A method and platform for adaptively controlling the working environment of excavating equipment

Abstract

The present invention discloses an adaptive control method and platform for the working environment of an excavation device, and relates to the field of excavation control technology. The method includes: obtaining basic environmental information of the working environment and establishing a spatial data set; performing key point-based positioning fusion on the spatial data set and establishing a three-dimensional reconstruction model; reading the working tasks of the excavation device and establishing a temporal attention constraint based on the task execution fitting result; and issuing a warning through an environmental warning channel built into the three-dimensional reconstruction model. The present invention solves the technical problems existing in the prior art of insufficient environmental perception capability and lack of intelligent warning mechanism, which lead to difficulty in timely identification and warning of potential risks, thereby affecting the safety and efficiency of the operation, and achieves the technical effect of realizing comprehensive monitoring of the working environment and instant abnormal warning, and improving the safety and efficiency of the operation.

Description

Technical Field

[0001] The present invention relates to the technical field of excavation control, and in particular to an adaptive control method and platform for the working environment of an excavation device. Background Art

[0002] In the operation process of modern excavation equipment, the equipment often needs to deal with complex and changing environmental factors, such as terrain undulations, obstacle distribution, and differences in material types. These environmental factors directly affect the operating efficiency, accuracy, and safety of the equipment. However, traditional excavation equipment usually relies on pre-set parameters and operating instructions, and lacks the ability to adapt to real-time environmental changes, which may lead to misoperation, equipment loss, and safety hazards during the operation. At the same time, the monitoring and early warning mechanism of the operating environment is relatively lagging, and it is impossible to perceive potential risks in real time and take countermeasures in advance.

[0003] Existing technologies have technical problems such as insufficient environmental perception capabilities and lack of intelligent early warning mechanisms, which make it difficult to identify and warn of potential risks in a timely manner, thus affecting operational safety and efficiency. Summary of the invention

[0004] The present application provides an adaptive control method and platform for the working environment of an excavation device, which is used to solve the technical problems in the prior art of insufficient environmental perception capability and lack of intelligent early warning mechanism, which make it difficult to identify and warn of potential risks in a timely manner, thereby affecting the safety and efficiency of operations.

[0005] In view of the above problems, the present application provides a method and platform for adaptively controlling the working environment of an excavation equipment.

[0006] In a first aspect of the present application, a method for adaptively controlling an operating environment of an excavation device is provided, the method comprising:

[0007] The basic environmental information of the working environment is obtained, and after evaluating the basic environmental information, the joint sensor is called to collect spatial data of the working environment to establish a spatial data set; the spatial data set is positioned and fused based on key points to establish a three-dimensional reconstruction model; the working task of the mining equipment is read, and the three-dimensional reconstruction model is used to perform task execution fitting on the working task, and a temporal attention constraint is established based on the task execution fitting result; the working environment is collected in real time through the temporal attention constraint, and the real-time collected data is updated to the three-dimensional reconstruction model, and constraint compensation of the real-time attention constraint is established with the update result, the data update iteration of the three-dimensional reconstruction model is executed, and a warning alarm is issued through the environmental warning channel built into the three-dimensional reconstruction model.

[0008] A second aspect of the present application provides an adaptive control platform for an excavation equipment working environment, the platform comprising:

[0009] A spatial data set establishment module, the spatial data set establishment module is used to obtain the basic environmental information of the working environment, and after evaluating the basic environmental information, call the joint sensor to collect the spatial data of the working environment to establish a spatial data set; a three-dimensional reconstruction model establishment module, the three-dimensional reconstruction model establishment module is used to perform key point-based positioning fusion on the spatial data set to establish a three-dimensional reconstruction model; a temporal attention constraint establishment module, the temporal attention constraint establishment module is used to read the working task of the mining equipment, use the three-dimensional reconstruction model to perform task execution fitting on the working task, and establish a temporal attention constraint based on the task execution fitting result; a warning alarm output module, the warning alarm output module is used to perform real-time acquisition of the working environment through the temporal attention constraint, and update the real-time acquisition data to the three-dimensional reconstruction model, and establish constraint compensation of the real-time attention constraint with the update result, execute data update iteration of the three-dimensional reconstruction model, and issue a warning alarm through the environmental warning channel built into the three-dimensional reconstruction model.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] Obtain basic environmental information of the working environment and establish a spatial data set; perform key point-based positioning fusion on the spatial data set to establish a three-dimensional reconstruction model; read the working tasks of the mining equipment and establish temporal attention constraints; perform real-time acquisition of the working environment through the temporal attention constraints, and update the real-time acquisition data to the three-dimensional reconstruction model, and establish constraint compensation of the real-time attention constraints with the update results, perform data update iterations of the three-dimensional reconstruction model, and issue early warnings through the environmental warning channel built into the three-dimensional reconstruction model. The technical effect of achieving comprehensive monitoring of the working environment and instant abnormal warnings, and improving work safety and efficiency has been achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic flow chart of a method for adaptively controlling the working environment of an excavation equipment provided in an embodiment of the present application;

[0014] Figure 2 A schematic diagram of the structure of an adaptive control platform for the working environment of an excavator provided in an embodiment of the present application.

[0015] Explanation of reference numerals: spatial data set establishing module 10 , three-dimensional reconstruction model establishing module 20 , temporal attention constraint establishing module 30 , early warning alarm output module 40 . DETAILED DESCRIPTION

[0016] This application provides an adaptive control method and platform for the working environment of an excavator, aiming to solve the technical problems in the prior art of insufficient environmental perception and lack of intelligent early warning mechanism, which make it difficult to identify and warn of potential risks in a timely manner, thereby affecting the safety and efficiency of operations.

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] Embodiment 1, as Figure 1 As shown, the present application provides a method for adaptively controlling the working environment of an excavation equipment, the method comprising:

[0019] Step S100: Acquire basic environmental information of the working environment, evaluate the basic environmental information, call the joint sensor to collect spatial data of the working environment, and establish a spatial data set.

[0020] Specifically, the basic information of the environment is first obtained. This basic information covers many aspects of the working environment, such as the undulation of the terrain, the type of surface material, and the distribution of surrounding obstacles. The basic information of the environment obtained is evaluated. The evaluation process will analyze the complexity of the terrain, determine whether there are special terrains that affect the operation of the excavation equipment, such as steep slopes or low-lying puddles, etc.; check the impact of the surface material on the grip and stability of the excavation equipment; count the number and size of obstacles and their positional relationship in the working area, etc. According to the evaluation results, the joint sensor is called to collect spatial data of the working environment. These joint sensors include various types such as lidar sensors and visual sensors. Lidar sensors can accurately measure distance information and construct a three-dimensional outline of the working environment by emitting laser beams and receiving reflected signals. Visual sensors can capture image information of the working environment and obtain the color, texture and other features of objects. Through the collaborative work of these sensors, a large amount of spatial data is collected, and these data are integrated and processed to establish a spatial data set. This spatial data set contains detailed information on each location of the working environment, providing a rich data foundation for the subsequent three-dimensional reconstruction model establishment and other operations.

[0021] Step S200: performing key point-based positioning fusion on the spatial data set to establish a three-dimensional reconstruction model.

[0022] Specifically, we first need to clarify the concept of key point-based positioning fusion and extract information from the spatial data set to determine the key points. These key points are usually representative locations in the spatial data. They are points with special characteristics in geometry or points with important significance in data distribution. Based on the basic information of the environment, the working environment is partitioned, that is, the working environment is divided into different areas such as excavation area, mobile area and unloading area. For each area, the distance attenuation coefficient of the working environment location point is generated by using its characteristics, which means that the location points in different areas will be given different weights according to the characteristics of the area in which they are located during the positioning fusion process. Then, the location points in the working environment partition are evaluated for geometric representativeness, and the curvature value of each location point is calculated by calling the collected data of any sensor in the spatial data set. The curvature value can reflect the change of the geometry around the location point, and the geometric representativeness evaluation result is generated according to the curvature value calculation result. Then, the optimization objective function is established to comprehensively consider the distance attenuation coefficient and the geometric representativeness evaluation result, which also includes a weight parameter to balance the influence of the two. After the comprehensive evaluation value of the location point is evaluated by the optimization objective function, the spatial distribution of the location point is used for combined optimization to determine the key points. Finally, these key points are located and fused, and the information of these key points is integrated to establish a 3D reconstruction model that can accurately reflect the working environment. This model can present the spatial structure and characteristics of the working environment more intuitively and accurately.

[0023] Step S300: reading the operation task of the mining equipment, performing task execution fitting on the operation task using the three-dimensional reconstruction model, and establishing a timing attention constraint based on the task execution fitting result.

[0024] Specifically, first read the information related to the operation task of the excavation equipment. These operation tasks include specific requirements such as the target location of excavation, the depth of excavation, and the path for material handling. Then, use the constructed 3D reconstruction model to perform task execution fitting on the operation task, and simulate the operation of the excavation equipment in the virtual environment of the 3D reconstruction model according to the requirements of the operation task. For example, according to the excavation position specified in the operation task, simulate the excavation head of the excavation equipment to move to the corresponding position and perform excavation in the 3D model; according to the requirements of the material handling path, simulate the walking of the equipment and the material transfer process. Through this task execution fitting, you can observe various situations that may occur when performing the operation task in the virtual environment, such as whether there will be a collision with the surrounding environment, whether the excavation efficiency meets expectations, etc. Temporal attention constraints are established based on the results of task execution fitting. If it is found during the fitting process that the excavation equipment is prone to collide with obstacles during certain time periods, then stricter attention constraints are established for these time periods. If the excavation equipment has the highest operating efficiency under a certain time sequence, then this time sequence is used as an important temporal attention constraint condition. These temporal attention constraints will be used to guide the subsequent real-time collection of the working environment and the actual operation of the excavation equipment, ensuring that the excavation equipment can complete the task more efficiently and safely in actual operation.

[0025] Step S400: Real-time acquisition of the working environment is performed through the timing attention constraint, and the real-time acquisition data is updated to the three-dimensional reconstruction model, and constraint compensation of the real-time attention constraint is established based on the update result, and data update iteration of the three-dimensional reconstruction model is executed, and early warning is issued through the environmental warning channel built into the three-dimensional reconstruction model.

[0026] Specifically, first, the real-time collection of the working environment is carried out according to the temporal attention constraint, which means that at a specific time point or time period, the data of the relevant working environment is collected according to the previously set constraints. For example, if the temporal attention constraint indicates that the environmental changes need to be paid attention to during the time period when the excavation equipment is turning, then the data of the surrounding environment of the excavation equipment will be collected in this time period. Then, the real-time collected data is updated to the 3D reconstruction model. These new data can reflect the latest status of the working environment, such as new obstacles and slight changes in the terrain. The model will adjust and optimize the original model structure according to these new data. Then, based on the update results, the constraint compensation of the real-time attention constraint is established. If it is found that the environmental changes in a certain area are faster than expected after the update, the temporal attention constraint corresponding to this area will be compensated, such as increasing the collection frequency or expanding the scope of attention. After that, the data update iteration of the 3D reconstruction model is performed. This is a continuous process. By continuously integrating new data and adjusting the model, the 3D reconstruction model can more accurately reflect the real-time status of the working environment. Finally, early warning is issued through the environmental warning channel built into the 3D reconstruction model. When the model detects that certain key indicators are out of the safe range or abnormal conditions occur, such as the distance between the excavation equipment and the obstacle is too close, or the temperature in the working environment is too high, the environmental warning channel will issue a warning message in time to remind the operator or the automated control system to take appropriate measures to avoid potential dangers or problems.

[0027] In a possible implementation, step S200 further includes:

[0028] Step S210: Acquire the working environment partitions based on the basic environmental information, wherein the working environment partitions include an excavation area, a moving area, and an unloading area.

[0029] Step S220: Generate a distance attenuation coefficient of a working environment location point using the working environment partition.

[0030] Step S230: Perform geometric representativeness evaluation on the position points within the work environment partition and establish geometric representativeness evaluation results.

[0031] Step S240: perform spatial distribution optimization according to the distance attenuation coefficient and the geometric representativeness evaluation result, determine the key points according to the spatial distribution optimization result, and complete the positioning fusion.

[0032] Specifically, firstly, the acquired basic environmental information is deeply analyzed. These basic environmental information contain rich contents, such as the undulation of topography, the distribution of soil or materials, the location of fixed facilities in the work site and other relevant data. Through the interpretation of topographic information, if there is a region with low terrain and concentrated material accumulation, then this region is likely to be divided into the excavation area. Because such an area meets the conditions for excavation equipment to carry out excavation operations, it can efficiently obtain materials. Next, observe the data related to the equipment movement path in the basic environmental information. Those channel areas that are relatively flat, without too many obstacles and connecting various key operation areas can be determined as mobile areas. The existence of mobile areas ensures that the excavation equipment can be safely and efficiently transferred between different operation areas. Finally, the unloading area is divided according to the information related to material handling and unloading. For example, if the parking position of transport vehicles or facilities dedicated to material unloading are found around a certain area, and the area is connected to the excavation area by a reasonable material transportation path, then this area can be identified as the unloading area. Through the above analysis process based on basic environmental information, the working environment is clearly divided into three different areas: excavation area, moving area and unloading area, which provides an important regional division basis for subsequent operations.

[0033] First, it is clear that the working environment has been divided into excavation area, mobile area and unloading area. For the excavation area, the core position points of the excavation operation are given a lower distance attenuation coefficient. This is because these position points are crucial in the excavation process and have a greater impact on the accuracy of the model. The closer to the excavation center area, the smaller the distance attenuation coefficient, which means that the data of these position points have a greater weight in subsequent processing. For example, the distance attenuation coefficient of the central excavation point in the excavation area is set to 0.1, while the coefficient at the edge of the excavation area increases to 0.5. In the mobile area, the position points related to the equipment movement path are mainly considered. The distance attenuation coefficient of the position points along the path where the equipment moves frequently is relatively small. For example, the coefficient of the middle position of the main moving channel of the equipment is set to 0.3, while the coefficient of the area far away from the moving path increases to 0.7. This is because during the movement of the equipment, the data of the position points close to the path has a more critical impact on the safety and efficiency of the operation. For the unloading area, the position points near the unloading equipment and the key position points of the unloading operation are given a lower distance attenuation coefficient. For example, the distance attenuation coefficient of the position point directly below the unloading hopper is set to 0.2, while the coefficient of the position around the unloading area can be set to 0.6. This can highlight the importance of the key positions of the unloading area in the model. By generating distance attenuation coefficients according to different working environment partitions, the data of different locations can be used more specifically in subsequent processing, making the construction of the overall model more scientific and reasonable.

[0034] In the excavation area, the geometric representativeness evaluation of the position points will focus on the geometric features such as the shape and slope of the excavation surface. For example, the position points with larger curvature on the excavation surface, that is, those points with more drastic shape changes, are more geometrically representative, because these points can reflect the complex geometric changes in the excavation area. By calculating the geometric parameters such as curvature and slope of these position points, their importance in the geometric structure of the excavation area is determined, and then the geometric representativeness evaluation results for the position points in the excavation area are established. In the mobile area, the main focus is on the geometric features related to the equipment movement path. The position points at the bends and the position points at the intersections of different paths have higher geometric representativeness. Because the geometric features of these position points will have an important impact on the smoothness and safety of the equipment movement. By analyzing the relationship between these position points and the overall geometric shape of the mobile path, calculating the curvature radius of the bend, the angle of the intersection, etc., the geometric representativeness evaluation results of the position points in the mobile area are obtained. For the unloading area, the relative geometric position between the unloading equipment and the unloading area is the key. The position points close to the discharge port of the unloading equipment and within the unloading drop range have higher geometric representativeness. By analyzing the relationship between these locations and the geometric structure of the unloading equipment, such as the distance from the discharge port, the position within the range of the drop cone, etc., the geometric representative evaluation results of the unloading area location points are established. Combining the analysis of each partition, the geometric representative evaluation results of the location points in the entire working environment partition are finally formed.

[0035] First of all, we need to comprehensively consider the two important factors of distance decay coefficient and geometric representative evaluation result. For the distance decay coefficient, it reflects the importance weight of different location points in different working environment partitions, while the geometric representative evaluation result gives the criticality information of the location point from the perspective of geometric characteristics. In the process of spatial distribution optimization, these two factors are integrated, and weights are set for the distance decay coefficient and the geometric representative evaluation result respectively, and then the comprehensive evaluation value of each location point is calculated. Assuming that the weight of the distance decay coefficient is a, the weight of the geometric representative evaluation result is b, the distance decay coefficient of the location point P is C1, and the geometric representative evaluation result is C2, then the comprehensive evaluation value V of the location point P is V = a × C 1+ b×C2. All the position points are compared and screened according to these comprehensive evaluation values, and those position points with higher comprehensive evaluation values ​​are determined as key points. These key points are of great significance in the entire working environment. They are both critical in spatial position and representative in geometric features. Finally, these determined key points are located and fused, which involves integrating the coordinates of these key points in three-dimensional space and fusing their related attribute information, so as to construct a model that can more accurately reflect the characteristics of the working environment and complete the positioning fusion work based on key points.

[0036] In a possible implementation, step S230 further includes:

[0037] The collected data of any sensor in the spatial data set is called, and the curvature value of each position point is calculated through the collected data, as follows:

[0038]

[0039] Among them, C(p) is the curvature value of the position point p, k represents the total number of adjacent points of the position point p, and n p is the normal vector of position point p, n i represents the normal vector of any point in the neighborhood, and ||·|| represents the vector distance.

[0040] Generate geometric representative evaluation results based on the curvature value calculation results.

[0041] Specifically, first, the collected data of any sensor is called from the spatial data set. These collected data contain relevant information of each position point in the working environment. In this formula, C(p) is the curvature value of the position point p, which reflects the curvature of the geometric shape around the position point. k represents the total number of adjacent points of the position point p, and these adjacent points jointly participate in the calculation of the curvature value. n p is the normal vector of the position point p, which describes the directional characteristics at this point. i Represents the normal vector of any point in the neighborhood, by calculating n p With n i The vector distance between them is calculated, and the curvature value of the position point p is obtained by summing and averaging all the adjacent points. Finally, the geometric representativeness evaluation result is generated based on the calculated curvature value result. Position points with larger curvature values ​​indicate that the geometric shapes around them change more dramatically and are more representative in geometry; while position points with smaller curvature values ​​have relatively flat geometric shapes around them and are relatively less representative. By analyzing and comparing the curvature values ​​of all position points, a corresponding geometric representativeness evaluation result can be established for each position point, thereby better describing the geometric importance and characteristics of each position point in the work environment partition.

[0042] In a possible implementation, step S240 further includes:

[0043] Establish the optimization objective function as follows:

[0044] F(x)=w1·α(x)+w2·G(x).

[0045] F(x) is the comprehensive evaluation value of the position point x, α(x) is the distance attenuation coefficient, G(x) is the geometric representative evaluation result, and w1 and w2 are weight parameters.

[0046] After evaluating the comprehensive evaluation value of the location points through the optimization objective function, the spatial distribution of the location points is used for combined optimization to establish the key points.

[0047] Specifically, an optimization objective function is established, which is F(x) = w1·α(x) + w2·G(x), where F(x) represents the comprehensive evaluation value of the location point x, α(x) is the distance decay coefficient, which reflects the importance weight of the location point in the work environment partition; G(x) is the geometric representative evaluation result, which reflects the criticality of the location point in terms of geometric features; w1 and w2 are weight parameters used to adjust the relative importance of the distance decay coefficient and the geometric representative evaluation result in the comprehensive evaluation value. The comprehensive evaluation value of the location point is evaluated by the optimization objective function, and the comprehensive evaluation value of each location point is calculated according to the set weight parameters, combined with the distance decay coefficient and the geometric representative evaluation result. The spatial distribution of the location points is used for combined optimization. After calculating the comprehensive evaluation value of each location point, the spatial position relationship of the location points is considered, and combined optimization is performed to find the optimal location point combination. The key points are determined according to the results of the spatial distribution optimization, and those location points that perform outstandingly in terms of comprehensive evaluation value, spatial position, etc. are determined as key points. Complete positioning fusion, integrate and fuse the determined key points so that they can accurately reflect the characteristics and key information of the working environment, thereby realizing positioning fusion based on key points and providing an important foundation for establishing a three-dimensional reconstruction model.

[0048] In a possible implementation, step S400 further includes:

[0049] Step S410: configuring an enhanced warning layer, which is a feedback processing layer in the environmental warning channel.

[0050] Step S420: extract warning features of each environmental warning channel through the enhanced warning layer, and establish a time series superposition coefficient.

[0051] Step S430: The identification of the next round of warning channels is enhanced by using the time series superposition coefficient to complete the early warning output.

[0052] Specifically, we start to configure the enhanced warning layer. The enhanced warning layer is an important part specially set up for the environmental warning channel. It plays a key role in feedback processing in the entire warning system. The enhanced warning layer is located inside the environmental warning channel. Its main function is to receive and process the information transmitted by the warning channel. When the environmental warning channel detects the existence of danger or abnormal conditions, it will send relevant information to the enhanced warning layer. The enhanced warning layer will conduct in-depth analysis of this information to determine the level, type and possible impact of the warning. It is not just a simple reception of information, but also a series of algorithms and logical judgments to process and process the warning information to extract more valuable feedback information. For example, the enhanced warning layer will assess the severity and development trend of the danger based on factors such as the frequency, intensity and duration of the warning. At the same time, it will also consider other relevant factors, such as changes in the operating environment and the operating status of the equipment, to provide more comprehensive and accurate feedback. By configuring such an enhanced warning layer, we can better deal with various potential risks in the operating environment, make corresponding decisions and measures in a timely manner, and ensure the safe operation of the excavation equipment.

[0053] The enhanced warning layer begins to play a key role. It will perform detailed feature extraction for each warning issued by the environmental warning channel. When the environmental warning channel detects an abnormal situation and issues a warning signal, the enhanced warning layer will respond quickly. It will deeply analyze the information contained in the warning signal and extract key features such as the type of warning (for example, about excessive temperature, abnormal pressure, or other specific types of danger), the intensity of the warning (indicating the severity of the danger), and the specific time when the warning occurs. On the basis of extracting these warning features, the enhanced warning layer will further establish a time series superposition coefficient. This coefficient is established to reflect the cumulative effect of the warning on the time series. It takes into account factors such as the continuity and frequency of the warning. If warnings appear frequently or the same type of warning occurs multiple times within a period of time, the time series superposition coefficient will increase accordingly, indicating that the potential danger may be gradually accumulating or intensifying. For example, if multiple warnings about excessive temperature are received in a short period of time, the enhanced warning layer will calculate a corresponding time series superposition coefficient based on the time interval and intensity of these warnings. This coefficient can more accurately reflect the development trend of the abnormal temperature situation and provide an important basis for subsequent warning processing. In this way, the enhanced early warning layer can more comprehensively understand the meaning and potential impact of early warning information, and provide strong support for timely and effective response measures.

[0054] The time series superposition coefficient is fully utilized to enhance the recognition ability of the next round of warning channels. After the time series superposition coefficient is obtained, the warning system will use it as an important reference indicator to adjust and optimize the recognition parameters of the next round of warning channels. If the time series superposition coefficient is high, it means that the previous warning situation is more frequent or serious. Then the system will increase the sensitivity of the warning channel accordingly and lower the threshold of the warning, so that the warning channel can more easily detect potential danger signals. For example, shorten the time interval of warning detection, or expand the scope of warning detection to ensure that possible dangerous situations can be discovered in time. At the same time, the time series superposition coefficient is also used to improve the warning algorithm so that it can better identify and judge complex dangerous patterns. Through the analysis and learning of historical warning data, combined with the trend reflected by the time series superposition coefficient, the warning algorithm can more accurately predict the development direction of potential dangers and issue warning signals in advance. After completing these adjustments and optimizations, the warning system will continue to monitor the operating environment in real time. Once a signal that meets the warning conditions is detected, an early warning report is immediately issued to convey the relevant information to the operator or the relevant control center in a timely manner. In this way, through the application of the time series superposition coefficient, the recognition ability of the early warning channel can be continuously enhanced, and the timeliness and accuracy of the early warning can be improved, thereby better ensuring the safety of the operating environment of the excavation equipment and effectively preventing the occurrence of potential dangers.

[0055] In a possible implementation, step S400 further includes:

[0056] Step S440: Establish a shovel tooth data set of the excavation equipment, and monitor the shovel tooth status through real-time monitoring equipment to generate an updated data set.

[0057] Step S450: identifying the change of the state of the shovel teeth on the updated data set, and providing a life warning based on the result of the identification of the change of the state of the shovel teeth.

[0058] Specifically, we first focus on establishing a shovel tooth dataset for excavation equipment. This dataset is established to comprehensively record the relevant information of the shovel teeth, including but not limited to basic information such as the model, material, and installation location of the shovel teeth, as well as historical data of the shovel teeth in previous use, such as usage time, degree of wear, and maintenance records. In order to grasp the status of the shovel teeth in real time, special real-time monitoring equipment is installed on the excavation equipment. These monitoring devices include various sensors, such as pressure sensors, displacement sensors, and image sensors. The pressure sensor can monitor the pressure on the shovel teeth during work, the displacement sensor can monitor the displacement change of the shovel teeth, and the image sensor can take images of the shovel teeth to obtain information about their appearance. These real-time monitoring devices will continuously collect the status data of the shovel teeth and transmit these data back to the system in real time. The data will be processed and analyzed, and compared and integrated with the historical data in the shovel tooth dataset to generate an updated dataset. This updated dataset can accurately reflect the current status of the shovel teeth, including their degree of wear, whether there is damage or deformation, etc. By establishing a shovel tooth dataset and performing real-time monitoring, the status changes of the shovel teeth can be understood in a timely manner, providing reliable data support for the subsequent identification of shovel tooth status changes and life warning, thereby ensuring the normal operation and work efficiency of the excavation equipment.

[0059] Convolutional neural network (CNN) is used to identify the state change of shovel teeth in the updated data set, and life warning is performed based on this. The specific process is as follows: First, data preparation is performed to collect a large amount of image data containing different states of shovel teeth. These images cover various situations such as normal state, wear state, crack state, etc. At the same time, these images are annotated in detail to clearly mark the state category of the shovel teeth. Subsequently, the data set is divided into training set, validation set and test set for subsequent model training and evaluation. Build a CNN model and select a CNN architecture suitable for the shovel tooth state recognition task, which usually includes convolution layer, pooling layer and fully connected layer. When determining the hyperparameters of the model, factors such as convolution kernel size, number and learning rate need to be considered. The selection of these hyperparameters will have an important impact on the performance of the model. Then, the model is trained, the training set data is input into the CNN model, and iterative training begins. During the training process, the loss function of the model is calculated, which is used to measure the difference between the model prediction result and the actual label. Through the back propagation algorithm, the parameters of the model are updated according to the gradient of the loss function to gradually optimize the model so that it can better fit the data. At the same time, the validation set data is used to monitor the training process, and the hyperparameters are adjusted in time to avoid overfitting of the model. After the training is completed, the model is evaluated. The trained model is tested using the test set data, and indicators such as accuracy and recall are calculated to evaluate the performance of the model in practical applications. In the actual recognition of the state change of the shovel tooth, the shovel tooth image collected by the real-time monitoring equipment is input into the trained CNN model. The model will automatically extract and classify the input image and output the state category prediction result of the shovel tooth. Finally, the life warning is carried out according to the result of the recognition of the state change of the shovel tooth. Combined with historical data and preset life prediction rules, if the shovel tooth is identified to be in a serious wear state for many consecutive times and the wear degree continues to increase, a life warning signal will be issued to remind relevant personnel to take timely measures, such as replacing the shovel tooth or performing maintenance, to ensure the normal operation and safe production of the mining equipment. By using the CNN algorithm, the information in the image data is fully utilized to achieve accurate recognition of the state change of the shovel tooth and life warning, providing strong support for the maintenance and management of mining equipment.

[0060] In a possible implementation, step S400 further includes:

[0061] Step S460: Identify the volume of the excavated material of the excavation equipment, and establish an additional risk factor based on the material volume identification result.

[0062] Step S470: After performing channel compensation for the environmental warning channel by using the additional risk factor, a warning alarm is issued.

[0063] Specifically, the volume of excavated materials is identified. First, the volume of materials excavated by the excavation equipment is identified using relevant volume measurement technology. This involves optical sensors, laser scanners and algorithms based on image analysis. Through these technical means, data can be collected from different angles, and then the volume of the materials can be accurately determined through complex calculations. For example, the image analysis algorithm processes the captured material image, identifies the pixel ratio of the material in the image, and then calculates the approximate volume of the material in combination with the known image scale information. After obtaining the material volume identification result, the additional risk factor is established. If the material volume is small, it is relatively easy to operate in the transportation and processing links, and the risk to the equipment and the entire operation process is low. At this time, the additional risk factor is small. However, when the material volume is identified as large, the situation becomes complicated. The introduction of bulk material identification technology plays a key role here. When the material is identified as a bulk material, it means that it is more likely to cause problems during the operation. For example, when a bulk material is transported on a conveyor belt, its volume and weight may cause the conveyor belt to be overloaded or even damaged, causing blockage in some narrow channels, affecting the continuity of the entire operation. In this case, the additional risk factor will be greatly increased according to the size of the bulk material and the degree to which it may cause blockage and equipment damage. Panoramic monitoring fused with AI analysis provides comprehensive protection for the entire operating environment in this step. In the process of identifying the material volume and establishing the additional risk factor, the panoramic monitoring system monitors the entire operating scene in real time. AI analysis analyzes the material status, equipment operating status, and operator operation in the monitoring screen. When potential dangers caused by the large volume of materials are found, such as excessive material accumulation or abnormal vibration of equipment operation (possibly due to the influence of large pieces of materials), an immediate alarm will be issued, which not only helps to adjust the operating process in time to prevent the occurrence of danger, but also can effectively improve the safety and efficiency of operations.

[0064] The additional risk factor reflects the risk level of the excavated material. When this factor is high, it means that the current excavated material has a greater risk of hidden dangers, such as the excessive volume of the material, which may easily lead to equipment failure or blockage. By applying this factor to the compensation process of the environmental warning channel, the sensitivity and response range of the warning system can be adjusted. If the additional risk factor is large, the warning threshold of the environmental warning channel will be reduced accordingly, and the warning range will be expanded. In this way, potential dangerous situations can be detected earlier and early warnings can be issued in advance. After completing the channel compensation, the early warning alarm operation is performed. The early warning alarm can be performed in a variety of ways, such as displaying eye-catching warning information on the operation interface, issuing sound and light alarm signals, or sending early warning notifications to the mobile devices of relevant staff. These warning messages will detail the current dangerous situation, such as the danger caused by the excessive volume of the material, and the approximate degree of the danger and other related information. Through such steps, the additional risk factor can be more effectively used to improve the accuracy and timeliness of environmental warnings, ensure the safety of the excavation process, and avoid equipment damage and operation stagnation caused by material-related risks.

[0065] Embodiment 2, based on the same inventive concept as the method for adaptively controlling the working environment of an excavating device in the above embodiment, Figure 2 As shown, the present application provides an adaptive control platform for the working environment of an excavation equipment. The platform and the method embodiments in the present application are based on the same inventive concept. The platform includes:

[0066] The spatial data set establishment module 10 is used to obtain basic environmental information of the working environment, and after evaluating the basic environmental information, call the joint sensor to collect spatial data of the working environment to establish a spatial data set.

[0067] The three-dimensional reconstruction model building module 20 is used to perform key point-based positioning fusion on the spatial data set to build a three-dimensional reconstruction model.

[0068] The timing attention constraint establishing module 30 is used to read the operation task of the mining equipment, perform task execution fitting on the operation task using the three-dimensional reconstruction model, and establish the timing attention constraint based on the task execution fitting result.

[0069] The early warning module 40 is used to perform real-time acquisition of the working environment through the timing attention constraint, and update the real-time acquisition data to the three-dimensional reconstruction model, and establish constraint compensation of the real-time attention constraint with the update result, execute data update iteration of the three-dimensional reconstruction model, and issue early warning through the environmental warning channel built into the three-dimensional reconstruction model.

[0070] Furthermore, the three-dimensional reconstruction model building module 20 also includes:

[0071] A working environment partition acquisition unit is used to acquire working environment partitions based on the basic environment information, wherein the working environment partitions include an excavation area, a moving area, and an unloading area.

[0072] A distance attenuation coefficient generating unit is used to generate a distance attenuation coefficient of a working environment location point by using the working environment partition.

[0073] A geometric representative evaluation result establishing unit is used to perform geometric representative evaluation on the position points within the working environment partition and establish a geometric representative evaluation result.

[0074] The positioning fusion unit is used to perform spatial distribution optimization according to the distance attenuation coefficient and the geometric representative evaluation result, determine the key points according to the spatial distribution optimization result, and complete the positioning fusion.

[0075] Furthermore, the geometric representative evaluation result establishing unit further includes:

[0076] The collected data of any sensor in the spatial data set is called, and the curvature value of each position point is calculated through the collected data, as follows:

[0077]

[0078] Among them, C(p) is the curvature value of the position point p, k represents the total number of adjacent points of the position point p, and n p is the normal vector of position point p, n i represents the normal vector of any point in the neighborhood, and ||·|| represents the vector distance.

[0079] Generate geometric representative evaluation results based on the curvature value calculation results.

[0080] Furthermore, the positioning fusion unit also includes:

[0081] Establish the optimization objective function as follows:

[0082] F(x)=w1·α(x)+w2·G(x).

[0083] F(x) is the comprehensive evaluation value of the position point x, α(x) is the distance attenuation coefficient, G(x) is the geometric representative evaluation result, and w1 and w2 are weight parameters.

[0084] After evaluating the comprehensive evaluation value of the location points through the optimization objective function, the spatial distribution of the location points is used for combined optimization to establish the key points.

[0085] Furthermore, the early warning module 40 further includes:

[0086] An enhanced warning layer configuration unit is used to configure an enhanced warning layer, which is a feedback processing layer in an environmental warning channel.

[0087] A time series superposition coefficient establishing unit is used to extract the warning features of each environmental warning channel through the enhanced warning layer and establish a time series superposition coefficient.

[0088] The early warning channel enhancement unit is used to enhance the recognition of the next round of early warning channels through the time series superposition coefficient to complete the early warning alarm.

[0089] Furthermore, the early warning module 40 further includes:

[0090] The update data set generation unit is used to establish a shovel tooth data set of the excavation equipment, and monitor the shovel tooth status through a real-time monitoring device to generate an update data set.

[0091] A life warning unit is used to identify the change of the state of the shovel teeth on the updated data set, and to issue a life warning based on the result of the identification of the change of the state of the shovel teeth.

[0092] Furthermore, the early warning module 40 further includes:

[0093] The additional risk factor establishment unit is used to identify the volume of excavated materials of the excavation equipment and establish the additional risk factor based on the material volume identification result.

[0094] The early warning execution unit is used to execute the early warning after performing channel compensation of the environmental early warning channel through the additional risk factor.

[0095] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. 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.

[0096] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0097] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A method for adaptively controlling the working environment of an excavation equipment, characterized in that: The method comprises: Obtaining basic environmental information of the working environment, evaluating the basic environmental information, calling the joint sensor to collect spatial data of the working environment, and establishing a spatial data set; Performing key point-based positioning fusion on the spatial data set to establish a three-dimensional reconstruction model; Reading the operation task of the mining equipment, performing task execution fitting on the operation task using the three-dimensional reconstruction model, and establishing a timing attention constraint based on the task execution fitting result; The operating environment is collected in real time through the time-series attention constraint, and the real-time collected data is updated to the three-dimensional reconstruction model, and the constraint compensation of the real-time attention constraint is established with the update result, and the data update iteration of the three-dimensional reconstruction model is executed, and the early warning alarm is issued through the environmental early warning channel built into the three-dimensional reconstruction model; The performing key point-based positioning fusion on the spatial data set to establish a three-dimensional reconstruction model also includes: Acquire the working environment partition based on the basic environment information, wherein the working environment partition includes an excavation area, a moving area, and an unloading area; Generating distance attenuation coefficients of working environment location points using the working environment partitions; Performing geometric representativeness evaluation on the position points within the working environment partition and establishing geometric representativeness evaluation results; The spatial distribution optimization is performed according to the distance attenuation coefficient and the geometric representative evaluation results, and the key points are determined according to the spatial distribution optimization results to complete the positioning fusion.

2. The method for adaptively controlling the working environment of an excavation equipment according to claim 1, characterized in that: The performing geometric representativeness evaluation on the position points within the working environment partition and establishing the geometric representativeness evaluation result also includes: The collected data of any sensor in the spatial data set is called, and the curvature value of each position point is calculated through the collected data, as follows: Among them, C(p) is the curvature value of the position point p, k represents the total number of adjacent points of the position point p, and n p is the normal vector of position point p, n i represents the normal vector of any point in the neighborhood, and ||·|| represents the vector spacing; Generate geometric representative evaluation results based on the curvature value calculation results.

3. The method for adaptively controlling the working environment of an excavation equipment according to claim 1, characterized in that: The performing spatial distribution optimization according to the distance attenuation coefficient and the geometric representativeness evaluation result, and determining the key point according to the spatial distribution optimization result, further includes: Establish the optimization objective function as follows: F(x)=w1·α(x)+w2·G(x); F(x) is the comprehensive evaluation value of the position point x, α(x) is the distance attenuation coefficient, G(x) is the geometric representative evaluation result, and w1 and w2 are weight parameters; After evaluating the comprehensive evaluation value of the location points through the optimization objective function, the spatial distribution of the location points is used for combined optimization to establish the key points.

4. The method for adaptively controlling the working environment of an excavation equipment according to claim 1, characterized in that: The early warning alarm is issued through the environmental early warning channel built into the three-dimensional reconstruction model, and further includes: Configuring an enhanced warning layer, which is a feedback processing layer in the environmental warning channel; Extract the warning features of each environmental warning channel through the enhanced warning layer and establish a time series superposition coefficient; The time series superposition coefficient is used to enhance the recognition of the next round of warning channels to complete the early warning alarm.

5. The method for adaptively controlling the working environment of an excavation equipment according to claim 1, characterized in that: The method further comprises: Establish a shovel tooth data set for excavation equipment, monitor the shovel tooth status through real-time monitoring equipment, and generate an updated data set; The updated data set is used to identify the change in the state of the shovel teeth, and a life warning is issued based on the result of the identification of the change in the state of the shovel teeth.

6. The method for adaptively controlling the working environment of an excavating device according to claim 1, characterized in that: The method further comprises: Identify the volume of excavated materials of the excavation equipment and establish an additional risk factor based on the material volume identification results; After channel compensation of the environmental warning channel is performed by using the additional risk factor, a warning alarm is issued.

7. An adaptive control platform for the working environment of an excavation equipment, characterized in that: The platform includes: A spatial data set establishment module, which is used to obtain basic environmental information of the working environment, and after evaluating the basic environmental information, call a joint sensor to collect spatial data of the working environment to establish a spatial data set; A three-dimensional reconstruction model building module, the three-dimensional reconstruction model building module is used to perform key point-based positioning fusion on the spatial data set to build a three-dimensional reconstruction model; A timing attention constraint establishment module, the timing attention constraint establishment module is used to read the operation task of the mining equipment, perform task execution fitting on the operation task using the three-dimensional reconstruction model, and establish a timing attention constraint based on the task execution fitting result; A pre-alarm output module, the pre-alarm output module is used to perform real-time acquisition of the operating environment through the temporal attention constraint, and update the real-time acquisition data to the three-dimensional reconstruction model, and establish constraint compensation of the real-time attention constraint with the update result, execute data update iteration of the three-dimensional reconstruction model, and issue a pre-alarm through the environmental warning channel built into the three-dimensional reconstruction model; The three-dimensional reconstruction model building module also includes: An operating environment partition acquisition unit, wherein the operating environment partition acquisition unit acquires operating environment partitions based on the basic environment information, wherein the operating environment partitions include an excavation area, a moving area, and an unloading area; A distance attenuation coefficient generating unit, the distance attenuation coefficient generating unit is used to generate a distance attenuation coefficient of a working environment location point by using the working environment partition; A geometric representative evaluation result establishing unit, the geometric representative evaluation result establishing unit is used to perform geometric representative evaluation on the position points within the working environment partition and establish a geometric representative evaluation result; The positioning fusion unit is used to perform spatial distribution optimization according to the distance attenuation coefficient and the geometric representative evaluation result, determine the key points according to the spatial distribution optimization result, and complete the positioning fusion.

Images