AI-based mechanical excavation operation dust fall control method and system

Abstract

The invention provides a mechanical excavation operation dust fall control method and system based on AI, and relates to the technical field of artificial intelligence. According to the control method, a first dust diffusion range generated during excavation is predicted according to excavated object information, the excavation direction of an excavation part, excavation information and weather information, and a first water mist area covering the first dust diffusion range is formed by controlling a holder and a power driving part; in this way, the concentration of flying dust generated in the excavation operation process can be reduced, environmental pollution is avoided, and it is guaranteed that on-site constructors have a healthy working environment.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an AI-based dust reduction control method and system for mechanical excavation operations. Background Art

[0002] During the on-site construction process, open-air excavation operations are inevitably involved. Dust will inevitably be generated during the excavation process. The dust will affect the surrounding air environment and cause pollution. It will also cause certain harm to the health of on-site construction workers. How to reduce dust during excavation operations is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0003] In order to at least overcome the above-mentioned deficiencies in the prior art, in a first aspect, the present invention provides an AI-based dust reduction control method for mechanical excavation operations, which is applied to an AI-based dust reduction control system for mechanical excavation operations. The system includes a computer device, a mechanical excavation device, a dust reduction device and an image acquisition device that are communicatively connected. The mechanical excavation device includes a mechanical base, a mechanical arm and an excavation component. One end of the mechanical arm is connected to the mechanical base, and the other end of the mechanical arm is connected to the excavation component. The image acquisition device is arranged on the mechanical excavation device. The dust reduction device is at least partially arranged on the mechanical arm. The dust reduction device includes a water storage container, a power drive unit, a pan-tilt head and a water spray head. The water spray head is fixed to the mechanical arm through a pan-tilt head, and the water storage container is connected to the water spray head through a water pipe. The method includes: Acquire video image information of the mechanical excavation equipment when it is working, and identify the excavated object information at the excavation position during excavation, the excavation direction of the excavation component in the mechanical excavation equipment, and the excavation information of the excavation component based on the video image information; Acquire the geographical location information of the mechanical excavation equipment, and acquire the weather information of the corresponding location area in real time according to the geographical location information, wherein the weather information includes temperature, humidity, wind force and wind direction; Inputting the excavated object information, the excavation direction of the excavation component, the excavation information and the weather information into a pre-trained excavation dust prediction model for prediction, thereby obtaining a first dust diffusion range generated during excavation; The movement of the pan / tilt head is controlled to adjust the water spraying direction of the water spray head, and the working power of the power driving unit is controlled to adjust the water outlet speed of the water spray head to form a first water mist area that at least covers the first dust diffusion range.

[0004] In a possible implementation, the method further includes: Acquire transfer information of the excavated object transferred by the mechanical excavation equipment through the excavation component through the video image information, wherein the transfer information includes a transfer speed and a transfer direction of the excavation component; Input the excavated object information, the weather information, the transfer speed and transfer direction of the excavation component into a pre-trained transfer dust prediction model for prediction, and obtain a second dust diffusion range generated during the transfer; The movement of the pan / tilt head is controlled to adjust the water spraying direction of the water spray head, and the working power of the power driving unit is controlled to adjust the water outlet speed of the water spray head, so as to form a second water mist area that at least covers the second dust diffusion range.

[0005] In a possible implementation, the method further includes: Obtaining unloading information of the excavated object by the mechanical excavation equipment through the video image information, wherein the unloading information includes the distance between the excavated object in the excavation component and the bearing surface during unloading, the unloading direction of the excavated object in the excavation component, and the unloading speed of the excavation component; Inputting the excavated object information, the weather information and the unloading information into a pre-trained unloading dust prediction model for prediction, and obtaining a third dust diffusion range generated during unloading; The movement of the pan / tilt head is controlled to adjust the water spraying direction of the water spray head, and the working power of the power drive unit is controlled to adjust the water outlet speed of the water spray head, so as to form a third water mist area that at least covers the third dust diffusion range.

[0006] In a possible implementation, the step of acquiring video image information of the mechanical excavation equipment when it is working, and identifying information of the excavated object at the excavation position during excavation, the excavation direction of the excavation component in the mechanical excavation equipment, and the excavation information of the excavation component based on the video image information includes: Acquire video image information acquired by the image acquisition device when the mechanical excavation device is working, wherein the video image information at least includes an excavation component of the mechanical excavation device and an excavation position area to which the excavation component is directed during excavation; Inputting the image corresponding to the excavation position area into the object recognition model for recognition, and obtaining the excavated object information at the excavation position, wherein the excavated object information includes the composition type and dryness degree of the excavated object; The excavation components of the mechanical excavation equipment in the video image information are identified to obtain the size of the excavation components, and the excavation speed of the excavation components is obtained by target tracking calculation on the excavation components in the video image information. The excavation information of the excavation components is composed of the size of the excavation components and the excavation speed.

[0007] In a possible implementation, the method further includes a step of training the excavation dust prediction model, the step comprising: Creating a first training sample set, the first training sample set comprising a plurality of first training samples and a first dust diffusion mark range corresponding to each of the first training samples, wherein the first training samples comprise an excavated object information sample, an excavation direction sample of an excavation component, an excavation information sample, and a weather information sample; Inputting the first training sample into a first neural network model for training to obtain a first dust prediction diffusion range, and calculating a first loss function value of the first neural network model based on the first dust prediction diffusion range and the first dust diffusion mark range; When the first loss function value is greater than a preset first loss function threshold, adjusting the network parameters of the first neural network model; Repeatedly input the first training sample in the first training sample set into the first neural network for training until the first loss function value is no greater than the preset first loss function domain value, and the training is completed. The first neural network obtained by training is used as the excavation dust prediction model.

[0008] In a possible implementation, the method further includes a step of training the transferred dust prediction model, the step comprising: Creating a second training sample set, the second training sample set comprising a plurality of second training samples and a second dust diffusion mark range corresponding to each of the second training samples, wherein the second training samples comprise excavated object information samples, weather information samples, transfer speed samples and transfer direction samples of the excavation components; Inputting the second training sample into a second neural network model for training to obtain a second dust predicted diffusion range, and calculating a second loss function value of the second neural network model based on the second dust predicted diffusion range and the second dust diffusion mark range; When the second loss function value is greater than a preset second loss function threshold, adjusting the network parameters of the second neural network model; Repeatedly input the second training sample in the second training sample set into the second neural network for training until the second loss function value is no greater than the preset second loss function domain value, and the training is completed. The second neural network obtained by training is used as the transferred dust prediction model.

[0009] In a possible implementation, the method further includes a step of training the unloading dust prediction model, the step comprising: Creating a third training sample set, the third training sample set comprising a plurality of third training samples and a third dust diffusion mark range corresponding to each of the third training samples, wherein the third training samples comprise excavated object information samples, weather information samples and unloading information samples; Inputting the third training sample into a third neural network model for training to obtain a third dust prediction diffusion range, and calculating a third loss function value of the third neural network model based on the third dust prediction diffusion range and the third dust diffusion mark range; When the third loss function value is greater than a preset third loss function threshold, adjusting the network parameters of the third neural network model; Repeatedly input the third training sample in the third training sample set into the third neural network for training until the third loss function value is no greater than the preset third loss function domain value, and the training is completed. The third neural network obtained by the training is used as the unloading dust prediction model.

[0010] In a possible implementation, the method further includes: comparing the first dust diffusion range with a preset maximum dust diffusion range, and reducing the excavation speed of the excavation component when the first dust diffusion range is greater than the preset maximum dust diffusion range; and / or, comparing the second dust diffusion range with a preset maximum dust diffusion range, and reducing the transfer speed of the excavating component when the second dust diffusion range is greater than the preset maximum dust diffusion range; and / or, The third dust diffusion range is compared with a preset maximum dust diffusion range. When the third dust diffusion range is greater than the preset maximum dust diffusion range, the distance between the excavated object in the excavation component and the bearing surface is reduced, and / or the unloading speed of the excavation component is reduced.

[0011] In a second aspect, the present invention further provides an AI-based dust reduction control system for mechanical excavation operations, the system comprising a computer device, a mechanical excavation device, a dust reduction device and an image acquisition device connected in communication, the mechanical excavation device comprising a mechanical base, a mechanical arm and an excavation component, one end of the mechanical arm is connected to the mechanical base, and the other end of the mechanical arm is connected to the excavation component, the image acquisition device is arranged on the mechanical excavation device, the dust reduction device is at least partially arranged on the mechanical arm, the dust reduction device comprises a water storage container, a power drive unit, a pan-tilt and a water spray head, the water spray head is fixed to the mechanical arm through the pan-tilt, and the water storage container is connected to the water spray head through a water pipe; The image acquisition device is used to acquire video image information at the excavation position when the mechanical excavation device is working; The computer device is used to identify and obtain information of the excavated object at the excavation position, the excavation direction of the excavation component in the mechanical excavation equipment, and the excavation information of the excavation component based on the video image information; The computer device is further used to obtain geographical location information of the mechanical excavation equipment, and obtain weather information of the corresponding location area in real time according to the geographical location information, wherein the weather information includes temperature, humidity, wind force and wind direction; The computer device is further used to input the excavated object information, the excavation direction of the excavation component, the excavation information and the weather information into a pre-trained excavation dust prediction model for prediction, so as to obtain a first dust diffusion range generated during excavation; The computer device is also used to control the movement of the pan / tilt platform to adjust the water spraying direction of the sprinkler head, and to control the working power of the power drive unit to adjust the speed of water discharge from the sprinkler head, so as to form a first water mist area that at least covers the first dust diffusion range.

[0012] In the third aspect, the present invention also provides a computer device, comprising a processor, a computer-readable storage medium and a communication interface, wherein the computer-readable storage medium, the communication interface and the processor are connected via a bus system, the computer-readable storage medium is used to store programs, instructions or codes, and the processor is used to execute the programs, instructions or codes in the computer-readable storage medium to execute the AI-based mechanical excavation dust reduction control method described in any possible implementation method of the first aspect.

[0013] In a fourth aspect, the present invention further provides a computer-readable storage medium, in which instructions are stored, which, when executed, enable a computer to execute an AI-based dust reduction control method for mechanical excavation operations in any possible implementation of the first aspect.

[0014] In the above scheme provided by the present invention, first, the information of the object to be excavated, the excavation direction of the excavation components and the excavation information are obtained based on the video image information when the mechanical excavation equipment is working; then, the geographical location information of the mechanical excavation equipment is obtained, and the weather information of the corresponding location area is obtained in real time according to the geographical location information; then, the information of the object to be excavated, the excavation direction of the excavation components, the excavation information and the weather information are input into the pre-trained excavation dust prediction model for prediction, and the first dust diffusion range generated during excavation is obtained; finally, the pan-tilt head and the power drive unit are controlled to form a first water mist area that at least covers the first dust diffusion range. The above scheme predicts the first dust diffusion range generated during excavation based on the information of the object to be excavated, the excavation direction of the excavation components, the excavation information and the weather information, and forms the first water mist area covering the first dust diffusion range by controlling the pan-tilt head and the power drive unit, so that the dust concentration generated during the excavation operation can be reduced, environmental pollution can be avoided, and a healthy working environment can be ensured for on-site construction personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0016] Figure 1 A block diagram of a mechanical excavation dust reduction control system based on AI provided by an embodiment of the present invention; Figure 2 A schematic flow chart of a dust reduction control method for mechanical excavation operations based on AI provided in an embodiment of the present invention; Figure 3 One of the partial flow diagrams of the AI-based dust reduction control method for mechanical excavation operations provided in an embodiment of the present invention; Figure 4 A second partial flow chart of a method for controlling dust reduction in mechanical excavation operations based on AI provided in an embodiment of the present invention; Figure 5 A schematic diagram of the structural framework of a computer device for implementing the above-mentioned AI-based mechanical excavation dust reduction control method provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] The present invention is described in detail below with reference to the accompanying drawings. The specific operation method in the method embodiment can also be applied to the device embodiment or the system embodiment.

[0018] Before introducing the specific solution provided by this embodiment, the application scenario of the AI-based mechanical excavation dust reduction control system to which the specific solution is applicable is introduced. Please refer to Figure 1 In this embodiment, the AI-based mechanical excavation dust reduction control system 1 includes a computer device 10, a mechanical excavation device 20, a dust reduction device 30 and an image acquisition device 40 that are connected in communication, wherein the mechanical excavation device 20 includes a mechanical base, a mechanical arm and an excavation component, one end of the mechanical arm is connected to the mechanical base, the other end of the mechanical arm is connected to the excavation component, the excavation component is movably connected to the mechanical arm, and the image acquisition device 40 is arranged on the mechanical excavation device 20. For example, the image acquisition device 40 is arranged on the mechanical arm. The dust reduction device 30 is at least partially arranged on the mechanical arm. The dust reduction device 30 includes a water storage container, a power drive unit, a pan-tilt and a sprinkler. A plurality of pan-tilts can be arranged in an array and fixed on the mechanical arm. The sprinkler is fixed to the mechanical arm through the pan-tilt. The water spraying direction of the sprinkler can be controlled by the pan-tilt. The water spraying height of the sprinkler can be controlled by the working power of the power drive component. Generally, the greater the working power of the power drive component, the greater the speed of water out of the sprinkler, and the greater the water spraying height. In this embodiment, the water storage container and the sprinkler can be connected by a water pipe.

[0019] Combine the following Figure 1 The application scenario shown is an exemplary description of the AI-based mechanical excavation dust control method provided in the embodiment of the present application. Figure 2 The AI-based mechanical excavation dust reduction control method provided in the embodiment of the present application can be executed by the aforementioned computer device 10. In other embodiments, the order of some steps in the AI-based mechanical excavation dust reduction control method in the embodiment of the present application can be interchanged according to actual needs, or some of the steps can be omitted or deleted. The detailed steps of the AI-based mechanical excavation dust reduction control method executed by the computer device 10 are introduced as follows.

[0020] Step S10, obtaining video image information of the mechanical excavation equipment at work collected by the image acquisition device, and identifying the excavated object information at the excavation position during excavation, the excavation direction of the excavation components in the mechanical excavation equipment, and the excavation information of the excavation components based on the video image information.

[0021] In this embodiment, step S10 can be implemented in the following manner.

[0022] First, the video image information collected by the image acquisition device 40 when the mechanical excavation device 20 is working is obtained. The video image information at least includes the excavation components of the mechanical excavation device 20 and the excavation position area to which the excavation components face during excavation.

[0023] Next, the image corresponding to the excavation location area is input into the object recognition model for recognition, and the excavated object information at the excavation location is obtained. The excavated object information includes the composition type and dryness of the excavated object, wherein the composition type of the excavated object may include sand, gravel, soil, and various mixtures, etc. The composition type of the excavated object can also be distinguished according to whether it is easy to generate dust, such as sticky objects and non-sticky objects. The dryness can be measured by the water content of the excavated object, wherein the higher the dryness, the easier it is to generate dust during the excavation process.

[0024] Then, the excavation components of the mechanical excavation equipment in the video image information are identified to obtain the size of the excavation components, and the excavation speed of the excavation components is obtained by target tracking and calculation of the excavation components in the video image information. The excavation information of the excavation components is composed of the size of the excavation components and the excavation speed.

[0025] Step S20, obtaining geographical location information of the mechanical excavation equipment, and obtaining weather information of the corresponding location area in real time according to the geographical location information.

[0026] Exemplarily, a positioning device can be configured on the mechanical excavation equipment 20, and the computer device 10 obtains the geographical location information of the mechanical excavation equipment 20 through the positioning device. The computer device 10 can obtain weather information of the location area in real time through the geographical location information, wherein the weather information includes temperature, humidity, wind force and wind direction, etc.

[0027] Step S30, inputting the excavated object information, the excavation direction of the excavation components, the excavation information and the weather information into a pre-trained excavation dust prediction model for prediction, thereby obtaining a first dust diffusion range generated during excavation.

[0028] The excavation dust prediction model can predict and output the first dust diffusion range generated during excavation based on the input excavated object information, the excavation direction of the excavation components, the excavation information, and the weather information. The first dust diffusion range represents the possible diffusion range of dust generated by the mechanical excavation equipment 20 during the excavation process. The diffusion range is closely related to the excavated object information, the excavation direction of the excavation components, the excavation information, and the weather information. For example, when the excavated object is an object that is prone to dust, the first dust diffusion range is generally larger; for another example, when the excavation direction and the wind direction are the same, the dust will be more easily diffused, and the first dust diffusion range is generally larger. In this embodiment, the first dust diffusion range is generally a spatial range covering the excavation components.

[0029] Step S40, controlling the movement of the pan / tilt to adjust the water spraying direction of the sprinkler head, and controlling the working power of the power drive unit to adjust the water outlet speed of the sprinkler head, so as to form a first water mist area that at least covers the first dust diffusion range.

[0030] The formed water mist area can be changed by adjusting the working power of the pan / tilt head and the power drive unit, thereby forming a first water mist area that can cover the first dust diffusion range, so as to achieve the purpose of reducing dust generated during excavation.

[0031] The above-mentioned scheme provided in this embodiment predicts the first dust diffusion range generated during excavation based on the information of the excavated object, the excavation direction of the excavation components, the excavation information and the weather information, and forms a first water mist area covering the first dust diffusion range by controlling the pan-tilt head and the power drive unit. This can reduce the dust concentration generated during the excavation operation, avoid environmental pollution, and ensure that on-site construction personnel have a healthy working environment.

[0032] Furthermore, before step S30, the method provided in this embodiment also includes a step of training an excavation dust prediction model, which can be implemented in the following manner.

[0033] First, a first training sample set is created, which includes multiple first training samples and a first dust diffusion mark range corresponding to each first training sample, wherein the first training sample includes an excavated object information sample, an excavation direction sample of an excavation component, an excavation information sample, and a weather information sample.

[0034] Next, the first training sample is input into the first neural network model for training to obtain a first dust prediction diffusion range, and a first loss function value of the first neural network model is calculated based on the first dust prediction diffusion range and the first dust diffusion mark range.

[0035] Then, when the first loss function value is greater than a preset first loss function threshold, the network parameters of the first neural network model are adjusted.

[0036] Finally, the first training sample in the first training sample set is repeatedly input into the first neural network for training until the first loss function value is no greater than the preset first loss function domain value, and the training is completed. The first neural network obtained by training is used as the mining dust prediction model.

[0037] Furthermore, the inventors have found that dust is generated not only during the excavation process of the mechanical excavation equipment 20, but also during the process of the mechanical excavation equipment 20 transferring the excavated objects. In order to solve this technical problem, please refer to Figure 3 , the method provided in this embodiment also includes the following steps.

[0038] Step S51, obtaining transfer information of the mechanical excavation equipment 20 transferring the excavated object through the excavation components through video image information, the transfer information including the transfer speed and transfer direction of the excavation components.

[0039] In detail, the transfer information of transferring the excavated object may be determined based on a target tracking method of the excavation component in the video image.

[0040] Step S52, the excavated object information, weather information, transfer speed and transfer direction of the excavating components are input into a pre-trained transfer dust prediction model for prediction, so as to obtain a second dust diffusion range generated during the transfer.

[0041] During the transfer, the information of the excavated object, weather information, the transfer speed of the excavation components and the transfer direction will affect the amount of dust. For example, when the excavated object is sand that is prone to dust, the faster the transfer speed, the easier it is to form dust. In the presence of wind, the stronger the wind, the easier it is to form dust, especially when the transfer direction is opposite to the wind direction. By combining the above factors to predict dust during the transfer process, the second dust diffusion range generated during the transfer can be predicted.

[0042] Step S53, controlling the movement of the pan / tilt to adjust the water spraying direction of the sprinkler head, and controlling the working power of the power drive unit to adjust the water outlet speed of the sprinkler head, so as to form a second water mist area that at least covers the second dust diffusion range.

[0043] The formed water mist area can be changed by adjusting the working power of the pan / tilt head and the power drive unit, thereby forming a second water mist area that can cover the second dust diffusion range, so as to achieve the purpose of reducing dust generated during the transfer process.

[0044] Furthermore, before step S52, the method provided in this embodiment also includes a step of training a dust transfer prediction model, which can be implemented in the following manner.

[0045] First, a second training sample set is created, the second training sample set includes a plurality of second training samples and a second dust diffusion mark range corresponding to each second training sample, wherein the second training sample includes an excavated object information sample, a weather information sample, a transfer speed sample of an excavation component, and a transfer direction sample.

[0046] Next, the second training sample is input into the second neural network model for training to obtain a second dust predicted diffusion range, and a second loss function value of the second neural network model is calculated based on the second dust predicted diffusion range and the second dust diffusion mark range.

[0047] Then, when the second loss function value is greater than a preset second loss function threshold, the network parameters of the second neural network model are adjusted.

[0048] Finally, the second training sample in the second training sample set is repeatedly input into the second neural network for training until the second loss function value is no greater than the preset second loss function domain value. The training is completed and the second neural network obtained by training is used as the transfer dust prediction model.

[0049] Furthermore, the inventors have also found that dust will also be generated during the process of unloading the excavated object by the mechanical excavation equipment 20. Specifically, during the process of unloading the excavated object from the excavation component to the bearing surface, due to the distance between the excavation component and the bearing surface, dust will be generated when the excavated object falls to the bearing surface. In order to solve this technical problem, please refer to Figure 4 , the method provided in this embodiment also includes the following steps.

[0050] Step S61, obtaining unloading information of the mechanical excavation equipment unloading the excavated object through video image information.

[0051] In this embodiment, the unloading information includes the distance between the excavated object in the excavating component and the bearing surface during unloading, the unloading direction of the excavated object in the excavating component, and the unloading speed of the excavating component.

[0052] Step S62, the excavated object information, weather information and unloading information are input into a pre-trained unloading dust prediction model for prediction, so as to obtain a third dust diffusion range generated during unloading.

[0053] During unloading, the information of the excavated object, weather information and unloading information will affect the amount of dust. By combining the above factors to predict the dust during unloading, the third dust diffusion range generated during unloading can be predicted.

[0054] Step S63, controlling the movement of the pan / tilt to adjust the water spraying direction of the sprinkler head, and controlling the working power of the power drive unit to adjust the water outlet speed of the sprinkler head, so as to form a third water mist area that at least covers a third dust diffusion range.

[0055] The formed water mist area can be changed by adjusting the working power of the pan / tilt head and the power drive unit, thereby forming a third water mist area that can cover the third dust diffusion range, so as to achieve the purpose of reducing dust generated during unloading.

[0056] Furthermore, before step S62, the method provided in this embodiment also includes a step of training an unloading dust prediction model, which can be implemented in the following manner.

[0057] First, a third training sample set is created, the third training sample set including a plurality of third training samples and a third dust diffusion mark range corresponding to each third training sample, wherein the third training sample includes an excavated object information sample, a weather information sample and an unloading information sample.

[0058] Next, the third training sample is input into the third neural network model for training to obtain a third dust predicted diffusion range, and a third loss function value of the third neural network model is calculated based on the third dust predicted diffusion range and the third dust diffusion mark range.

[0059] Then, when the third loss function value is greater than a preset third loss function threshold, the network parameters of the third neural network model are adjusted.

[0060] Finally, the third training sample in the third training sample set is repeatedly input into the third neural network for training until the third loss function value is no greater than the preset third loss function domain value. The training is completed and the third neural network obtained by the training is used as the unloading dust prediction model.

[0061] Furthermore, in order to prevent the dust generated during the construction process from spreading beyond the water mist area, thereby reducing the dust reduction effect, the method adopted in this embodiment also includes the following steps.

[0062] comparing the first dust diffusion range with a preset maximum dust diffusion range, and reducing the excavation speed of the excavation component when the first dust diffusion range is greater than the preset maximum dust diffusion range; and / or, comparing the second dust diffusion range with a preset maximum dust diffusion range, and reducing the transfer speed of the excavation component when the second dust diffusion range is greater than the preset maximum dust diffusion range; and / or, The third dust diffusion range is compared with a preset maximum dust diffusion range. When the third dust diffusion range is greater than the preset maximum dust diffusion range, the distance between the excavated object in the excavation component and the bearing surface is reduced, and / or the unloading speed of the excavation component is reduced.

[0063] In this embodiment, the preset maximum dust diffusion range can correspond to the maximum water mist area that can be formed by the dust reduction device 30. Through the above control, it can be ensured that the water mist area formed by the dust reduction device 30 can cover the dust diffusion range generated during the construction process, thereby achieving the purpose of reducing dust.

[0064] Please refer again Figure 1The present embodiment provides a mechanical excavation dust reduction control system based on AI. The mechanical excavation dust reduction control system 1 based on AI includes a computer device 10, a mechanical excavation device 20, a dust reduction device 30 and an image acquisition device 40 which are communicatively connected. The mechanical excavation device 20 includes a mechanical base, a mechanical arm and an excavation component. One end of the mechanical arm is connected to the mechanical base, and the other end of the mechanical arm is connected to the excavation component. The excavation component is movably connected to the mechanical arm. The image acquisition device 40 is arranged on the mechanical excavation device 20. For example, the image acquisition device 40 is arranged on the mechanical arm. The dust reduction device 30 is at least partially arranged on the mechanical arm. The dust reduction device 30 includes a water storage container, a power drive unit, a pan-tilt and a sprinkler. A plurality of pan-tilts can be arranged in an array and fixed on the mechanical arm. The sprinkler is fixed to the mechanical arm through the pan-tilt. The spray direction of the sprinkler can be controlled by the pan-tilt. The spray height of the sprinkler can be controlled by the working power of the power drive component. Generally, the greater the working power of the power drive component, the greater the speed of water out of the sprinkler, and the greater the spray height. In this embodiment, the water storage container and the water spray head can be connected through a water pipe.

[0065] The image acquisition device 40 is used to acquire video image information at the excavation position when the mechanical excavation device 20 is working, and send the video image information to the computer device 10.

[0066] The computer device 10 is used to identify the excavated object information at the excavation position, the excavation direction of the excavation component in the mechanical excavation equipment, and the excavation information of the excavation component based on the video image information.

[0067] In this embodiment, the computer device 10 can implement the above process in the following manner.

[0068] First, the computer device 10 obtains video image information collected by the image acquisition device 40 when the mechanical excavation device 20 is working. The video image information at least includes the excavation components of the mechanical excavation device 20 and the excavation position area to which the excavation components face during excavation.

[0069] Next, the computer device 10 inputs the image corresponding to the excavation position area into the object recognition model for recognition, and obtains the excavated object information at the excavation position. The excavated object information includes the composition type and dryness of the excavated object, wherein the composition type of the excavated object may include sand, gravel, soil, and various mixtures, etc. The composition type of the excavated object can also be distinguished according to whether it is easy to generate dust, such as sticky objects and non-sticky objects, and the dryness can be measured by the water content of the excavated object, wherein the higher the dryness, the easier it is to generate dust during the excavation process.

[0070] Then, the computer device 10 identifies the excavation components of the mechanical excavation equipment in the video image information to obtain the size of the excavation components, performs target tracking calculation on the excavation components in the video image information to obtain the excavation speed of the excavation components, and the excavation information of the excavation components is composed of the size of the excavation components and the excavation speed.

[0071] The computer device 10 is also used to obtain the geographical location information of the mechanical excavation equipment, and obtain the weather information of the corresponding location area in real time according to the geographical location information, wherein the weather information includes temperature, humidity, wind force and wind direction.

[0072] Exemplarily, a positioning device can be configured on the mechanical excavation equipment 20, and the computer device 10 obtains the geographical location information of the mechanical excavation equipment 20 through the positioning device. The computer device 10 can obtain weather information of the location area in real time through the geographical location information, wherein the weather information includes temperature, humidity, wind force and wind direction, etc.

[0073] The computer device 10 is also used to input the excavated object information, the excavation direction of the excavation component, the excavation information and the weather information into a pre-trained excavation dust prediction model to perform prediction, and obtain the first dust diffusion range generated during excavation. The excavation dust prediction model can predict and output the first dust diffusion range generated during excavation based on the input excavated object information, the excavation direction of the excavation components, the excavation information, and the weather information. The first dust diffusion range represents the possible diffusion range of dust generated by the mechanical excavation equipment 20 during the excavation process. The diffusion range is closely related to the excavated object information, the excavation direction of the excavation components, the excavation information, and the weather information. For example, when the excavated object is an object that is prone to dust, the first dust diffusion range is generally larger; for another example, when the excavation direction and the wind direction are the same, the dust will be more easily diffused, and the first dust diffusion range is generally larger. In this embodiment, the first dust diffusion range is generally a spatial range covering the excavation components.

[0074] The computer device 10 is also used to control the movement of the pan / tilt head to adjust the water spraying direction of the sprinkler head, and to control the working power of the power drive unit to adjust the water outlet speed of the sprinkler head, so as to form a first water mist area that at least covers the first dust diffusion range.

[0075] The computer device 10 can change the formed water mist area by adjusting the working power of the pan / tilt head and the power drive unit, thereby forming a first water mist area that can cover the first dust diffusion range, so as to achieve the purpose of reducing dust generated during the excavation process.

[0076] Please refer to Figure 5 , Figure 5FIG. 1 is a schematic diagram of the hardware structure of a computer device 10 for implementing the above-mentioned AI-based mechanical excavation dust reduction control method provided by an embodiment of the present disclosure. The computer device 10 can be implemented on a cloud server. Figure 5 As shown, the computer device 10 may include a processor 101 , a computer-readable storage medium 102 , a bus 103 , and a communication interface 104 .

[0077] During the specific implementation process, at least one processor 101 executes computer-executable instructions stored in a computer-readable storage medium 102, so that the processor 101 can execute the AI-based mechanical excavation dust reduction control method of the above method embodiment, wherein the processor 101, the computer-readable storage medium 102 and the communication interface 104 are connected via a bus 103, and the processor 101 can be used to control the sending and receiving actions of the communication interface 104.

[0078] The specific implementation process of the processor 101 can refer to the various method embodiments executed by the above-mentioned computer device 10. The implementation principles and technical effects are similar and will not be repeated in this embodiment.

[0079] The computer-readable storage medium 102 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk storage.

[0080] The bus 103 can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, the bus in the drawings of the present invention is not limited to only one bus or one type of bus.

[0081] In addition, an embodiment of the present invention further provides a readable storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the above-mentioned AI-based dust reduction control method for mechanical excavation operations is implemented.

[0082] In summary, the technical solution provided by the embodiment of the present invention firstly obtains the information of the excavated object, the excavation direction of the excavation component and the excavation information during excavation based on the video image information when the mechanical excavation equipment is working; then, obtains the geographical location information of the mechanical excavation equipment, and obtains the weather information of the corresponding location area in real time according to the geographical location information; then, inputs the information of the excavated object, the excavation direction of the excavation component, the excavation information and the weather information into the pre-trained excavation dust prediction model for prediction, and obtains the first dust diffusion range generated during excavation; finally, controls the pan-tilt head and the power drive unit to form a first water mist area that at least covers the first dust diffusion range. The above scheme predicts the first dust diffusion range generated during excavation based on the information of the excavated object, the excavation direction of the excavation component, the excavation information and the weather information, and forms the first water mist area covering the first dust diffusion range by controlling the pan-tilt head and the power drive unit, so as to reduce the dust concentration generated during the excavation operation, avoid environmental pollution, and ensure that the on-site construction personnel have a healthy working environment.

[0083] The above is a description of a specific embodiment of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A dust reduction control method for mechanical excavation operations based on AI, characterized in that: The invention is applied to a dust reduction control system for mechanical excavation operation based on AI, the system comprises a computer device, a mechanical excavation device, a dust reduction device and an image acquisition device connected in communication, the mechanical excavation device comprises a mechanical base, a mechanical arm and an excavation component, one end of the mechanical arm is connected to the mechanical base, and the other end of the mechanical arm is connected to the excavation component, the image acquisition device is arranged on the mechanical excavation device, the dust reduction device is at least partially arranged on the mechanical arm, the dust reduction device comprises a water storage container, a power drive unit, a pan-tilt and a water spray head, the water spray head is fixed on the mechanical arm through the pan-tilt, the water storage container and the water spray head are connected through a water pipe, and the method comprises: Acquire video image information of the mechanical excavation equipment when it is working, and identify the excavated object information at the excavation position during excavation, the excavation direction of the excavation component in the mechanical excavation equipment, and the excavation information of the excavation component based on the video image information; Acquire the geographical location information of the mechanical excavation equipment, and acquire the weather information of the corresponding location area in real time according to the geographical location information, wherein the weather information includes temperature, humidity, wind force and wind direction; Inputting the excavated object information, the excavation direction of the excavation component, the excavation information and the weather information into a pre-trained excavation dust prediction model for prediction, thereby obtaining a first dust diffusion range generated during excavation; The movement of the pan / tilt head is controlled to adjust the water spraying direction of the water spray head, and the working power of the power driving unit is controlled to adjust the water outlet speed of the water spray head to form a first water mist area that at least covers the first dust diffusion range.

2. The AI-based dust reduction control method for mechanical excavation operations according to claim 1, characterized in that: The method further comprises: Acquire transfer information of the excavated object transferred by the mechanical excavation equipment through the excavation component through the video image information, wherein the transfer information includes a transfer speed and a transfer direction of the excavation component; Input the excavated object information, the weather information, the transfer speed and transfer direction of the excavation component into a pre-trained transfer dust prediction model for prediction, and obtain a second dust diffusion range generated during the transfer; The movement of the pan / tilt head is controlled to adjust the water spraying direction of the water spray head, and the working power of the power driving unit is controlled to adjust the water outlet speed of the water spray head, so as to form a second water mist area that at least covers the second dust diffusion range.

3. The AI-based dust reduction control method for mechanical excavation operations according to claim 2, characterized in that: The method further comprises: Obtaining unloading information of the excavated object by the mechanical excavation equipment through the video image information, wherein the unloading information includes the distance between the excavated object in the excavation component and the bearing surface during unloading, the unloading direction of the excavated object in the excavation component, and the unloading speed of the excavation component; Inputting the excavated object information, the weather information and the unloading information into a pre-trained unloading dust prediction model for prediction, and obtaining a third dust diffusion range generated during unloading; The movement of the pan / tilt head is controlled to adjust the water spraying direction of the water spray head, and the working power of the power drive unit is controlled to adjust the water outlet speed of the water spray head, so as to form a third water mist area that at least covers the third dust diffusion range.

4. The AI-based dust reduction control method for mechanical excavation operations according to claim 3, characterized in that: The step of acquiring video image information of the mechanical excavation equipment when it is working, and identifying the excavated object information at the excavation position during excavation, the excavation direction of the excavation component in the mechanical excavation equipment, and the excavation information of the excavation component based on the video image information, comprises: Acquire video image information acquired by the image acquisition device when the mechanical excavation device is working, wherein the video image information at least includes an excavation component of the mechanical excavation device and an excavation position area to which the excavation component is directed during excavation; Inputting the image corresponding to the excavation position area into the object recognition model for recognition, and obtaining the excavated object information at the excavation position, wherein the excavated object information includes the composition type and dryness degree of the excavated object; The excavation components of the mechanical excavation equipment in the video image information are identified to obtain the size of the excavation components, and the excavation speed of the excavation components is obtained by target tracking calculation on the excavation components in the video image information. The excavation information of the excavation components is composed of the size of the excavation components and the excavation speed.

5. The AI-based dust reduction control method for mechanical excavation operations according to claim 3, characterized in that: The method further comprises the step of training the excavation dust prediction model, the step comprising: Creating a first training sample set, the first training sample set comprising a plurality of first training samples and a first dust diffusion mark range corresponding to each of the first training samples, wherein the first training samples comprise an excavated object information sample, an excavation direction sample of an excavation component, an excavation information sample, and a weather information sample; Inputting the first training sample into a first neural network model for training to obtain a first dust prediction diffusion range, and calculating a first loss function value of the first neural network model based on the first dust prediction diffusion range and the first dust diffusion mark range; When the first loss function value is greater than a preset first loss function threshold, adjusting the network parameters of the first neural network model; Repeatedly input the first training sample in the first training sample set into the first neural network for training until the first loss function value is no greater than the preset first loss function domain value, and the training is completed. The first neural network obtained by training is used as the excavation dust prediction model.

6. The AI-based dust reduction control method for mechanical excavation operations according to claim 3, characterized in that: The method further comprises the step of training the transferred dust prediction model, the step comprising: Creating a second training sample set, the second training sample set comprising a plurality of second training samples and a second dust diffusion mark range corresponding to each of the second training samples, wherein the second training samples comprise excavated object information samples, weather information samples, transfer speed samples and transfer direction samples of the excavation components; Inputting the second training sample into a second neural network model for training to obtain a second dust predicted diffusion range, and calculating a second loss function value of the second neural network model based on the second dust predicted diffusion range and the second dust diffusion mark range; When the second loss function value is greater than a preset second loss function threshold, adjusting the network parameters of the second neural network model; Repeatedly input the second training sample in the second training sample set into the second neural network for training until the second loss function value is no greater than the preset second loss function domain value, and the training is completed. The second neural network obtained by training is used as the transferred dust prediction model.

7. The AI-based dust reduction control method for mechanical excavation operations according to claim 3, characterized in that: The method further comprises the step of training the unloading dust prediction model, the step comprising: Creating a third training sample set, the third training sample set comprising a plurality of third training samples and a third dust diffusion mark range corresponding to each of the third training samples, wherein the third training samples comprise excavated object information samples, weather information samples and unloading information samples; Inputting the third training sample into a third neural network model for training to obtain a third dust prediction diffusion range, and calculating a third loss function value of the third neural network model based on the third dust prediction diffusion range and the third dust diffusion mark range; When the third loss function value is greater than a preset third loss function threshold, adjusting the network parameters of the third neural network model; Repeatedly input the third training sample in the third training sample set into the third neural network for training until the third loss function value is no greater than the preset third loss function domain value, and the training is completed. The third neural network obtained by the training is used as the unloading dust prediction model.

8. The AI-based dust reduction control method for mechanical excavation operations according to claim 3, characterized in that: The method further comprises: comparing the first dust diffusion range with a preset maximum dust diffusion range, and reducing the excavation speed of the excavation component when the first dust diffusion range is greater than the preset maximum dust diffusion range; and / or, comparing the second dust diffusion range with a preset maximum dust diffusion range, and reducing the transfer speed of the excavating component when the second dust diffusion range is greater than the preset maximum dust diffusion range; and / or, The third dust diffusion range is compared with a preset maximum dust diffusion range. When the third dust diffusion range is greater than the preset maximum dust diffusion range, the distance between the excavated object in the excavation component and the bearing surface is reduced, and / or the unloading speed of the excavation component is reduced.

9. An AI-based dust reduction control system for mechanical excavation operations, characterized in that: The system includes a computer device, a mechanical excavation device, a dust suppression device and an image acquisition device that are in communication connection. The mechanical excavation device includes a mechanical base, a mechanical arm and an excavation component. One end of the mechanical arm is connected to the mechanical base, and the other end of the mechanical arm is connected to the excavation component. The image acquisition device is arranged on the mechanical excavation device. The dust suppression device is at least partially arranged on the mechanical arm. The dust suppression device includes a water storage container, a power drive unit, a pan-tilt and a water spray head. The water spray head is fixed to the mechanical arm through the pan-tilt, and the water storage container is connected to the water spray head through a water pipe. The image acquisition device is used to acquire video image information at the excavation position when the mechanical excavation device is working; The computer device is used to identify and obtain information of the excavated object at the excavation position, the excavation direction of the excavation component in the mechanical excavation equipment, and the excavation information of the excavation component based on the video image information; The computer device is further used to obtain geographical location information of the mechanical excavation equipment, and obtain weather information of the corresponding location area in real time according to the geographical location information, wherein the weather information includes temperature, humidity, wind force and wind direction; The computer device is further used to input the excavated object information, the excavation direction of the excavation component, the excavation information and the weather information into a pre-trained excavation dust prediction model for prediction, so as to obtain a first dust diffusion range generated during excavation; The computer device is also used to control the movement of the pan / tilt platform to adjust the water spraying direction of the sprinkler head, and to control the working power of the power drive unit to adjust the speed of water discharge from the sprinkler head, so as to form a first water mist area that at least covers the first dust diffusion range.

10. A computer device, characterized in that: The computer device includes a processor, a readable storage medium and a communication interface. The readable storage medium, the communication interface and the processor are connected via a bus system. The readable storage medium is used to store programs, instructions or codes. The processor is used to execute the programs, instructions or codes in the readable storage medium to execute the AI-based mechanical excavation dust reduction control method as described in any one of claims 1-8.

Images