Asynchronous data processing method, device, system and storage medium for coal pile

Abstract

The present application discloses an asynchronous data processing method, device and system for coal piles, the method comprising: obtaining three-dimensional point cloud data of the current coal pile; inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile; based on the three-dimensional space size and the three-dimensional layout coordinate value, obtaining the storage capacity of the current coal pile through a preset first weight model; obtaining multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage capacities of multiple current coal piles according to a time series, and constructing a comprehensive data set; inputting the comprehensive data set into a prediction model based on a time series to obtain the predicted three-dimensional space size of the target coal pile, the three-dimensional layout coordinate value of the target coal pile and the storage capacity of the target coal pile. The present application can accurately predict the volume, spatial position and storage capacity of the target coal pile, help managers adjust management strategies, and improve the operational efficiency of coal pile management.

Description

Technical Field

[0001] The present application relates to the technical field of data processing, and discloses an asynchronous data processing method, device, system and storage medium for a coal pile. Background Art

[0002] With the acceleration of industrialization, the storage, transportation and monitoring requirements of materials such as coal are increasing in the material yard management. As a storage and management place for combustible materials such as coal, the material yard plays a vital role in the entire fuel supply chain. For material yard managers, it is crucial to accurately predict the changes in the volume, spatial location and storage volume of coal piles in the future, because this information is directly related to the efficiency, safety and cost control of material yard operations. Traditional coal pile management methods usually rely on manual observation and statistical analysis, and there are problems such as untimely data collection and inaccurate analysis, which cannot meet the urgent needs of real-time and efficient management. Managers often face challenges such as the inability to accurately grasp the changes in the coal pile situation, difficulty in adjusting management strategies in a timely manner, and inability to effectively prevent the risk of spontaneous combustion. Summary of the invention

[0003] The present application provides an asynchronous data processing method, device, system and storage medium for coal piles to solve the technical problems in related technologies such as untimely collection of coal pile data, inaccurate analysis, and difficulty in real-time and efficient management.

[0004] In a first aspect, the present application provides an asynchronous data processing method for a coal pile, comprising:

[0005] Acquire the three-dimensional point cloud data of the current coal pile based on the set laser sensor;

[0006] Inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, wherein the three-dimensional space size represents the volume of the current coal pile, and the three-dimensional layout coordinate value represents the spatial layout position of the current coal pile;

[0007] Based on the size of the three-dimensional space and the three-dimensional layout coordinate values, obtaining the storage capacity of the current coal pile through a preset first weight model;

[0008] According to the time series, multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage volumes of multiple current coal piles are obtained to construct a comprehensive data set based on the time series information corresponding to the time series;

[0009] The comprehensive data set is input into a prediction model based on a time series to obtain the predicted three-dimensional spatial size of the target coal pile, the three-dimensional layout coordinate value of the target coal pile, and the storage capacity of the target coal pile.

[0010] Wherein, the prediction model based on time series includes:

[0011] ,

[0012] in, represents the predicted value corresponding to time t, and is the model correlation parameter, is the variance of the full data set at time t, is the density function of the forecasting model based on time series, is the autoregressive coefficient, which is used to measure and The linear relationship strength between them, j is the lag order of the prediction model based on time series, is the penalty factor for the time series-based forecasting model, represents the comprehensive data set at time t, represents the comprehensive dataset at time tj, represents the change of the comprehensive data set at time t, that is, the first-order difference sequence of the comprehensive data set at time t, Represents the change in the comprehensive data set at time tj.

[0013] Furthermore, in the time series-based prediction model,

[0014] ,

[0015] in, represents the error factor at time t, represents the lag factor of lag order j,

[0016] The autoregressive coefficient By updating the lag order j, we solve

[0017] If the autoregressive coefficient , then the time series is stationary and the comprehensive data set is not updated.

[0018] If the autoregressive coefficient , then the time series is non-stationary, through the autoregressive coefficient Update the comprehensive dataset at time t , get the comprehensive data set at time t The amount of change ,

[0019] Among them, through the following formula, through the autoregressive coefficient Update the comprehensive dataset at time t :

[0020] ,

[0021] .

[0022] Furthermore, the model correlation parameter and Determined by the error function, the expression of the error function is,

[0023] ,

[0024] in, represents the error function,

[0025] By the error function Derivatives to update model correlation parameters and , the derivation process is as follows:

[0026] ,

[0027] ,

[0028] Among them, the model correlation parameter and The initial value of is 1,

[0029] When the error function When the minimum, the best model correlation parameter is obtained and .

[0030] Furthermore, the three-dimensional point cloud data of the current coal pile is input into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, including:

[0031] Inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system;

[0032] The three-dimensional laser scanning mobile measurement system obtains the map features and / or wall features of the current coal pile based on the three-dimensional point cloud data of the current coal pile.

[0033] Map information of the current coal pile is constructed based on the map features and / or wall features, and the three-dimensional space size and three-dimensional layout coordinate values ​​of the current coal pile are obtained in the map information.

[0034] Furthermore, after obtaining multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage capacities of multiple current coal piles according to the time series and constructing a comprehensive data set based on the time series information corresponding to the time series, it also includes:

[0035] The comprehensive data set is input into a preset curvature model, and curvature calculation is performed on all data in the comprehensive data set to obtain multiple curvature values ​​of the comprehensive data set data.

[0036] When the curvature value of some data in the comprehensive data set is greater than a preset critical curvature value, the some data are removed from the comprehensive data set:

[0037] Wherein, the preset curvature model is:

[0038] ,

[0039] in, and is the eigenvalue of the second-order derivative matrix, that is, the eigenvalue of the second-order derivative matrix of the surface area formed by the comprehensive data set, is the preset critical curvature value, P1 is the new comprehensive data set obtained by removing some data in the comprehensive data set, P represents the comprehensive data set, and p i represents the i-th subset in the full dataset.

[0040] Furthermore, in the preset curvature model, and Obtained through the following process:

[0041] For a point (x, y) in the full data set, a surface S is defined in the local neighborhood of the point (x, y). The first-order derivative matrix is ​​the normal vector of the tangent plane of the surface S at the point (x, y).

[0042] ,

[0043] in, is the first-order derivative matrix,

[0044] Based on the first-order derivative matrix, calculate the second-order derivative matrix:

[0045] ,

[0046] Where H is the second-order derivative matrix, , , , is the second-order derivative of the surface S with respect to x or y,

[0047] Eigenvalues ​​of the second-order derivative matrix H and Obtained by the following formula:

[0048] ,

[0049] ,

[0050] get:

[0051] ,

[0052] ,

[0053] = ,

[0054] Where I is the identity matrix.

[0055] Furthermore, the method further comprises:

[0056] Obtaining the layout environment image of the material yard where the current coal pile is located, and the layout position image of the current coal pile;

[0057] Get the current temperature value of the coal pile;

[0058] According to the layout environment image of the material yard, the layout position image of the coal pile and the current temperature value of the coal pile, spontaneous combustion fire alarm warning information is output and displayed.

[0059] In a second aspect, the present application also provides an asynchronous data processing device for a coal pile, comprising:

[0060] A point cloud data acquisition module is used to acquire three-dimensional point cloud data of the current coal pile based on a set laser sensor;

[0061] A scanning and measuring module, used for inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, wherein the three-dimensional space size represents the volume of the current coal pile, and the three-dimensional layout coordinate value represents the spatial layout position of the current coal pile;

[0062] A weight calculation module, used for obtaining the storage capacity of the current coal pile through a preset first weight model based on the size of the three-dimensional space and the three-dimensional layout coordinate value;

[0063] A data set construction module is used to obtain multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage volumes of multiple current coal piles according to the time series, and to construct a comprehensive data set based on the time series information corresponding to the time series;

[0064] A target value output module is used to input the comprehensive data set into a prediction model based on a time series to obtain the predicted three-dimensional space size of the target coal pile, the three-dimensional layout coordinate value of the target coal pile, and the storage capacity of the target coal pile.

[0065] Wherein, the prediction model based on time series includes:

[0066] ,

[0067] in, represents the predicted value corresponding to time t, and is the model correlation parameter, is the variance of the full data set at time t, is the density function of the forecasting model based on time series, is the autoregressive coefficient, which is used to measure and The linear relationship strength between them, j is the lag order of the prediction model based on time series, is the penalty factor for the time series-based forecasting model, represents the comprehensive data set at time t, represents the comprehensive dataset at time tj, represents the change of the comprehensive data set at time t, that is, the first-order difference sequence of the comprehensive data set at time t, Represents the change in the comprehensive data set at time tj.

[0068] In a third aspect, the present application also provides an asynchronous data processing system for a coal pile, comprising: one or more processors and a memory;

[0069] The memory is used to store one or more programs;

[0070] When the one or more programs are executed by the one or more processors, the one or more processors implement the asynchronous data processing method for the coal pile as described in the first aspect.

[0071] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the asynchronous data processing method for a coal pile as described in the first aspect.

[0072] The present invention proposes a data processing technology based on a prediction model of a time series, which can build an accurate prediction model through historical data analysis and prediction algorithms, and obtain the accurate predicted volume of the target coal pile, the predicted spatial layout position information of the target coal pile, and the predicted storage volume of the target coal pile in real time, so as to help managers accurately predict the future coal pile situation. The material yard manager can optimize the coal pile layout, adjust the storage strategy, and reasonably arrange the procurement and sales plan according to the changes in the coal pile situation, so as to improve the operation efficiency of the material yard and reduce costs. The present invention aims to help managers optimize management decisions and promote the development of intelligent and digital material yard management through precise monitoring and prediction technology. For the time series prediction model of coal pile volume, spatial position and storage volume, considering the temporal relationship and trend between data, a suitable time series prediction algorithm is established to predict the changes in the volume, spatial position and storage volume of coal piles at future moments. According to the prediction results, managers are assisted in making operational decisions and resource allocation, optimizing the operation efficiency of the material yard and reducing costs.

[0073] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0075] Figure 1 The present invention is a schematic diagram of the structure of an asynchronous data processing system for a coal pile provided according to an embodiment of the present application.

[0076] Figure 2 The present invention is a flowchart of an asynchronous data processing method for a coal pile provided according to an embodiment of the present application.

[0077] Figure 3 The present invention is a module diagram of an asynchronous data processing device for a coal pile provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0078] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only the parts related to the present application, rather than all structures, are shown in the accompanying drawings.

[0079] The asynchronous data processing method for coal piles provided in the embodiment of the present application is executed on a coal pile data processing system. The coal pile data processing system can be implemented by software and / or hardware, and the coal pile data processing system can be composed of two or more physical entities, or can be composed of one physical entity. In one embodiment of the present application, the coal pile data processing system can be a computer (computer) with data processing function, a server connected to sensors, cameras and other external devices.

[0080] Figure 1 The structure diagram of an asynchronous data processing system for coal piles provided according to an embodiment of the present application is shown in FIG. Figure 1 The asynchronous data processing system for coal piles at least includes a processor 11 and a memory 12. The processor 11 and the memory 12 may be connected via a bus or other means.

[0081] The number of processors 11 is one or more. Figure 1 A processor 11 is taken as an example. The processor 11 may include processing units such as an application processor (AP), a graphics processing unit (GPU), and a central processing unit (CPU).

[0082] The memory 12, as a computer-readable storage medium, can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the asynchronous data processing system for coal piles in the embodiment of the present application. The memory 12 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the asynchronous data processing system for coal piles, etc. In addition, the memory 12 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage devices. In some instances, the memory 12 may further include a memory 12 remotely arranged relative to the processor 11, and these remote memories may be connected to the asynchronous data processing system for coal piles through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0083] In addition, the asynchronous data processing system for the coal pile may include, in addition to the processor 11 and the memory 12, one or more components such as a display screen, a component for accessing a network (such as the Internet), a communication interface, a power supply, a speaker, a camera, an input device, etc., which are not limited in the embodiment.

[0084] In this article, coal piles, coal and other fuel storage media have equivalent meanings to material yards and will not be elaborated here.

[0085] In one embodiment, at least one camera and at least one sensor are arranged at different heights and positions of the material yard (or coal pile), and these cameras are used to monitor the volume and spatial position changes of the coal pile; the sensor can be set in the coal pile to monitor the storage capacity, weight, ambient temperature and humidity of the coal pile in real time.

[0086] In one embodiment, after acquiring the coal pile related parameter data from the camera and the sensor, the coal pile related parameter data is transmitted to the computing center processing unit or the cloud through the wireless communication interface for processing. After preliminary processing and screening of the data by the computing center processing unit, the data is transmitted to the cloud equipped with a large neural network model for deep processing to obtain parameters such as the predicted volume of the target coal pile, the predicted spatial layout position information of the target coal pile, and the predicted storage capacity of the target coal pile.

[0087] The asynchronous data processing system for the coal pile executes the asynchronous data processing method for the coal pile. Figure 2 This is a flowchart of an asynchronous data processing method for a coal pile provided according to an embodiment of the present application, with reference to Figure 2 The asynchronous data processing method for the coal pile includes steps 210 to 250.

[0088] Step 210: Acquire the three-dimensional point cloud data of the current coal pile based on the set laser sensor.

[0089] Specifically, in one embodiment of the present application, a material yard where a coal pile is placed is provided with many sensors and cameras, and the sensors and cameras are arranged at different heights and positions in the material yard; wherein the cameras are used to capture the layout environment image of the material yard, as well as the layout position image of the coal pile and other environmental information. The sensors include temperature sensors and humidity sensors, which are arranged in the coal pile or other places in the material yard to obtain the temperature and humidity information in the material yard or the coal pile. In addition, in one embodiment, a laser sensor is also arranged in the material yard (or a position related to the coal pile), and the laser sensor is used to obtain the three-dimensional point cloud data (X, Y, Z) of the current coal pile, that is, the three-dimensional spatial position information of the coal pile.

[0090] Step 220: Input the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional spatial size and three-dimensional layout coordinate value of the current coal pile, wherein the three-dimensional spatial size represents the coal pile volume of the current coal pile, and the three-dimensional layout coordinate value represents the spatial layout position of the current coal pile.

[0091] In one embodiment of the present application, a high-precision three-dimensional laser scanning mobile measurement system GoSLAM (Generalized Online Simultaneous Localization and Mapping) can be used. The core of the algorithm performs positioning and builds an environmental map based on the data scanned by the laser sensor. It mainly includes extracting key features from the acquired three-dimensional point cloud data, such as map features or wall features, etc., for building a map of the environment and positioning. Then, a more accurate three-dimensional space size and three-dimensional layout coordinate value of the current coal pile are obtained from the map. Specifically, the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile can be displayed on a display screen.

[0092] In addition, it should be noted that the acquired three-dimensional point cloud data can be preprocessed (preprocessing can include filtering and denoising) to obtain denoised three-dimensional point cloud data to improve data quality and accuracy. Further, the denoised three-dimensional point cloud data is analyzed through the above process. In order to distinguish it from the processing below, the preprocessing here can be referred to as the first preprocessing.

[0093] Furthermore, the denoised three-dimensional point cloud data can be updated in real time, so as to match and optimize the previously obtained map and positioning information, so as to improve the output accuracy of the entire three-dimensional laser scanning mobile measurement system GoSLAM, and obtain more accurate three-dimensional spatial size and three-dimensional layout coordinate values ​​of the current coal pile from the map.

[0094] Step 230: Based on the three-dimensional space size and the three-dimensional layout coordinate values, the storage capacity of the current coal pile is obtained through a preset first weight model.

[0095] Specifically, the first weight model may be obtained by training based on historical data. Specifically, in one embodiment, the core algorithm of the first weight model may be implemented using a volume method or a discrete method.

[0096] In one embodiment, the volume method estimates the storage weight of the material by calculating the volume of space occupied by the material. Assuming that the material (coal) is evenly distributed, the following formula can be used to calculate the current material storage volume: Storage volume = material volume × material density, where the material volume can be calculated by the three-dimensional space size and three-dimensional layout coordinate value of the material, and the material density can be pre-set according to the type and properties of the material. In this application, the material can be a substance such as coal, and the state of the material is not limited to solid, but can also be liquid and gaseous, but not limited to this.

[0097] In one embodiment, the discrete method estimates the storage weight by discretizing the space and calculating the weight of the coal at each discrete point. The specific steps are as follows: according to the size of the three-dimensional space and the three-dimensional layout coordinate value, the three-dimensional space is divided into discrete voxels or grids, and the volume of each voxel or grid is calculated respectively; then, according to the distribution of the material in the voxel or grid, the weight of the material in the voxel or grid is calculated, and finally, the weight of the material in all voxels or grids is summed to obtain the current storage capacity of the coal pile.

[0098] Step 240: Acquire multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage capacities of multiple current coal piles according to the time series, and construct a comprehensive data set based on the time series information corresponding to the time series.

[0099] Specifically, according to the time series, the three-dimensional space size, three-dimensional layout coordinate value and the storage capacity of the current coal pile corresponding to multiple time points are obtained. The three-dimensional space size, three-dimensional layout coordinate value and the storage capacity of the current coal pile corresponding to the collected multiple time points are preprocessed (here can be referred to as the second preprocessing), including data cleaning, removing outliers, filling missing values ​​and other operations to ensure the quality and integrity of the comprehensive data set. According to the order of the time series, the three-dimensional space size, three-dimensional layout coordinate value and the storage capacity of the current coal pile at each time point are combined into a data sample. Each data sample can be represented as a set containing the three-dimensional space size, three-dimensional layout coordinate value and the storage capacity of the current coal pile.

[0100] Furthermore, in order to retain the information of the time series, it is necessary to add corresponding time series information to each data sample. The order of the time series can be used as the time series information, or the time series information of each data sample can be marked using a timestamp or other means. The three-dimensional space size, three-dimensional layout coordinate value, current storage capacity of the coal pile and the corresponding time series information of each data sample are combined into a comprehensive data set. Through the above steps, a comprehensive data set based on the time series information corresponding to the time series can be constructed. This data set contains the three-dimensional space size, three-dimensional layout coordinate value and current storage capacity of the coal pile at multiple time points, and retains the time series information. This data set can be used for subsequent analysis and modeling, such as trend analysis based on time series, construction of prediction models, etc.

[0101] Exemplarily, in one embodiment of the present application, example data of a comprehensive data set is as follows:

[0102] ,

[0103] In the example data, in January 2023, the three-dimensional space size representing the volume of the coal pile is 10,000 cubic meters, the three-dimensional layout coordinate value representing the spatial position is (100, 200, 5) meters, and the current storage capacity of the coal pile is 5,000 tons.

[0104] Furthermore, multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage capacities of multiple current coal piles are obtained according to the time series, and a comprehensive data set based on the timing information corresponding to the time series is constructed. The method also includes preprocessing the three-dimensional space sizes, three-dimensional layout coordinate values ​​and current storage capacities of the coal piles corresponding to the collected multiple time points, including data cleaning, removal of outliers, filling in missing values ​​and other operations, so as to ensure the quality and integrity of the comprehensive data set.

[0105] Further, after acquiring multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage capacities of multiple current coal piles according to the time series and constructing a comprehensive data set based on the time series information corresponding to the time series, the method further includes:

[0106] The comprehensive data set is input into a preset curvature model, and curvature calculation is performed on all data in the comprehensive data set to obtain multiple curvature values ​​of the comprehensive data set data.

[0107] When the curvature value of some data in the comprehensive data set is greater than a preset critical curvature value, the some data are removed from the comprehensive data set:

[0108] Wherein, the preset curvature model is:

[0109] ,

[0110] in, and is the eigenvalue of the second-order derivative matrix, that is, the eigenvalue of the second-order derivative matrix of the surface area formed by the comprehensive data set, is the preset critical curvature value, P1 is the new comprehensive data set obtained by removing some data in the comprehensive data set, P represents the comprehensive data set, and p i represents the i-th subset in the full dataset.

[0111] Specifically, the present application removes those points with abnormally large curvature changes by collecting the curvature information of the point cloud data in the comprehensive data set. The curvature can reflect the curvature of the point cloud surface, and noise points often have abnormally high curvature values. Curvature-based point cloud pre-filtering is a method that uses the geometric characteristics of point cloud data to identify and remove noise, and can more accurately identify abnormal points that are inconsistent with the surrounding surface, such as noise points or outliers. Curvature is a local attribute, so this method has good robustness to local noise and can work effectively even in noisy areas. The curvature method can adapt to the different densities and complexities of point cloud data because it relies on local neighborhoods rather than global attributes. Once the local curvature is calculated, the filtering process can be performed quickly, especially when using effective algorithms and data structures. By adjusting the curvature threshold, the strictness of the filtering can be controlled to meet different application requirements. Compared with commonly used statistical methods, which may be greatly affected by outliers, curvature-based methods directly focus on the geometric structure of the point cloud. Compared with commonly used morphological methods, which may not be accurate enough when dealing with complex shapes, curvature-based methods can identify surface features more meticulously. Compared with conditional filtering methods, which may smooth out some important details (such as Gaussian filtering), curvature-based methods can retain these details while removing noise. Compared with machine learning-based methods, which require a large amount of labeled data for training, curvature-based methods do not require a training phase and can be directly applied to new datasets. Compared with graph-based methods, which may require complex graph construction and algorithm implementation, curvature-based methods usually have more intuitive implementation methods. Compared with region growing methods, which may require an iterative process to determine the region boundary, curvature-based methods can determine whether to retain a point in one go. Compared with local feature-based methods, which may require additional feature extraction and matching steps, curvature-based methods extract information directly from point cloud data.

[0112] Processing process:

[0113] For a point (x, y) in the full data set, a surface S is defined in the local neighborhood of the point (x, y). The first-order derivative matrix is ​​the normal vector of the tangent plane of the surface S at the point (x, y).

[0114] ,

[0115] in, is the first-order derivative matrix,

[0116] Based on the first-order derivative matrix, calculate the second-order derivative matrix:

[0117] ,

[0118] Where H is the second-order derivative matrix, , , , is the second-order derivative of the surface S with respect to x or y,

[0119] Eigenvalues ​​of the second-order derivative matrix H and Obtained by the following formula:

[0120] ,

[0121] ,

[0122] get:

[0123] ,

[0124] ,

[0125] = ,

[0126] Where I is the identity matrix.

[0127] and Obtained through the above process.

[0128] Define a curvature threshold , the curvature can be used to filter out the noise points in the comprehensive data set to obtain the preprocessed comprehensive data set:

[0129] ,

[0130] Curvature can reflect the curvature of the point cloud data surface, and noise points often have abnormally high curvature values. The absolute value of the principal curvature is greater than , this point is a noise point.

[0131] Through the above methods, noise points and invalid points in the comprehensive data set are removed, thereby improving the efficiency and accuracy of subsequent processing.

[0132] Furthermore, in one embodiment, to correspond the comprehensive data set to each point (x, y) in the point cloud data, the conversion relationship can be achieved by the following steps:

[0133] First, use 3D laser scanning technology, such as the GoSLAM mobile measurement system, to obtain 3D point cloud data of the coal pile. This technology can collect 3D information of spatial targets in complex field environments in real time, accurately and efficiently. Then, use a large neural network model preset in the cloud, such as Point-Net, to perform semantic segmentation on the 3D point cloud data of the coal pile to distinguish the coal pile point cloud data from other object point cloud data. This step is to obtain point cloud data containing only the coal pile. Then, the segmented coal pile point cloud data is Delaunay triangulated, and the volume data of the coal pile is obtained using the integral calculation method. Finally, the 3D coordinates (X, Y, Z) in the point cloud data are matched with the 3D layout coordinate values ​​in the comprehensive data set to ensure that each point cloud data is associated with the correct 3D space size (characterizing the volume of the coal pile) and the current storage capacity of the coal pile.

[0134] Step 250: Input the comprehensive data set into a time series-based prediction model to obtain the predicted three-dimensional spatial size of the target coal pile, the three-dimensional layout coordinate value of the target coal pile, and the storage capacity of the target coal pile.

[0135] Wherein, the prediction model based on time series includes:

[0136] ,

[0137] in, represents the predicted value corresponding to time t, and is the model correlation parameter, is the variance of the full data set at time t, is the density function of the forecasting model based on time series, is the autoregressive coefficient, which is used to measure and The linear relationship strength between them, j is the lag order of the prediction model based on time series, is the penalty factor for the time series-based forecasting model, represents the comprehensive data set at time t, represents the comprehensive dataset at time tj, represents the change of the comprehensive data set at time t, that is, the first-order difference sequence of the comprehensive data set at time t, Represents the change in the comprehensive data set at time tj.

[0138] This forecasting model combines the advantages of AR model (autoregressive model) and MA model (moving average model), and can capture the autocorrelation and moving average characteristics of time series data at the same time; by introducing the penalty factor , the model can avoid overfitting and improve the robustness of prediction; through the lag order j and the autoregressive coefficient The update can adapt to the changes of time series data; density function The introduction of enables the model to better fit the data distribution and improve the prediction accuracy.

[0139] Specifically, the prediction model based on time series is previously trained based on a large amount of sample data. In one embodiment of the present application, in order to ensure the integrity and accuracy of the data, the data is subjected to a stationarity test to determine whether the time series needs to be differentiated (whether the comprehensive data set needs to be updated). Differentiation helps to remove abnormal trend components in the data.

[0140] In one embodiment, in the time series based prediction model,

[0141] ,

[0142] in, represents the error factor at time t, represents the lag factor of lag order j,

[0143] The autoregressive coefficient By updating the lag order j, we solve

[0144] If the autoregressive coefficient , then the time series is stationary and the comprehensive data set is not updated.

[0145] If the autoregressive coefficient , then the time series is non-stationary, through the autoregressive coefficient Update the comprehensive dataset at time t , get the comprehensive data set at time t The amount of change ,

[0146] Among them, through the following formula, through the autoregressive coefficient Update the comprehensive dataset at time t :

[0147] ,

[0148] .

[0149] Through the implementation of difference and autoregressive coefficient methods, the comprehensive data set The update helps to transform non-stationary time series into stationary series, which is usually a prerequisite for time series forecasting. After the stationarity is improved, the mean and variance of the data set remain relatively unchanged in different time periods, thereby enhancing the reliability of the model. Update comprehensive data set The process can capture the dynamic relationship in the time series. This dynamic modeling allows the model to reflect the changing trend and periodic characteristics of time series data over time. and hysteresis factor The introduction of makes the model more adaptable to different data characteristics. By adjusting these parameters, the model can be effectively configured for different time series data, thereby improving flexibility. , the model can alleviate the impact of outliers on the prediction results to a certain extent, making the comprehensive data set Ability to maintain stability in the face of data noise.

[0150] Furthermore, the model correlation parameter and Determined by the error function, the expression of the error function is,

[0151] ,

[0152] in, represents the error function,

[0153] By the error function Derived to update the model correlation parameters and , the derivation process is as follows:

[0154] ,

[0155] ,

[0156] Among them, the model correlation parameter and The initial value of is 1,

[0157] When the error function When the minimum, the best model correlation parameter is obtained and .

[0158] In this application, the model correlation parameter and The solution process allows the model to be adjusted according to the data characteristics, which improves the flexibility of the model.

[0159] Furthermore, in one embodiment, the predicted volume of the target coal pile represents the predicted volume of the coal pile at a specific time point, and this parameter is one of the main indicators of the model output; the predicted spatial layout position information of the target coal pile can be the predicted coordinates of the surface space points of the coal pile output by the preset time series prediction model, and the coordinate data can be used to analyze the spatial distribution changes of the coal pile and predict possible spontaneous combustion risk areas, so as to facilitate the managers of the material yard to carry out spatial planning of the material yard according to the spatial distribution changes. In addition, in addition to outputting the predicted volume of the target coal pile and the predicted spatial layout position information of the target coal pile, the preset time series prediction model can also include the predicted storage capacity of the target coal pile, which can be in tons; the predicted storage capacity is used to remind the material yard management to adjust the procurement and sales strategies according to the changes in storage capacity to avoid excess and shortage of coal.

[0160] Specifically, in one embodiment of the present application, the asynchronous data processing method for a coal pile provided in the embodiment of the present application is executed on a coal pile data processing system. The relevant description of the coal pile data processing system is as described above.

[0161] Specifically, the computing center processing unit of the coal pile data processing system can be a computer installed in the main control room, and a display screen is connected to the computer to present the real-time map information of the current coal pile including the three-dimensional space size and the three-dimensional layout coordinate value, as well as the layout environment image of the material yard taken by the camera installed in the material yard, and the layout position image of the coal pile and other environmental information. The environmental temperature information and humidity information obtained by multiple temperature sensors and humidity sensors will also be displayed on the large screen synchronously. In addition, when the temperature sensor detects that the temperature value in the current coal pile is higher than the temperature critical value, and the layout environment image of the material yard taken by the camera and the layout position image of the coal pile are analyzed to obtain that the location of the coal pile belongs to the flammable area, the sound and light information with warning effects such as highlighting will be issued on the display screen in time to remind the manager of the material yard that there may be fire hazards.

[0162] Through the implementation of the above control method, the computing center processing unit can obtain and display the map information of the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile in real time, as well as the layout environment image of the material yard and the layout position image of the coal pile and other environmental information. At the same time, by monitoring the data of the temperature sensor and analyzing the images taken by the camera, the fire hazards of the coal pile can be discovered in time, and warning information can be issued through the display screen to remind the managers of the material yard to take corresponding measures. This can help material yard managers to discover and deal with potential fire risks in a timely manner, improve safety and prevent the occurrence of spontaneous combustion fire accidents.

[0163] That is to say, the present application can not only realize the prediction of the predicted volume of the target coal pile, the predicted spatial layout position information of the target coal pile and the predicted storage capacity of the target coal pile, but also realize the prediction of fire hazards of the coal pile and the like.

[0164] The present application also provides an asynchronous data processing device for a coal pile, Figure 3 This is a schematic diagram of the structure of an asynchronous data processing device for a coal pile provided according to an embodiment of the present application, with reference to Figure 3 The asynchronous data processing device for coal pile includes a point cloud data acquisition module 301, a scanning measurement module 302, a weight calculation module 303, a data construction module 304 and a target value output module 305.

[0165] The point cloud data acquisition module 301 is used to acquire the three-dimensional point cloud data of the current coal pile based on the set laser sensor;

[0166] The scanning measurement module 302 is used to input the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, wherein the three-dimensional space size represents the volume of the current coal pile, and the three-dimensional layout coordinate value represents the spatial layout position of the current coal pile;

[0167] A weight calculation module 303 is used to obtain the storage capacity of the current coal pile through a preset first weight model based on the size of the three-dimensional space and the three-dimensional layout coordinate value;

[0168] The data set construction module 304 is used to obtain multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage volumes of multiple current coal piles according to the time series, and to construct a comprehensive data set based on the time series information corresponding to the time series;

[0169] The target value output module 305 is used to input the comprehensive data set into the prediction model based on the time series to obtain the predicted three-dimensional space size of the target coal pile, the three-dimensional layout coordinate value of the target coal pile and the storage capacity of the target coal pile.

[0170] Wherein, the prediction model based on time series includes:

[0171] ,

[0172] in, represents the predicted value corresponding to time t, and is the model correlation parameter, is the variance of the full data set at time t, is the density function of the forecasting model based on time series, is the autoregressive coefficient, which is used to measure and The linear relationship strength between them, j is the lag order of the prediction model based on time series, is the penalty factor for the time series-based forecasting model, represents the comprehensive data set at time t, represents the comprehensive dataset at time tj, represents the change of the comprehensive data set at time t, that is, the first-order difference sequence of the comprehensive data set at time t, Represents the change in the comprehensive data set at time tj.

[0173] Furthermore, in the asynchronous data processing device for the coal pile, the scanning measurement module 302 inputs the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, including:

[0174] Inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system;

[0175] The three-dimensional laser scanning mobile measurement system obtains the map features and / or wall features of the current coal pile based on the three-dimensional point cloud data of the current coal pile.

[0176] Map information of the current coal pile is constructed based on the map features and / or wall features, and the three-dimensional space size and three-dimensional layout coordinate values ​​of the current coal pile are obtained in the map information.

[0177] In one embodiment of the present application, the asynchronous data processing device for a coal pile further includes a preprocessing module, wherein the preprocessing module is used to:

[0178] The comprehensive data set is input into a preset curvature model, and curvature calculation is performed on all data in the comprehensive data set to obtain multiple curvature values ​​of the comprehensive data set data.

[0179] When the curvature value of some data in the comprehensive data set is greater than a preset critical curvature value, the some data are removed from the comprehensive data set:

[0180] Wherein, the preset curvature model is:

[0181] ,

[0182] in, and is the eigenvalue of the second-order derivative matrix, that is, the eigenvalue of the second-order derivative matrix of the surface area formed by the comprehensive data set, is the preset critical curvature value, P1 is the new comprehensive data set obtained by removing some data in the comprehensive data set, P represents the comprehensive data set, and p i represents the i-th subset in the full dataset.

[0183] In one embodiment of the present application, the asynchronous data processing device for a coal pile further includes a temperature sensor and an alarm module, wherein the alarm module is used to:

[0184] Acquire a layout environment image of the material yard and a layout position image of the coal pile taken by at least one camera disposed in the material yard where the coal pile is located;

[0185] Obtain the current temperature value of the coal pile monitored by the temperature sensor;

[0186] According to the layout environment image of the material yard, the layout position image of the coal pile and the current temperature value of the coal pile, the spontaneous combustion fire alarm warning information is output and displayed on the display screen.

[0187] This application builds an accurate time series prediction model through the three-dimensional point cloud data of historical coal piles and prediction algorithms, and obtains accurate target predicted coal pile volume and target predicted spatial position information in real time, helping managers to accurately predict future coal pile conditions, thereby improving the operational efficiency of the material yard and reducing costs. In addition, this application can also predict coal pile fire hazards and other situations.

[0188] The division of each module in the above-mentioned asynchronous data processing device for coal piles is only for illustration. In other embodiments, the asynchronous data processing device for coal piles can be divided into different modules as needed to complete all or part of the functions of the above-mentioned asynchronous data processing device for coal piles. The implementation of each module in the asynchronous data processing device for coal piles provided in the embodiment of the present application can be in the form of a computer program. The computer program can be run on an electronic device such as a terminal or a server. The program modules constituted by the computer program can be stored in the memory of an electronic device such as a terminal or a server. When the computer program is executed by the processor, all or part of the steps of the asynchronous data processing method for coal piles described in the embodiment of the present application are implemented.

[0189] An embodiment of the present application further provides a non-volatile computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the asynchronous data processing method for a coal pile as described in the first aspect is implemented, and has corresponding functions and beneficial effects. It should be understood by those skilled in the art that the embodiments of the present application can provide a method, system or computer program product.

[0190] Therefore, the present application may take the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware. Moreover, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processing module of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processing module of the computer or other programmable data processing device generate instructions for implementing the processes in the process Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0191] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined in this article, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0192] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0193] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. An asynchronous data processing method for a coal pile, characterized in that: The method comprises: Acquire the three-dimensional point cloud data of the current coal pile based on the set laser sensor; Inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, wherein the three-dimensional space size represents the volume of the current coal pile, and the three-dimensional layout coordinate value represents the spatial layout position of the current coal pile; Based on the size of the three-dimensional space and the three-dimensional layout coordinate values, obtaining the storage capacity of the current coal pile through a preset first weight model; According to the time series, multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage volumes of multiple current coal piles are obtained to construct a comprehensive data set based on the time series information corresponding to the time series; The comprehensive data set is input into a prediction model based on a time series to obtain the predicted three-dimensional spatial size of the target coal pile, the three-dimensional layout coordinate value of the target coal pile, and the storage capacity of the target coal pile. Wherein, the prediction model based on time series includes: ， in, represents the predicted value corresponding to time t, and is the model correlation parameter, is the variance of the full data set at time t, is the density function of the forecasting model based on time series, is the autoregression coefficient, which is used to measure and The linear relationship strength between them, j is the lag order of the prediction model based on time series, is the penalty factor for the time series-based forecasting model, represents the comprehensive data set at time t, represents the comprehensive dataset at time tj, represents the change of the comprehensive data set at time t, that is, the first-order difference sequence of the comprehensive data set at time t, Represents the change in the comprehensive data set at time tj.

2. The asynchronous data processing method for coal pile according to claim 1, characterized in that: In the time series based forecasting model, ， in, represents the error factor at time t, represents the lag factor of lag order j, The autoregressive coefficient By updating the lag order j, we solve If the autoregressive coefficient , then the time series is stationary and the comprehensive data set is not updated. If the autoregressive coefficient , then the time series is non-stationary, through the autoregressive coefficient Update the comprehensive dataset at time t , get the comprehensive data set at time t The amount of change , Among them, through the following formula, through the autoregressive coefficient Update the comprehensive dataset at time t : ， 。 3. The asynchronous data processing method for coal pile according to claim 1, characterized in that: The model correlation parameters and Determined by the error function, the expression of the error function is, ， in, represents the error function, By the error function Derived to update the model correlation parameters and , the derivation process is as follows: ， ， Among them, the model correlation parameter and The initial value of is 1, When the error function When the minimum, the best model correlation parameter is obtained and .

4. The asynchronous data processing method for coal pile according to claim 1, characterized in that: Inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, including: Inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system; The three-dimensional laser scanning mobile measurement system obtains the map features and / or wall features of the current coal pile based on the three-dimensional point cloud data of the current coal pile. Map information of the current coal pile is constructed based on the map features and / or wall features, and the three-dimensional space size and three-dimensional layout coordinate values ​​of the current coal pile are obtained in the map information.

5. The asynchronous data processing method for coal pile according to claim 1, characterized in that: After obtaining multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage volumes of multiple current coal piles according to the time series, and constructing a comprehensive data set based on the time series information corresponding to the time series, it also includes: The comprehensive data set is input into a preset curvature model, and curvature calculation is performed on all data in the comprehensive data set to obtain multiple curvature values ​​of the comprehensive data set data. When the curvature value of some data in the comprehensive data set is greater than a preset critical curvature value, the some data are removed from the comprehensive data set: Wherein, the preset curvature model is: ， in, and is the eigenvalue of the second-order derivative matrix, that is, the eigenvalue of the second-order derivative matrix of the surface area formed by the comprehensive data set, is the preset critical curvature value, P1 is the new comprehensive data set obtained by removing some data in the comprehensive data set, P represents the comprehensive data set, and p i represents the i-th subset in the full dataset.

6. The asynchronous data processing method for coal pile according to claim 5, in the preset curvature model, and Obtained through the following process: For a point (x, y) in the full data set, a surface S is defined in the local neighborhood of the point (x, y). The first-order derivative matrix is ​​the normal vector of the tangent plane of the surface S at the point (x, y). ， in, is the first-order derivative matrix, Based on the first-order derivative matrix, calculate the second-order derivative matrix: ， Where H is the second-order derivative matrix, , , , is the second-order derivative of the surface S with respect to x or y, Eigenvalues ​​of the second-order derivative matrix H and Obtained by the following formula: ， ， get: ， ， ， Where I is the identity matrix.

7. The asynchronous data processing method for coal pile according to claim 1, characterized in that: The method further comprises: Obtaining the layout environment image of the material yard where the current coal pile is located, and the layout position image of the current coal pile; Get the current temperature value of the coal pile; According to the layout environment image of the material yard, the layout position image of the coal pile and the current temperature value of the coal pile, spontaneous combustion fire alarm warning information is output and displayed.

8. An asynchronous data processing device for a coal pile, characterized in that: include: A point cloud data acquisition module is used to acquire three-dimensional point cloud data of the current coal pile based on a set laser sensor; A scanning and measuring module, used for inputting the three-dimensional point cloud data of the current coal pile into a preset three-dimensional laser scanning mobile measurement system to obtain the three-dimensional space size and three-dimensional layout coordinate value of the current coal pile, wherein the three-dimensional space size represents the volume of the current coal pile, and the three-dimensional layout coordinate value represents the spatial layout position of the current coal pile; A weight calculation module, used for obtaining the storage capacity of the current coal pile through a preset first weight model based on the size of the three-dimensional space and the three-dimensional layout coordinate value; A data set construction module is used to obtain multiple three-dimensional space sizes, multiple three-dimensional layout coordinate values ​​and corresponding storage volumes of multiple current coal piles according to the time series, and to construct a comprehensive data set based on the time series information corresponding to the time series; A target value output module is used to input the comprehensive data set into a prediction model based on a time series to obtain the predicted three-dimensional space size of the target coal pile, the three-dimensional layout coordinate value of the target coal pile, and the storage capacity of the target coal pile. Wherein, the prediction model based on time series includes: ， in, represents the predicted value corresponding to time t, and is the model correlation parameter, is the variance of the full data set at time t, is the density function of the forecasting model based on time series, is the autoregression coefficient, which is used to measure and The linear relationship strength between them, j is the lag order of the prediction model based on time series, is the penalty factor for the time series-based forecasting model, represents the comprehensive data set at time t, represents the comprehensive dataset at time tj, represents the change of the comprehensive data set at time t, that is, the first-order difference sequence of the comprehensive data set at time t, Represents the change in the comprehensive data set at time tj.

9. An asynchronous data processing system for coal piles, characterized in that: include: one or more processors and memory; The memory is used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the asynchronous data processing method for a coal pile according to any one of claims 1-7.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed, the asynchronous data processing method for a coal pile according to any one of claims 1 to 7 is implemented.

Images