the exhaust area Q in the production workshop is determined. From among the plurality of reference positions, the area position located between the exhaust area Q and the first processing target area F2 is selected as at least one risk diffusion area correlated with the first processing target area G2, that is, risk diffusion area H4, risk diffusion area H5, risk diffusion area H6, and risk diffusion area H7.

[0095] Referring to FIG. 3C, when the adjacent area of the third boiler 303 is taken as the first processing target area F3, there is a candidate diffusion possible area G3 correlated with the first processing target area F3. At this time, within the candidate diffusion possible area G3, a reference circumference centered on the first processing target area F3 can be drawn, and a plurality of reference positions (not fully shown in FIG. 3C) can be marked according to a preset circumferential distance on the reference circumference. Based on the internal configuration information, the shape, dimensions, and installation position of the exhaust equipment 304 in the production workplace are obtained, and based on the shape, dimensions, and installation position of the exhaust equipment 304, the exhaust area Q in the production workplace is determined. From among the plurality of reference positions, the area position located between the exhaust area Q and the first processing target area F3 is selected as at least one risk diffusion area correlated with the first processing target area G3, that is, risk diffusion area H8, risk diffusion area H9, and risk diffusion area H10.

[0096] In step S103-2, based on the plurality of first inspection areas and the M risk diffusion areas, a plurality of second inspection areas are obtained.

[0097] In one example, the plurality of first inspection areas and the M risk diffusion areas can both be used as the plurality of second inspection areas, or the plurality of first inspection areas and at least one risk diffusion integration area obtained based on the M risk diffusion areas can also be used as the plurality of second inspection areas.

[0098] Through the above steps included in step S103, in the embodiment of the present disclosure, each first inspection area among the plurality of first inspection areas is taken as the first processing target area, at least one risk diffusion area correlated with the first processing target area is determined from the production workplace, M risk diffusion areas correlated with the plurality of first inspection areas are obtained, and based on the plurality of first inspection areas and the M risk diffusion areas, a plurality of second inspection areas can be obtained. In this way, the expansion of the inspection area for the production workplace can be realized, not only ensuring that the candidate dataset (the plurality of second inspection areas) has a high coverage rate for the production workplace, but also ensuring the reliability of the candidate dataset.

[0099] Furthermore, when using a plurality of first inspection regions and at least one risk diffusion integration region obtained based on M risk diffusion regions together as a plurality of second inspection regions, the plurality of second inspection regions can be obtained as follows.

[0100] (1) Based on the M risk diffusion regions, obtain at least one group of integratable regions.

[0101] Here, each group of integratable regions in the at least one group of integratable regions includes N integratable regions out of the M risk diffusion regions. Here, 2 ≤ N ≤ M, and M and N are integers.

[0102] In one example, for each group of integratable regions in the at least one group of integratable regions, the first interval distance between any two integratable regions is less than or equal to a preset distance threshold. Here, the preset distance threshold can be set according to the application requirements, and the embodiments of the present disclosure are not limited thereto.

[0103] (2) For each integratable region in the at least one group of integratable regions, perform region integration to obtain at least one risk diffusion integration region.

[0104] Here, the at least one risk diffusion integration region corresponds one-to-one with the at least one group of integratable regions.

[0105] In one example, the region integration for the group of integratable regions can integrate the N integratable regions included in the group of integratable regions to obtain a risk diffusion integration region.

[0106] (3) Use the plurality of first inspection regions and the at least one risk diffusion integration region together as the plurality of second inspection regions.

[0107] Referring to FIGS. 3A, 3B, and 3C, there are 10 risk diffusion regions correlated with a plurality of first inspection regions. The 10 risk diffusion regions are respectively risk diffusion region H1, risk diffusion region H2, risk diffusion region H3, risk diffusion region H4, risk diffusion region H5, risk diffusion region H6, risk diffusion region H7, risk diffusion region H8, risk diffusion region H9, and risk diffusion region H10. Here, since the first interval distance between risk diffusion region H3 and risk diffusion region H4 is equal to or less than a preset distance threshold, risk diffusion region H3 and risk diffusion region H4 belong to one integratable region group. Since the first interval distance between risk diffusion region H7 and risk diffusion region H8 is equal to or less than a preset distance threshold, risk diffusion region H9 and risk diffusion region H8 belong to one integratable region group. Further referring to FIG. 3D, by integrating risk diffusion region H3 and risk diffusion region H4, a risk diffusion integrated region H34 can be obtained. By integrating risk diffusion region H7 and risk diffusion region H8, a risk diffusion integrated region H78 can be obtained.

[0108] Finally, a plurality of first inspection regions (the first adjacent region of the first boiler 301, the adjacent region of the second boiler 302, and the adjacent region of the third boiler 303), the risk diffusion integrated region H34, and the risk diffusion integrated region H78 can be used as a plurality of second inspection regions together.

[0109] In the above manner, based on M risk diffusion regions, at least one integratable region group can be obtained, and region integration is performed on each integratable region in at least one integratable region group to obtain at least one risk diffusion integrated region. Further, a plurality of first inspection regions and at least one risk diffusion integrated region can be used as a plurality of second inspection regions together. In this way, the redundancy of the second inspection region can be avoided, thereby reducing the data processing amount of the safety inspection method and further improving the execution efficiency of the safety inspection method.

[0110] In some alternative embodiments, step S104 may include the following steps.

[0111] In step S104-1, each risk diffusion integration region in the plurality of second inspection regions is set as a second processing target region, and at least one reference region correlated with the second processing target region is determined from among the plurality of first inspection regions.

[0112] Referring to FIG. 3D, when the risk diffusion integration region H34 is set as the second processing target region, at least one reference region correlated with the second processing target region determined from among the plurality of first inspection regions includes the neighboring region of the first boiler 301 (i.e., the first processing target region F1) and the neighboring region of the second boiler 302 (i.e., the first processing target region F2). When the risk diffusion integration region H89 is set as the second processing target region, at least one reference region correlated with the second processing target region determined from among the plurality of first inspection regions includes the neighboring region of the second boiler 302 (i.e., the first processing target region F2) and the neighboring region of the third boiler 303 (i.e., the first processing target region F3).

[0113] In step S104-2, based on the first environmental parameter prediction results of at least one reference region, the second environmental parameter prediction results of the second processing target region in the second period are obtained.

[0114] In one example, the second environmental parameter prediction results of the second processing target region in the second period can be obtained as follows.

[0115] (1) Each reference region in at least one reference region is set as a target reference region, and a second interval distance between the target reference region and the second processing target region is obtained.

[0116] (2) A parameter weight correlated with the second interval distance is set.

[0117] In one specific example, according to the parameter setting rule, a parameter weight having a negative correlation with the second interval distance can be set. Here, the negative correlation with the second interval distance can be understood as meaning that the larger the second interval distance, the smaller the parameter weight, and conversely, the smaller the second interval distance, the larger the parameter weight.

[0118] In another specific example, according to the first parameter setting rule, a first weight having a negative correlation with the second interval distance is set, and according to the second parameter setting rule, a second weight having a negative correlation with the reference angle is set, and the product of the first weight and the second weight can be used as a parameter weight correlated with the second interval distance. Here, the negative correlation with the second interval distance means that the larger the second interval distance, the smaller the first weight, and conversely, the smaller the second interval distance, the larger the first weight. The negative correlation with the reference angle means that the larger the reference angle, the smaller the second weight, and conversely, the smaller the reference angle, the larger the second weight. Here, the reference angle may be the angle between the first reference line and the second reference line. The first reference line may be the connection line between the center point of the target reference area and the center point of the exhaust area, and the second reference line may be the connection line between the center point of the target reference area and the center point of the second processing target area.

[0119] (3) Based on the predicted result of the first environmental parameter of the target reference area and the weight parameter, obtain the predicted result of the addable environmental parameter of the second processing target area in the second period.

[0120] In one specific example, the predicted result of the first environmental parameter of the target reference area is divided to obtain at least one predicted value of the first environmental parameter, and each predicted value of the first environmental parameter among the at least one predicted value of the first environmental parameter is multiplied by the weight parameter respectively to obtain at least one predicted value of the first addable environmental parameter corresponding one-to-one to the at least one predicted value of the first environmental parameter. Furthermore, the at least one predicted value of the first addable environmental parameter can be used as the predicted result of the addable environmental parameter of the second processing target area in the second period.

[0121] (4) Based on at least one predicted result of the addable environmental parameter corresponding one-to-one to at least one reference area, obtain the predicted result of the second environmental parameter of the second processing target area in the second period.

[0122] In one specific example, for each addable environmental parameter prediction result among at least one addable environmental parameter prediction result, a division is performed to obtain at least one first addable environmental parameter prediction value. All the first addable environmental parameter prediction values having the same parameter attribute are respectively added to obtain at least one second environmental parameter prediction value that corresponds one-to-one with at least one first addable environmental parameter prediction value. Further, at least one first addable environmental parameter prediction value and at least one second environmental parameter prediction value that corresponds one-to-one with it can be used together as the second environmental parameter prediction result of the second processing target area in the second period.

[0123] Referring to FIG. 3D, when the risk diffusion integration area H34 is the second processing target area, at least one reference area correlated with the second processing target area includes the adjacent area of the first boiler 301 (i.e., the first processing target area F1) and the adjacent area of the second boiler 302 (i.e., the first processing target area F2). Here, the adjacent area of the first boiler 301 (i.e., the first processing target area F1) is the target reference area I1, and the adjacent area of the second boiler 302 (i.e., the first processing target area F2) is the target reference area I2.

[0124] Suppose the production workplace is a common process workplace. For the target reference area I1, its first environmental parameter prediction result includes a first temperature prediction value x1 and a first harmful gas concentration prediction value y1. The second interval distance between the target reference area I1 and the second processing target area is h1, and the parameter weight correlated with the second interval distance h1 is z1. Then, based on the first environmental parameter prediction result (the first temperature prediction value x1 and the first harmful gas concentration prediction value y1) of the target reference area I1 and the weight parameter z1, the obtained first addable environmental parameter prediction result of the second processing target area in the second period is First addable temperature prediction value: x1 × z1, First addable harmful gas concentration value: y1 × z1, and can include them.

[0125] Similarly, for the target reference area I2, the first environmental parameter prediction result thereof includes a first temperature prediction value x2 and a first harmful gas concentration prediction value y2. The second interval distance between the target reference area I2 and the second processing target area is h2, and the parameter weight correlated with the second interval distance h2 is z2. Then, based on the first environmental parameter prediction result (the first temperature prediction value x2 and the first harmful gas concentration prediction value y2) of the target reference area I2 and the weight parameter z2, the second addable environmental parameter prediction result of the second processing target area in the obtained second period is Second addable temperature prediction value: x2 × z2, Second additive harmful gas concentration value: y2 × z2, can be included.

[0126] Based on the first addable environmental parameter prediction result and the second addable environmental parameter prediction result, the second environmental parameter prediction result of the second processing target area in the obtained second period is Second temperature prediction value: x1 × z1 + x2 × z2, Second harmful gas concentration prediction value: y1 × z1 + y2 × z2, can be included.

[0127] Through the above steps included in step S104, in the embodiments of the present disclosure, each risk diffusion integration area among the plurality of second inspection areas is used as the second processing target area, at least one reference area correlated with the second processing target area is determined from among the plurality of first inspection areas, and based on the first environmental parameter prediction result of the at least one reference area, the second environmental parameter prediction result of the second processing target area in the second period can be obtained. Thereby, the accuracy of the second environmental parameter prediction result can be ensured.

[0128] It should be understood that in the embodiments of the present disclosure, for each first inspection area among the plurality of second inspection areas, the first environmental parameter prediction result of the first inspection area is not directly related to the second environmental parameter prediction result of the second inspection area after the first inspection area is used as the second inspection area, and this will not be repeatedly described here.

[0129] Furthermore, in the embodiments of the present disclosure, it is possible to determine that there is a risk area among a plurality of second inspection areas as follows.

[0130] (1) Each second inspection area among the plurality of second inspection areas is set as a third processing target area, and at least one risk evaluation parameter is obtained based on the predicted result of the second environmental parameter of the third processing target area.

[0131] In one example, when the production workplace is a common process workplace, at least one risk evaluation parameter obtained based on the predicted result of the second environmental parameter of the third processing target area can include temperature and harmful gas concentration. When the production workplace is a spinning workplace, at least one risk evaluation parameter obtained based on the predicted result of the second environmental parameter of the third processing target area can include temperature, noise intensity, and the concentration of chemical fiber lubricant volatile substances.

[0132] (2) Obtain a risk parameter threshold corresponding to each risk evaluation parameter among the at least one risk evaluation parameter.

[0133] (3) Based on the at least one risk evaluation parameter and the risk parameter threshold corresponding to each risk evaluation parameter among the at least one risk evaluation parameter, if the third processing target area is determined to be a risk area, it is confirmed that there is a risk area among the plurality of second inspection areas.

[0134] In one example, each risk evaluation parameter among the at least one risk evaluation parameter can be set as a target evaluation parameter, and a target risk parameter threshold corresponding to the target evaluation parameter can be obtained. When the parameter value of the target evaluation parameter is equal to or greater than the target risk parameter threshold, the third processing target area is determined to be a risk area, and it is confirmed that there is a risk area among the plurality of second inspection areas.

[0135] As described above, in the embodiments of the present disclosure, each second inspection area among a plurality of second inspection areas is used as a third processing target area, and based on the predicted result of the second environmental parameter of the third processing target area, after obtaining at least one risk evaluation parameter, a corresponding risk parameter threshold is set for each risk evaluation parameter in the at least one risk evaluation parameter, and a classification determination is performed. Thereby, when there is a risk area in the production workplace, the risk area can be inspected more accurately.

[0136] In some alternative embodiments, "generating safety warning information corresponding to the risk area" in step S105 may include the following steps.

[0137] In step S105-1, based on the internal structure of the production workplace, an initial workplace image is constructed.

[0138] Here, the initial workplace image may be a two-dimensional planar image or a three-dimensional solid model, and the embodiments of the present disclosure are not limited thereto.

[0139] In step S105-2, in the initial workplace image, the target position corresponding to the risk area is identified.

[0140] In step S105-3, in the initial workplace image, a special display is performed on the target position to obtain a target workplace image, and based on the target workplace image, safety warning information is obtained.

[0141] Note that the special display may be performed in a prominent color or with a danger icon, and the embodiments of the present disclosure are not limited thereto.

[0142] After obtaining the target workplace image, the target workplace image may be used as safety warning information, or voice information and the target workplace image may be combined to be used as safety warning information.

[0143] In the embodiments of the present disclosure, through the above steps included in step S105, an initial workplace image is constructed based on the internal structure of the production workplace. In the initial workplace image, the target positions corresponding to the risk areas are determined. Further, in the initial workplace image, special display is performed on the target positions to obtain a target workplace image, and the target workplace image is used as safety warning information, thereby enhancing the presentation effect of the safety warning information.

[0144] Furthermore, referring to FIG. 4, FIG. 4 is a schematic diagram showing an application scenario of the safety inspection method according to the embodiments of the present disclosure.

[0145] As described above, the embodiments of the present disclosure provide a safety inspection method applicable to an electronic device. The electronic device can represent a server or various forms of terminal devices, such as a computer (desktop computer, notebook computer, etc.), an IoT device, or other similar computing devices.

[0146] The electronic device obtains the actual environmental parameters of each first inspection area among a plurality of first inspection areas in the first period. Here, the plurality of first inspection areas are located within the production workplace. Based on the actual environmental parameters of each first inspection area among the plurality of first inspection areas, the predicted results of the first environmental parameters of each first inspection area among the plurality of first inspection areas in the second period are obtained. Here, the second period is a future period of the first period. Based on the area positions of each first inspection area in the production workplace among the plurality of first inspection areas, an inspection area expansion operation on the production workplace is performed to obtain a plurality of second inspection areas. Based on the predicted results of the first environmental parameters of each first inspection area among the plurality of first inspection areas, the predicted results of the second environmental parameters of each second inspection area among the plurality of second inspection areas in the second period are obtained. When it is determined that there is a risk area in a plurality of second inspection areas based on the predicted results of the second environmental parameter of each second inspection area in the plurality of second inspection areas, safety warning information corresponding to the risk area can be generated and transmitted to the target terminal.

[0147] Here, the target terminal may be a computer (such as a desktop computer or a notebook computer) for broadcasting safety warning information, a smartphone, a tablet, a wearable device, or other similar computing devices.

[0148] In addition, in the embodiments of the present disclosure, the actual environmental parameters of each first inspection area in a plurality of first inspection areas during a first period can be collected by having a robot perform a patrol inspection or by pre-installing environmental parameter sensors in each first inspection area in the plurality of first inspection areas.

[0149] It should be noted that in the embodiments of the present disclosure, the schematic diagram showing the application scenario shown in FIG. 4 is not limiting but is schematic, and those skilled in the art can make various obvious changes and / or substitutions based on the example shown in FIG. 4, and it should be noted that the obtained technical solutions still fall within the scope disclosed by the embodiments of the present disclosure.

[0150] To better implement the safety inspection method, the embodiments of the present disclosure also provide a safety inspection device applied to an electronic device. The electronic device can represent a server or various forms of terminal devices, such as a computer (such as a desktop computer or a notebook computer), an IoT device, or other similar computing devices. Hereinafter, the safety inspection device 500 provided by the disclosed embodiments will be described with reference to the schematic block diagram shown in FIG. 5.

[0151] The safety inspection device 500 includes A parameter acquisition unit 501 for obtaining the actual environmental parameters of each first inspection area among a plurality of first inspection areas in the first period, where the plurality of first inspection areas are located in a production workplace, the parameter acquisition unit 501, A first prediction unit 502 for obtaining the first environmental parameter prediction results of each first inspection area among a plurality of first inspection areas in the second period based on the actual environmental parameters of each first inspection area among the plurality of first inspection areas, where the second period is a future period of the first period, the first prediction unit 502, An area selection unit 503 for performing an inspection area expansion operation on the production workplace based on the area positions of each first inspection area in the production workplace among the plurality of first inspection areas to obtain a plurality of second inspection areas, A second prediction unit 504 for obtaining the second environmental parameter prediction results of each second inspection area among the plurality of second inspection areas in the second period based on the first environmental parameter prediction results of each first inspection area among the plurality of first inspection areas, An inspection result acquisition unit 504 for generating safety warning information corresponding to a risk area and transmitting the safety warning information to a target terminal for broadcasting the safety warning information when it is determined that there is a risk area in the plurality of second inspection areas based on the second environmental parameter prediction results of each second inspection area among the plurality of second inspection areas.

[0152] In some alternative embodiments, the first prediction unit 502 uses each first inspection area among the plurality of first inspection areas as a first processing target area, constructs a first input sequence based on the actual environmental parameters of the first processing target area, constructs a second input sequence based on the actual environmental parameters of the first processing target area and a preset input sequence, inputs the first input sequence and the second input sequence into a trained time series model to obtain an overall output sequence of the time series model, and obtains the first environmental parameter prediction results of the first processing target area in the second period based on the overall output sequence.

[0153] In some alternative embodiments, the trained time series model includes an encoder and a decoder, and the first prediction unit 502 inputs the first input sequence into the encoder, processes the first input sequence using the first self-attention module and the distillation module in the encoder to obtain a first input feature mapping result, inputs the second input sequence into the second self-attention module in the encoder, processes the second input sequence using the second self-attention module to obtain a second input feature mapping result, inputs the first input feature mapping result and the second input feature mapping result into the cross-attention module in the encoder, processes the first input feature mapping result and the second input feature mapping result using the cross-attention module to obtain an overall output sequence of the time series model, and is used for

[0154] In some alternative embodiments, the region selection unit 503 uses each first inspection region among a plurality of first inspection regions as a first processing target region, determines at least one risk diffusion region correlated with the first processing target region from a production workplace, and obtains M risk diffusion regions correlated with the plurality of first inspection regions, where M≥2 and M is an integer, and is used for obtaining a plurality of second inspection regions based on the plurality of first inspection regions and the M risk diffusion regions.

[0155] In some alternative embodiments, the region selection unit 503 obtains at least one integratable region group based on the M risk diffusion regions, where each integratable region group in the at least one integratable region group includes N integratable regions among the M risk diffusion regions, 2≤N≤M, and N is an integer, Performing area integration for each integrable area group in at least one integrable area group to obtain at least one risk diffusion integration area, where the at least one risk diffusion integration area corresponds one-to-one with at least one integrable area group, and It is used for forming a plurality of second inspection areas from a plurality of first inspection areas and at least one risk diffusion integration area together.

[0156] In some alternative embodiments, the second prediction unit 504 uses each risk diffusion integration area in the plurality of second inspection areas as a second processing target area, and determines at least one reference area correlated with the second processing target area from among the plurality of first inspection areas; and is used for obtaining a predicted result of a second environmental parameter of the second processing target area in a second period based on a predicted result of a first environmental parameter of at least one reference area.

[0157] In some alternative embodiments, the safety inspection device 500 further includes a risk area determination unit, and the risk area determination unit uses each second inspection area in the plurality of second inspection areas as a third processing target area, and obtains at least one risk evaluation parameter based on a predicted result of a second environmental parameter of the third processing target area; obtains a risk parameter threshold corresponding to each risk evaluation parameter in the at least one risk evaluation parameter; and is used for determining that there is a risk area in the plurality of second inspection areas when it is determined that the third processing target area is a risk area based on the at least one risk evaluation parameter and the risk parameter threshold corresponding to each risk evaluation parameter in the at least one risk evaluation parameter.

[0158] In some alternative embodiments, the inspection result acquisition unit 505 constructs an initial workplace image based on the internal structure of the production workplace; In the initial workplace image, determining a target position corresponding to a risk area, In the initial workplace image, it is used for performing a special display on the target position to obtain a target workplace image, and obtaining safety warning information based on the target workplace image.

[0159] For the specific functions and exemplary descriptions of each module of the safety inspection device 500 according to the embodiments of the present disclosure, reference can be made to the related descriptions of the corresponding steps in the above-described method embodiments, and will not be repeated here.

[0160] In the technical solution of the present disclosure, the acquisition, storage, and application of user personal information comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0161] FIG. 6 is a structural block diagram of an electronic device according to an embodiment of the present disclosure. As shown in FIG. 6, the electronic device includes a memory 610 and a processor 620, and a computer program executable by the processor 620 is stored in the memory 610. The numbers of the memory 610 and the processor 620 can be one or more. The memory 610 can store one or more computer programs, and when the one or more computer programs are executed by the electronic device, the electronic device is caused to execute the method provided by the above method embodiments. The electronic device can further include the following. The communication interface 630 is used for communicating with an external device and performing data interaction and transmission.

[0162] When the memory 610, the processor 620, and the communication interface 630 are implemented independently, the memory 610, the processor 620, and the communication interface 630 are connected to each other via a bus and can communicate with each other. The bus can be, for example, an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus. The bus can be classified into an address bus, a data bus, a control bus, etc. For ease of explanation, only a single thick line is shown in FIG. 6, but it does not mean only a single bus or a single type of bus.

[0163] Optionally, in a specific implementation form, when the memory 610, the processor 620, and the communication interface 630 are integrated on one chip, the memory 610, the processor 620, and the communication interface 630 can communicate with each other via an internal interface.

[0164] It should be understood that the above processor may be a central processing unit (CPU), and may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware assemblies, etc. The general-purpose processor can be, for example, a microprocessor or any conventional processor. In addition, the processor can be a processor that supports the Advanced RISC Machines (ARM) architecture.

[0165] Furthermore, optionally, the memory may include a read-only memory and a random access memory, and may further include a non-volatile random access memory. The memory can be either a volatile memory or a non-volatile memory, or can include both a volatile memory and a non-volatile memory. Here, the non-volatile memory can include a ROM (Read-Only Memory), PROM (Programmable ROM), EPROM (Erasable PROM), EEPROM (Electrically EPROM), or flash memory. The volatile memory can include a random access memory (Random Access Memory, RAM) that functions as an external cache. By way of example and not limitation, many forms of RAM are available. For example, static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic Random Access Memory, DRAM), synchronous DRAM (Synchronous DRAM, SDRAM), double data rate SDRAM (Double Data Rate SDRAM, DDR SDRAM), enhanced SDRAM (Enhanced SDRAM, ESDRAM), synchlink DRAM (Synchlink DRAM, SLDRAM), and direct RAMBUS RAM (Direct RAMBUS RAM, DR RAM).

[0166] In the above embodiments, they may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented by software, the whole or part may be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present disclosure are generated in whole or in part. The computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium, or may be transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as coaxial cable, optical fiber, Digital Subscriber Line (DSL)) or wireless (such as infrared, Bluetooth®, microwave, etc.). The computer-readable storage medium may be any available medium accessible by a computer, or a data storage device including a server, a data center, etc. integrated with one or more available media. The available medium may be a magnetic medium (such as a floppy (registered trademark) disk, a hard disk, a magnetic tape), an optical medium (such as a Digital Versatile Disc (DVD)), or a semiconductor medium (such as a Solid State Disk (SSD)). It should be noted that the computer-readable storage medium mentioned in the present disclosure may be a non-volatile storage medium, in other words, a non-transitory storage medium.

[0167] Those skilled in the art can understand that all or some of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by instructing the relevant hardware through a program, and the program may be stored in a computer-readable storage medium, and the above storage medium may be a read-only memory, a magnetic disk, an optical disk, or the like.

[0168] In the description of the embodiments of the present disclosure, the descriptions of reference terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or features described in relation to the embodiment or example are included in at least one embodiment or example of the present disclosure. And the specific features, structures, materials, or features described can be combined in any one or more embodiments or examples in an appropriate manner. Furthermore, those skilled in the art may combine different embodiments or examples described in the present disclosure and the features of different embodiments or examples without contradiction.

[0169] In the description of the embodiments of the present disclosure, " / " represents the meaning of "or" unless otherwise specified. For example, A / B may represent either A or B. "And / or" in the present disclosure only explains the relationship between related objects and indicates that there may be three types of relationships. For example, A and / or B can indicate the following. There are three situations where A exists alone, A and B exist simultaneously, and B exists alone.

[0170] In the description of the embodiments of the present disclosure, the terms "first" and "second" are used only for the purpose of description and should not be construed as indicating or implying relative importance, nor should they be construed as implying the number of technical features shown. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present disclosure, "a plurality" means two or more unless otherwise specified.

[0171] The above are only exemplary embodiments of the present disclosure and do not limit the present disclosure. Any modifications, equivalent replacements, improvements, etc. made within the spirit and scope of the principles of the present disclosure should all be included within the protection scope of the present disclosure.

Claims

In an electronic device including a processor and a memory, a safety inspection method implemented by the processor executing instructions stored in the memory, comprising: obtaining actual environmental parameters of each first inspection area among a plurality of first inspection areas in a first period, wherein the plurality of first inspection areas are located in a production workplace; obtaining a first environmental parameter prediction result of each first inspection area among the plurality of first inspection areas in a second period based on the actual environmental parameters of each first inspection area among the plurality of first inspection areas, wherein the second period is a future period of the first period; performing an inspection area expansion operation on the production workplace based on the area location of each first inspection area among the plurality of first inspection areas in the production workplace to obtain a plurality of second inspection areas; obtaining a second environmental parameter prediction result of each second inspection area among the plurality of second inspection areas in the second period based on the first environmental parameter prediction result of each first inspection area among the plurality of first inspection areas; when it is determined that there is a risk area in the plurality of second inspection areas based on the second environmental parameter prediction result of each second inspection area among the plurality of second inspection areas, generating safety warning information corresponding to the risk area and transmitting the safety warning information to a target terminal for broadcasting the safety warning information; A safety inspection method.

2. Obtaining the first environmental parameter prediction result of each first inspection area among the plurality of first inspection areas in the second period based on the actual environmental parameters of each first inspection area among the plurality of first inspection areas includes: taking each first inspection area among the plurality of first inspection areas as a first processing target area, and constructing a first input sequence based on the actual environmental parameters of the first processing target area; constructing a second input sequence based on the actual environmental parameters of the first processing target area and a preset input sequence; inputting the first input sequence and the second input sequence into a trained time series model to obtain an overall output sequence of the time series model; obtaining a first environmental parameter prediction result of the first processing target area in the second period based on the overall output sequence; The safety inspection method according to claim 1.

3. The trained time series model includes an encoder and a decoder, Inputting the first input sequence and the second input sequence into the trained time series model to obtain the overall output sequence of the time series model, Inputting the first input sequence into the encoder, and processing the first input sequence by using a first self-attention module and a distillation module in the encoder to obtain a first input feature mapping result, Inputting the second input sequence into a second self-attention module in the encoder, and processing the second input sequence by using the second self-attention module to obtain a second input feature mapping result, Inputting the first input feature mapping result and the second input feature mapping result into a cross-attention module in the encoder, and processing the first input feature mapping result and the second input feature mapping result by using the cross-attention module to obtain the overall output sequence of the time series model, including The safety inspection method according to claim 2.

4. Executing an inspection area expansion operation on the production workplace based on the area position of each first inspection area in the plurality of first inspection areas in the production workplace to obtain a plurality of second inspection areas, Regarding each first inspection area in the plurality of first inspection areas as a first processing target area, determining at least one risk diffusion area correlated with the first processing target area from the production workplace, and obtaining M risk diffusion areas correlated with the plurality of first inspection areas, where M≥2 and M is an integer, Obtaining the plurality of second inspection areas based on the plurality of first inspection areas and the M risk diffusion areas, including The safety inspection method according to claim 1.

5. The safety inspection method is Regarding each second inspection area in the plurality of second inspection areas as a third processing target area, obtaining at least one risk evaluation parameter based on the predicted result of the second environmental parameter of the third processing target area, Obtaining a risk parameter threshold corresponding to each risk evaluation parameter in the at least one risk evaluation parameter Based on the at least one risk assessment parameter and the risk parameter threshold corresponding to each risk assessment parameter in the at least one risk assessment parameter, when it is determined that the third processing target area is a risk area, determining that the risk area exists in the plurality of second inspection areas. The safety inspection method according to claim 1.

6. Generating the safety warning information corresponding to the risk area includes: Constructing an initial workplace image based on the internal structure of the production workplace; Determining a target position corresponding to the risk area in the initial workplace image; In the initial workplace image, performing a special display on the target position to obtain a target workplace image, and obtaining the safety warning information based on the target workplace image. The safety inspection method according to claim 1.

7. A safety inspection device, comprising: A parameter acquisition unit for obtaining the actual environmental parameters of each first inspection area in a plurality of first inspection areas during a first period, wherein the plurality of first inspection areas are located in a production workplace; A first prediction unit for obtaining a first environmental parameter prediction result of each first inspection area in the plurality of first inspection areas during a second period based on the actual environmental parameters of each first inspection area in the plurality of first inspection areas, wherein the second period is a future period of the first period; An area selection unit for performing an inspection area expansion operation on the production workplace based on the area position of each first inspection area in the plurality of first inspection areas in the production workplace to obtain a plurality of second inspection areas; A second prediction unit for obtaining a second environmental parameter prediction result of each second inspection area in the plurality of second inspection areas during the second period based on the first environmental parameter prediction result of each first inspection area in the plurality of first inspection areas; An inspection result acquisition unit for generating safety warning information corresponding to the risk area when it is determined that the risk area exists in the plurality of second inspection areas based on the second environmental parameter prediction result of each second inspection area in the plurality of second inspection areas, and transmitting the safety warning information to a target terminal for broadcasting the safety warning information. A safety inspection device.

8. At least one processor; a memory communicatively connected to the at least one processor wherein the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, cause the at least one processor to execute the method according to any one of claims 1 to 6 An electronic device **Claim 9** A non-transitory computer-readable storage medium storing instructions for causing a computer to execute the method according to any one of claims 1 to 6 **Claim 10** A program which, when executed by a processor in a computer, implements the method according to any one of claims 1 to 6

Images