AI material management and control method and system, electronic equipment and storage medium

Abstract

The invention relates to an AI material management and control method and system, electronic equipment and a storage medium, and is applied to the technical field of material management. Factory entering management: determining a vehicle factory entering sequence based on the information of the vehicles to enter the factory; route guidance: planning a driving route of the vehicle entering the factory; unmanned metering: obtaining metering data of the vehicle; factory management: verifying the waybill completion condition of the vehicle based on the measurement data; inventory management: counting inventory information, and making a purchase plan based on the inventory information; coal blending management: determining a coal blending scheme based on an AI calculation model; data management: performing statistical analysis on the data; logistics monitoring: monitoring the position of the vehicle, and estimating the arrival time of the vehicle; light guidance: indicating the advancing route and the coal unloading position through laser light; material sampling: generating a sampling instruction and a packaging instruction; and material testing: acquiring and analyzing sample testing data. The method has the effect of improving the material management efficiency.

Description

Technical Field

[0001] The present application relates to the technical field of material management, and in particular to an AI material management and control method, system, electronic device and storage medium. Background Art

[0002] With the increasing complexity of material transportation and management in various industries, efficient enterprise management is particularly important. It can not only significantly improve the production efficiency and service quality of the enterprise, but also effectively reduce costs, enhance market competitiveness and promote the sustainable development of the industry.

[0003] In current practical operations, many companies still use manual methods to manage materials. For example, manual dispatchers need to manually record the arrival time and material type of each transport vehicle. This way of working is not only cumbersome and error-prone, but also very likely to cause congestion during peak hours, delaying production progress. For another example, for different types of materials, there is a lack of scientific basis for manual allocation of storage areas and unloading locations, and material backlogs or shortages are often caused by wrong decisions. Another problem under the manual management model is the lag in information transmission. Due to the lack of an effective data sharing mechanism, communication and coordination between departments are difficult, which seriously affects the smooth operation of the entire production process. In general, manual management is inefficient and has a high error rate, which is far from meeting the needs of efficient operations of modern enterprises. Summary of the invention

[0004] In order to improve the efficiency of material management, the present application provides an AI material management method, system, electronic device and storage medium.

[0005] In the first aspect, the present application provides an AI material control method, which adopts the following technical solution: An AI material control method, comprising: Driver management: obtaining driver information, identifying the driver identity and driver status based on the driver information, and matching the vehicle identity and vehicle status based on the driver identity and driver status; Factory entry management: determining the order of vehicles entering the factory based on the information of vehicles to be entered into the factory, wherein the information of vehicles to be entered into the factory refers to the information of vehicles whose status is waiting to enter the factory; Route guidance: planning the driving routes for vehicles entering the factory; Unmanned weighing: guide the vehicle to be weighed and obtain the vehicle's measurement data; Factory management: Verify the vehicle's waybill completion status based on measurement data and generate exception information; The method further comprises: Inventory management: collect inventory information and formulate purchasing plans based on the inventory information; Coal blending management: Determine the coal blending plan based on the AI ​​calculation model, which includes the coal pit number and the amount of coal to be blended; Data management: perform statistical analysis on data and generate data reports; Logistics monitoring: monitor the location of vehicles and estimate their arrival time; Light guidance: In response to vehicles entering the coal bunker, laser lights are used to indicate the approach and coal unloading location on the route inside the coal bunker; Material sampling: Generate sampling instructions based on sampling points and sampling times, and generate packaging instructions based on packaging parameters; Material testing: Obtain sample testing data, analyze the sample testing data, and generate abnormal warning information.

[0006] By adopting the above technical solution, the driver information is automatically obtained and identified without manual identity authentication, which improves the efficiency and accuracy of identification; the vehicle status is determined by the driver information, and the vehicle entry order is planned for vehicles whose vehicle status is waiting to enter the factory, so that the vehicles can enter the factory in an optimized order, effectively reducing waiting time and traffic congestion, and improving transportation efficiency; by planning the driving routes of vehicles entering the factory, the vehicles can travel along more suitable routes, reducing the situation where drivers are unfamiliar with the routes and causing detours, and further improving transportation efficiency; by guiding the vehicles to perform automated weighing processes, the waste of human resources is reduced, and the weighing efficiency is improved; by verifying the completion of the waybill, when there is abnormal information in the measurement data, the vehicle can be automatically reminded to wait for the abnormality to be processed, thereby improving the accuracy and completeness of material transportation; at the same time, inventory statistics can be performed regularly and irregularly according to demand, which greatly improves the efficiency and accuracy of inventory statistics and procurement plan formulation; The AI ​​calculation model determines the coal blending plan, which improves the convenience of the coal blending process; by statistically analyzing various data and generating intuitive data reports, it can provide decision-making support for managers; by real-time monitoring of vehicle positions and estimating arrival times, it provides strong guarantees for production scheduling and material distribution; by using laser lights to mark the approach and indicate the coal unloading location, the driver can reach the coal unloading location more quickly and conveniently, improving the accuracy and efficiency of coal unloading; the material sampling module automates the material sampling and packaging process, reducing the possibility of manual contact with samples, thereby improving the reliability of samples; the material testing module automatically analyzes sample testing data and generates abnormal warning information, improving the efficiency of testing and the reliability of test results. In summary, by improving the efficiency of driver management, factory entry management, route guidance, unmanned metering, factory exit management, inventory management, coal blending management, data management, logistics monitoring, lighting guidance, material sampling and material testing, the efficiency of material management is improved.

[0007] Optionally, determining the order of vehicles entering the factory based on the information of vehicles to be entered into the factory includes: Determine the material type and waiting time of each vehicle based on the information of the vehicles to be transported into the factory; Obtain inventory information and material consumption information; Sorting the material types based on the inventory information and the material consumption information to obtain a first sorting result; Determine the order of vehicles entering the factory based on the first sorting result and the waiting time; The sorting of the material types based on the inventory information and the material consumption information to obtain a first sorting result includes: Determine a current material consumption rate based on the material consumption information; Determine the inventory quantity of each material based on the inventory information; Determine the consumable time based on the inventory quantity and the current material consumption speed; The material types are sorted based on the consumable time to obtain a first sorting result.

[0008] By adopting the above technical solution, the consumable time of each material is calculated according to the current material consumption speed and inventory quantity, and the material types are sorted according to the consumable time to obtain a first sorting result. Therefore, the shortage degree of each material can be determined according to the first sorting result, and the order of vehicles entering the factory is determined according to the first sorting result and the waiting time of the vehicles, so that vehicles with higher shortages and longer waiting times can enter the factory first, which improves the rationality of the order of vehicles entering the factory.

[0009] Optionally, the planning of the driving route of the vehicle entering the factory includes: Obtain the material type corresponding to the vehicle entering the factory; Determine candidate locations for each step based on the material type and the preset process; Obtaining current queue information of each candidate position; Calculate the distance between each two candidate positions in two adjacent steps; Determine the target position of each step based on the current queue information and the distance; Sorting the target locations according to the preset process to obtain a driving route; The guiding vehicle to be weighed and obtaining the vehicle's measurement data includes: Obtaining the vehicle identity based on a measurement method, wherein the measurement method includes on-vehicle measurement and off-vehicle measurement; Acquire vehicle images; Identify the vehicle image based on a preset AI image recognition model to obtain driver information and vehicle information; If the driver's seat information and the vehicle information both meet the preset measurement rules, then obtain the vehicle's measurement data; The verification of the vehicle's waybill completion status based on the metering data and the generation of abnormal information include: verifying the identity of the vehicle; Calculating the difference of the measurement data; comparing the difference with the net weight of the material; If the difference is the same as the net weight, then a confirmation factory information is generated; If the difference is not the same as the net weight, an exception message is generated.

[0010] By adopting the above technical solution, when determining the vehicle's driving route, the current queue information of each candidate position and the distance between the candidate positions are taken into consideration, thereby improving the rationality of the driving route and improving the transportation efficiency according to the driving route; the vehicle image is recognized by a preset AI image recognition model, and if the driver's seat information and vehicle information both meet the preset measurement rules, the vehicle's measurement data is obtained, otherwise the vehicle is guided by voice to adjust according to the preset measurement rules for weighing. No staff is required to be present during the whole process, thereby reducing personnel consumption; by comparing the difference in measurement data with the net weight of the material, the factory information or abnormal information is generated and determined, and no human participation is required during the whole process, thereby improving the efficiency of factory management.

[0011] Optionally, collecting inventory information and formulating a purchase plan based on the inventory information includes: Inventory measurement: count inventory information and record inventory in and out information; Procurement: Determine the procurement plan based on the inventory information and material consumption information; The determining of the procurement plan based on the inventory information and the material consumption information includes: Determine the inventory quantity of each material based on the inventory information; Determine a current material consumption rate based on the material consumption information; Get the historical material consumption rate of the current cycle; Calculate an estimated material consumption rate based on the historical material consumption rate, the current material consumption rate and a preset weight; Determine the estimated inventory quantity of various materials at various times based on the inventory quantity and the estimated material consumption rate; Determine the time when the estimated inventory quantity is less than a preset inventory threshold as the latest purchase time; A procurement plan is determined based on the latest procurement time and the corresponding material type.

[0012] By adopting the above technical solution, by regularly or irregularly counting inventory information and recording inventory in and out information, inventory management can be more refined; when calculating the latest purchase time, not only the current material consumption rate is taken into account, but also the historical material consumption rate of the current cycle is taken into account, so as to determine the estimated material consumption rate and improve the reliability of the latest purchase time.

[0013] Optionally, determining the coal blending scheme based on the AI ​​computing model includes: Obtain the coal blending quantity, coal blending requirement parameters and material parameters of various material types; Determine the coal blending ratio of various material types based on the AI ​​calculation model; Determine the required quantity of various types of materials based on the coal blending ratio and the coal blending quantity; Get the coal pit number corresponding to each material type and the quantity of materials in each coal pit; Determine the coal quantity for each coal pit number based on the material quantity and the required quantity; Determine a coal blending scheme based on the coal blending ratio and the coal blending quantity; Before determining the coal blending scheme based on the AI ​​computing model, the method further includes: Acquire historical coal blending information, wherein the historical coal blending information includes historical demand parameters, historical material parameters, and historical coal blending ratios; Dividing the historical coal blending information according to the types of the historical demand parameters to obtain a plurality of historical coal blending information combinations; The coal blending algorithm is trained based on multiple combinations of the historical coal blending information to obtain the AI ​​calculation model.

[0014] By adopting the above technical scheme, the coal blending algorithm is trained respectively through the divided multiple historical coal blending information combinations to obtain an AI calculation model, thereby improving the reliability of the AI ​​calculation model. The coal blending ratio of various material types is determined through the AI ​​calculation model, and the coal blending quantity of each coal pit number is determined by analyzing the quantity of materials in each coal pit. The coal blending process does not require human participation, thereby improving the coal blending efficiency.

[0015] Optionally, the statistical analysis of the data and generation of data reports may include: Obtaining user needs, wherein the user needs include irregular needs and regular needs; Analyze the data based on the user needs and preset tools, and visualize the analysis results, which include the data report; If there is an abnormality in the analysis result, a warning message is generated; The monitoring of the vehicle's location and estimating the vehicle's arrival time include: Obtain historical transportation data of vehicles; The AI ​​calculation algorithm is trained based on the historical transportation data to obtain an arrival time prediction model; Acquiring vehicle transportation data, weather data, and traffic data, wherein the vehicle transportation data includes a vehicle location and a destination; Inputting the vehicle location, the weather data, the traffic data, and the destination into the arrival time prediction model to estimate the arrival time of the vehicle; In response to a vehicle entering a coal bunker, a guide path is drawn on the route in the coal bunker by means of laser lights and a coal unloading position is indicated, including: In response to the vehicle entering the coal bunker, determining a travel route within the coal bunker based on the travel route; An approach road is marked on the travel route by means of laser lights, and the coal unloading position is indicated by means of laser lights.

[0016] By adopting the above technical solution, data can be analyzed regularly or irregularly according to user needs, early warning information can be generated, and the analysis results can be visualized, which improves the reliability of data management and makes the data management process clearer and more intuitive. The arrival time of the vehicle can be estimated through the arrival time prediction model, so that the factory can prepare for receiving the goods in advance, and the approach road and coal unloading location can be indicated by laser lights, which improves the accuracy and efficiency of coal unloading.

[0017] Optionally, generating a sampling instruction based on a sampling point and a sampling number, and generating a packing instruction based on a packing parameter, includes: Obtain vehicle monitoring data, material monitoring data and vehicle quantity, where the vehicle quantity is the number of transport vehicles corresponding to one waybill information; determining a sampling number for each vehicle based on the number of vehicles; Performing image recognition on the vehicle monitoring data to determine a sampling range, where the sampling range is the range where the material is located; Performing image recognition on the material monitoring data, and determining a sampling point based on the recognition result, the sampling times and the sampling range; Generate a sampling instruction based on the sampling point and the sampling number; Get packaging parameters; Generate a packing instruction based on the packing parameters; The obtaining of sample test data, analyzing the sample test data, and generating abnormal warning information includes: Acquire and store sample analysis data; Determine whether the sample test data is abnormal; If there is an abnormality in the sample test data, abnormality warning information is generated.

[0018] By adopting the above technical solution, the possibility of manual contact with samples is reduced through the automation of material sampling and packaging processes, thereby improving the reliability of samples. In addition, the sampling and packaging processes do not require manual participation, reducing the waste of human resources. By automatically analyzing sample test data and generating abnormal warning information, the efficiency and reliability of the test are improved.

[0019] In the second aspect, the present application provides an AI material management and control system, which adopts the following technical solutions: An AI material control system, comprising: A camera, used to collect image data or video data at each location; ID card recognition equipment, used to recognize ID card information; Voice guidance equipment, used to play preset audio when the vehicle reaches a specified location or completes a specified step; Automatic weighing equipment, used to weigh vehicles or materials and obtain measurement data; Sampling equipment for sampling materials on the vehicle; Packaging equipment, used to pack samples; Testing equipment, used to test materials according to preset requirements and obtain sample testing data; Coal pan meter, used to collect images of materials in the coal bunker, Lighting equipment, used to emit laser lights to guide the vehicle's route and indicate the coal unloading location; Vehicle monitoring equipment, used to monitor the status of vehicles and materials in real time and obtain monitoring data, the vehicle monitoring equipment includes GPS equipment and vehicle-mounted sensors; Handheld terminal devices are used by drivers or staff to view and operate information; The electronic device is used to obtain the image data, the video data, the measurement data, the sample test data, the material image, the monitoring data and the operation information, and is also used to control the voice guidance device to play audio, control the sampling device to sample, control the packaging device to package, and control the lighting device to emit laser light.

[0020] By adopting the above technical solution, by installing cameras, ID card recognition equipment, voice guidance equipment, sampling equipment, packaging equipment, testing equipment, automatic weighing equipment, coal pan and lighting equipment in the factory, it is possible to more conveniently obtain various data and guide the transport vehicles to complete various processes. The transport vehicles are equipped with GPS equipment and on-board sensors, which can monitor the status of the vehicles and materials in real time. The drivers or staff can view and operate information more conveniently through handheld terminal devices. Electronic equipment can combine various equipment to complete material management by obtaining image data, video data, measurement data, sample testing data, material images, monitoring data, operation information, and controlling the voice guidance equipment to play audio, the sampling equipment to take samples, the packaging equipment to pack, and the lighting equipment to emit laser lights.

[0021] In a third aspect, the present application provides an electronic device, which adopts the following technical solution: An electronic device comprises a processor, wherein the processor is coupled to a memory; The memory stores a computer program that can be loaded by the processor and execute the AI ​​material management and control method described in any one of the first aspects.

[0022] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium storing a computer program that can be loaded by a processor and execute the AI ​​material control method described in any one of the first aspects.

[0023] In summary, the present application includes at least one of the following beneficial technical effects: 1. Automatically obtain and identify driver information without manual identity authentication, which improves the efficiency and accuracy of identification; determine the vehicle status through driver information, and plan the vehicle entry order for vehicles waiting to enter the factory, so that vehicles can enter the factory in an optimized order, effectively reducing waiting time and traffic congestion, and improving transportation efficiency; by planning the driving routes of vehicles entering the factory, vehicles can travel along more suitable routes, reducing the situation where drivers are unfamiliar with the routes and causing detours, further improving transportation efficiency; by guiding vehicles to perform automated weighing processes, the waste of human resources is reduced, and at the same time, the weighing efficiency is improved; by verifying the completion of the waybill, when there is abnormal information in the measurement data, the vehicle can be automatically reminded to wait for the abnormality to be processed, thereby improving the accuracy and completeness of material transportation; at the same time, inventory statistics can be performed regularly and irregularly according to demand, which greatly improves the efficiency and accuracy of inventory statistics and procurement plan formulation; through AI calculation model The coal blending scheme is determined based on the model, which improves the convenience of the coal blending process; by statistically analyzing various data and generating intuitive data reports, it can provide decision-making support for managers; by real-time monitoring of vehicle positions and estimating arrival times, it provides strong guarantees for production scheduling and material distribution; by using laser lights to mark the approach and indicate the coal unloading location, the driver can reach the coal unloading location more quickly and conveniently, which improves the accuracy and efficiency of coal unloading; the material sampling module automates the material sampling and packaging process, reduces the possibility of manual contact with samples, and thus improves the reliability of samples; the material testing module automatically analyzes sample test data and generates abnormal warning information, which improves the efficiency of testing and the reliability of test results. In summary, by improving the efficiency of driver management, factory entry management, route guidance, unmanned metering, factory exit management, inventory management, coal blending management, data management, logistics monitoring, light guidance, material sampling and material testing, the efficiency of material management is improved; 2. The consumable time of each material is calculated based on the current material consumption speed and inventory quantity, and the material types are sorted according to the consumable time to obtain a first sorting result, so that the shortage of each material can be determined according to the first sorting result. The order of vehicles entering the factory is determined according to the first sorting result and the waiting time of the vehicles, so that vehicles with higher shortage and longer waiting time can enter the factory first, that is, the rationality of the order of vehicles entering the factory is improved; 3. When determining the vehicle's driving route, the current queue information of each candidate position and the distance between the candidate positions are taken into account, which improves the rationality of the driving route, thereby improving the transportation efficiency according to the driving route; the vehicle image is recognized by the preset AI image recognition model. If the driver's seat information and vehicle information meet the preset measurement rules, the vehicle's measurement data is obtained. Otherwise, the voice guides the vehicle to adjust according to the preset measurement rules, so as to perform weighing. No staff is required to be present during the whole process, reducing personnel consumption; by comparing the difference in measurement data with the net weight of the material, the factory information or abnormal information is generated and determined. No human participation is required during the whole process, which improves the efficiency of factory management; 4. The possibility of manual contact with samples is reduced through the automation of material sampling and packaging, thereby improving the reliability of samples. In addition, the sampling and packaging process does not require manual participation, reducing the waste of manual resources. By automatically analyzing sample test data and generating abnormal warning information, the efficiency and reliability of the test are improved; 5. By installing cameras, ID card recognition equipment, voice guidance equipment, sampling equipment, packaging equipment, testing equipment, automatic weighing equipment, coal pans and lighting equipment in the factory, it is easier to obtain various data and guide transport vehicles to complete various processes. Transport vehicles are equipped with GPS equipment and on-board sensors to monitor the status of vehicles and materials in real time. Drivers or staff can view and operate information more conveniently through handheld terminal devices. Electronic equipment can combine various equipment to complete material management by obtaining image data, video data, measurement data, sample testing data, material images, monitoring data, operation information, and controlling voice guidance equipment to play audio, sampling equipment to take samples, packaging equipment to pack, and lighting equipment to emit laser lights. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is a structural block diagram of an AI material management and control system provided in an embodiment of the present application; Figure 2 It is a flow chart of an AI material control method provided in an embodiment of the present application; Figure 3 A schematic diagram of a process for determining the order in which vehicles enter a factory provided in an embodiment of the present application; Figure 4 A schematic diagram of a process for determining a first sorting result of a material type provided in an embodiment of the present application; Figure 5 A schematic diagram of a process for determining a vehicle driving route provided in an embodiment of the present application; Figure 6 A schematic diagram of a process for determining a procurement plan provided in an embodiment of the present application; Figure 7 A schematic diagram of a process for determining a coal blending scheme provided in an embodiment of the present application; Figure 8 A schematic diagram of a process for generating an AI computing model provided in an embodiment of the present application; Fig. 9 A schematic diagram of the sampling and packaging process provided in the embodiment of the present application; Fig.10 It is a structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are 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.

[0026] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0027] The present application embodiment provides an AI material management and control system, such as Figure 1 As shown, the AI ​​material control system includes central electronic equipment and edge devices scattered in various locations in the factory. The edge devices include but are not limited to cameras, ID card recognition equipment, voice guidance equipment, sampling equipment, packaging equipment, testing equipment, automatic weighing equipment, coal panning equipment, lighting equipment, GPS equipment installed on transport vehicles, vehicle-mounted sensors, and handheld terminal devices of drivers or staff. The specific installation location of the edge devices is not specifically limited here. The electronic equipment is communicatively connected to each edge device, and the GPS equipment and vehicle-mounted sensors are communicatively connected to the driver's handheld terminal device.

[0028] Among them, the camera is used to collect image data or video data at various locations. For example, cameras for real-time collection or collection as needed can be installed at the check-in point, factory entry point, sampling point, skinning point, unloading point, hair passing point, and factory gate.

[0029] ID card recognition equipment can be installed at locations where identity verification is required, such as check-in locations, weighing locations, factory exits, etc., to identify ID card information, so that drivers can quickly identify their identities by swiping their ID cards.

[0030] In locations where the driver has to perform a lot of operations, has to wait for a long time, or encounters many problems, a voice guidance device can be installed to guide the vehicle to move or complete various steps. The voice guidance device can be controlled by the staff to play audio, or it can have an internal preset program to play the preset audio when the vehicle reaches the designated location or completes the designated steps.

[0031] Automatic weighing equipment can weigh vehicles or materials. For example, it can be a floor scale. When the vehicle arrives at the weighing location and the vehicle identity is verified, the electronic equipment will obtain the measurement data from the automatic weighing equipment.

[0032] The sampling device is used to sample materials on the vehicle. The sampling device may have a built-in camera. For example, when the vehicle arrives at the sampling device, the vehicle image can be captured by the built-in camera, and the sampling position can be determined by the built-in program, thereby generating a sampling instruction based on the sampling position. That is, the sampling process does not require passing through the electronic device, but in order to improve the traceability of the data, the vehicle image can be transmitted to the electronic device for recording.

[0033] The packaging equipment for packaging samples is the same as the sampling equipment mentioned above, and may have a built-in camera, which will not be described in detail here.

[0034] The testing equipment is used to test materials according to preset requirements and obtain sample testing data.

[0035] The coal pan meter is located above the coal bunker and is used to collect images of materials in the coal bunker to assist in inventory statistics.

[0036] The lighting equipment is installed in the coal bunker to emit laser lights to guide the vehicle's route and indicate the coal unloading location. It is worth noting that the lighting equipment can also be installed in other locations as needed. This is only an example and is not specifically limited.

[0037] Through handheld terminal devices, drivers or staff can easily view and operate information.

[0038] Vehicle monitoring equipment is installed on the transport vehicle, including GPS equipment and on-board sensors. The on-board sensors include but are not limited to temperature sensors, humidity sensors, smoke concentration sensors, pressure sensors, vibration sensors, etc. The GPS equipment and on-board sensors can monitor the conditions of the vehicle and materials in real time to obtain monitoring data. The GPS equipment and on-board sensors are both communicatively connected to the driver's handheld terminal device, so that the vehicle's location data and sensor data can be directly transmitted to the driver's handheld terminal.

[0039] The electronic equipment is used to obtain image data, video data, measurement data, sample testing data, material images, monitoring data and operation information collected by each edge device. It is also used to control the voice guidance device to play audio, control the sampling device to sample, control the packaging equipment to package, and control the lighting equipment to emit laser light.

[0040] It is worth noting that each edge device may also be provided with a data processing device for performing data processing. The data processing device may be a small computer or a small server, which is not specifically limited here.

[0041] The above-mentioned AI material control system can be applied to various industries such as coking industry, coal industry, steel industry, automobile manufacturing industry, electronics manufacturing industry, food processing industry, pharmaceutical manufacturing industry, machinery manufacturing industry, chemical and energy industry, logistics and warehousing industry, etc. The materials include but are not limited to coal, steel, auto parts, electronic parts, food and medicine, etc., and no specific limitation is made here.

[0042] Figure 2 A flowchart of an AI material management method provided in an embodiment of the present application.

[0043] like Figure 2 As shown, an AI material control method includes (steps S101-S112, it is worth noting that the steps included in different application scenarios and the order of use of each step may be different, and the order of step numbers does not represent the process order of the steps): Step S101, driver management: obtaining driver information, identifying the driver identity and driver status based on the driver information, and matching the vehicle identity and vehicle status based on the driver identity and driver status.

[0044] In this embodiment, after the transport vehicle is loaded with materials, the driver will scan the QR code through the mobile phone and enter the driver's identity, waybill information, material information and vehicle information, so as to bind the driver's identity and the vehicle driven by the driver to the waybill information. It is worth noting that one waybill information can be bound to multiple vehicles, that is, multiple vehicles jointly complete the transportation work of one waybill. At this time, the position of the transport vehicle begins to be monitored in real time. When monitoring the position of the transport vehicle, the vehicle's position data and / or sensor data can be transmitted to the electronic device in real time through the vehicle monitoring equipment installed on the vehicle, or the vehicle's position data and / or sensor data can be transmitted to the electronic device in real time through the handheld terminal device (for example, mobile phone) carried by the driver, so as to realize real-time monitoring. Among them, the waybill information includes but is not limited to material batch, material type, shipping address and receiving address, etc., the material information includes material net weight, material type, material batch, etc., and the vehicle information includes vehicle type, license plate number, vehicle size (for example, vehicle height, vehicle length), vehicle service life, etc.

[0045] After the driver has loaded the materials, he will go through multiple states (i.e., driver status), such as driving, signing in, waiting to enter the factory, sampling, skinning, unloading, roughing, and leaving the factory. It is worth noting that the order of driver status changes may vary in different application scenarios, which is not specifically limited here. Each state corresponds to a position, and the driver information needs to be obtained at each position. The method of obtaining the driver information includes collecting the driver's facial image, obtaining the driver's identity through AI image recognition of the facial image, and updating the driver status according to the location where the facial image is collected (the vehicle status is the same as the driver status). The method of obtaining the driver information also includes the driver swiping his ID card for identification, and updating the driver status according to the location where the ID card is swiped.

[0046] To facilitate subsequent description, this application takes an application scenario of a coking plant as an example. When the driver arrives at the sign-in place of the coking plant, the camera installed at the entrance of the sign-in place will capture the vehicle image and use the preset AI image recognition model to determine whether the vehicle type and license plate number correspond to the vehicle. If not, an abnormal mark is added to the license plate number. The license plate number with the abnormal mark cannot be signed in subsequently. If the vehicle type corresponds to the license plate number, there is no need to add an abnormal mark. Vehicles corresponding to the license plate number that enter the sign-in place and do not have an abnormal mark added can sign in. After signing in, you can wait for the call to enter the coking plant. The signed-in vehicle is determined as a vehicle to be entered into the plant. The sign-in process is that the driver signs in through a handheld terminal device. After entering the sign-in place, the driver can decide the sign-in time as needed. There is no need to sign in immediately after entering the sign-in place (you can rest for a while first). Among them, the preset AI image recognition model includes but is not limited to a convolutional neural network model (CNN), a YOLO target detection model, etc., and the model type for a specific application can be selected according to needs.

[0047] By identifying whether the vehicle type and license plate number correspond, it is possible to reduce the occurrence of people entering the check-in area with license plates or vehicles of other types entering the check-in area with license plates corresponding to waybill information, thereby making the vehicle entry process more rigorous and reliable.

[0048] Step S102, factory entry management: determining the order in which vehicles enter the factory based on the information of vehicles to be entered into the factory.

[0049] Among them, the information of vehicles waiting to enter the factory is the information of vehicles whose vehicle status is waiting to enter the factory.

[0050] When a vehicle arrives at the sign-in area of ​​the coking plant and signs in, it needs to wait for its number to be called before it can enter the coking plant. The calling is divided into automatic calling and manual calling. Automatic calling is based on the order in which the vehicles enter the plant. Manual calling is based on the order determined by the coal management staff under preset circumstances (for example, when the inventory information of the plant is incomplete or when there is a temporary demand in the plant). It does not call in the order in which the vehicles arrive. This takes into account the needs of the coking plant, thereby improving the production efficiency of the coking plant.

[0051] like Figure 3 As shown, as an optional implementation of this embodiment, the order of vehicles entering the factory is determined based on the information of vehicles to be entering the factory, including (steps Sa to Sd): Step Sa, determining the material type and waiting time of each vehicle based on the information of the vehicles to be transported into the factory; Step Sb, obtaining inventory information and material consumption information; Step Sc, sorting the material types based on the inventory information and the material consumption information to obtain a first sorting result; Step Sd: determining the order in which vehicles enter the factory based on the first sorting result and the waiting time.

[0052] In this embodiment, the information of vehicles to be admitted to the factory includes the waiting time of the driver at the entrance of the coking plant and the waybill information. The material type and waiting time of each vehicle are searched from the information of vehicles to be admitted to the factory, and the inventory information and material consumption information are obtained from the database. The inventory information and material consumption information of various material types are analyzed to sort the material types to obtain a first sorting result. The first sorting result is the sorting result of various material types. The vehicles to be admitted to the factory are sorted according to the corresponding material types according to the first sorting result. If the material types corresponding to multiple vehicles to be admitted to the factory are the same, they are sorted in order from long to short according to the waiting time to obtain the order of vehicles entering the factory, that is, vehicles with the corresponding material types that are at the front of the first sorting result and with longer waiting times are given priority to enter the factory. It is worth noting that vehicles corresponding to the same waybill are in adjacent positions in the order of vehicles entering the factory, which is convenient for subsequent statistical analysis.

[0053] like Figure 4 As shown, as an optional implementation of this embodiment, the material types are sorted based on the inventory information and the material consumption information to obtain a first sorting result, including (steps Sc1 to Sc4): Step Sc1, determining the current material consumption rate based on the material consumption information; Step Sc2: determining the inventory quantity of each material based on the inventory information; Step Sc3, determining the available consumption time based on the inventory quantity and the current material consumption speed; Step Sc4: sort the material types based on the consumable time to obtain a first sorting result.

[0054] In this embodiment, the current material consumption speed is found from the material consumption information (the material consumption speed is the average consumption speed within a preset period, the preset period can be one month or two months, which is not specifically limited here), and the inventory quantity of each material is found from the inventory information. The consumable time = inventory quantity / current material consumption speed. The material types are sorted from short to long according to the consumable time to obtain the first sorting result.

[0055] As another optional implementation of this embodiment, the material types are sorted based on the inventory information and the material consumption information to obtain a first sorting result, including: obtaining the historical consumption information and the current inventory quantity of each material from the database, the historical consumption information including the historical production order information, the production consumption time period corresponding to each historical production order information (the division of the consumption time period is pre-set according to the needs of the factory area, for example, it can be one month, one week, or one day) and the consumption speed; dividing the historical consumption information into a training set, a verification set, and a test set according to a preset division rule, training the AI ​​learning model through the training set, verifying the AI ​​learning model through the verification set, and verifying the AI ​​learning model through the test set. The test set is used to test the AI ​​learning model to obtain an AI estimated consumption model. The AI ​​estimated consumption model can analyze the inventory consumption rate in different time periods, so as to obtain the consumable time according to the inventory quantity and the inventory consumption rate. The current inventory quantity is input into the AI ​​estimated consumption model to obtain the consumable time of each material, and the material types are sorted from short to long according to the consumable time to obtain a first sorting result, wherein the preset division rule can be to select 75% of the historical consumption information from each production consumption time period as a training set, 15% of the historical consumption information as a verification set, and 15% of the historical consumption information as a test set. The AI ​​learning model includes but is not limited to support vector machine models (Support Vector Machines, SVM), decision tree models, K-means clustering models (K-Means Clustering), feedforward neural network models (Feedforward Neural Networks, FNN), deep belief network models (Deep Belief Networks, DBN), etc.

[0056] When the vehicle is called, the driver drives the vehicle to the card-making office and provides the card-making information to the card-making staff. The card-making information includes but is not limited to the driver's ID card, the original receipt of the waybill, the vehicle's driving license, etc. After the staff verifies that the information is correct, they generate electronic card-making information through a handheld terminal device. The electronic card-making information includes but is not limited to waybill information, vehicle information, material information, etc. Vehicles with electronic card-making information will be identified as vehicles entering the factory.

[0057] Step S103, route guidance: planning the driving route of vehicles entering the factory.

[0058] When a vehicle enters the coking plant, a driving route is planned in advance for the vehicle. The vehicle drives according to the route, which can improve the vehicle's driving efficiency and reduce detours caused by unfamiliar routes.

[0059] like Figure 5As shown, as an optional implementation of this embodiment, planning the driving route of the vehicle entering the factory includes (step S11 to step S16): Step S11, obtaining the material type corresponding to the vehicle entering the factory; Step S12: determining candidate locations for each step based on material type and preset process; Step S13, obtaining the current queue information of each candidate position; Step S14, calculating the distance between every two candidate positions in two adjacent steps; Step S15, determining the target position of each step based on the current queue information and distance; Step S16: sort the target locations according to a preset process to obtain a driving route.

[0060] In this embodiment, the material type corresponding to the vehicle entering the factory is searched from the vehicle information. Different material types correspond to different operating positions. The preset process can be: coking plant entrance (swiping ID card to enter the coking plant) → sampling point (sampling from the material for testing) → roughing point (weighing the vehicle loaded with materials) → unloading point → skinning point (weighing the vehicle after unloading again) → coking plant exit. It is worth noting that the preset processes corresponding to different application scenarios may be different. This is only an example and is not specifically limited. According to the material type, the candidate positions of each step in the preset process are searched from the database or the staff. Monitoring equipment is installed at each candidate position to monitor the queuing information of the position. For example, the monitoring equipment can be a camera, which uses AI image recognition technology (convolutional neural network, YOLO target detection and other technologies can be used, which are not specifically limited here) to identify the queue. The number of vehicles is determined, and the number of identified vehicles is stored in a database. The current queue information of each candidate position, that is, the number of vehicles currently in the queue, is obtained from the database. The distance between each two candidate positions in two adjacent steps is calculated. A candidate position is selected from each step of the preset process in turn to obtain multiple candidate position combinations. Each candidate position combination corresponds to a candidate driving route. The total distance of the candidate driving route is calculated according to the distance between two adjacent candidate positions. The estimated driving time of the candidate driving route = the total distance of the candidate driving route / the preset driving speed + the waiting time of each candidate position. The waiting time of each candidate position = the number of vehicles currently queuing at the candidate position × the unit waiting time (the unit waiting time of each candidate position is obtained from the database). The candidate driving route with the shortest estimated driving time is determined as the driving route of the vehicle, and the candidate position in the driving route is determined as the target position.

[0061] As an optional implementation of this embodiment, when planning the driving route of vehicles entering the factory, physical structure information within the factory area (including but not limited to the building locations within the factory area, all actual passable paths between every two buildings, and the distances of the actual passable paths, etc.) can also be obtained from the staff, and an AI route planning model of the factory area can be constructed according to the physical structure information within the factory area. The AI ​​route planning model can be implemented through dynamic programming algorithms, particle swarm optimization algorithms, etc., and can also be combined with 3D modeling or digital twin technologies to construct a 3D AI route planning model. The AI ​​route planning model can obtain the vehicle's driving route based on the queuing situation at each location and the preset flow speed at each location. Path planning is performed through the AI ​​route planning model, that is, each candidate location, the number of vehicles currently queuing at each candidate location, and the preset flow speed of each candidate location (which can be obtained from the staff) are input into the AI ​​route planning model to obtain the vehicle's driving route.

[0062] Step S104, unmanned weighing: guiding the vehicle to be weighed and obtaining the vehicle's measurement data.

[0063] Unmanned metering can be applied to the hair-passing and skin-passing areas. Both the hair-passing and skin-passing areas are equipped with automatic weighing equipment (for example, floor scales), voice guidance equipment and cameras. The metering methods for weighing vehicles include on-vehicle metering and off-vehicle metering (the two metering methods are applicable to different application scenarios, for example, on-vehicle metering is applicable to the weighing of vehicles corresponding to the waybill information, and off-vehicle metering is applicable to the precise weighing of materials transferred by coking plant staff). On-vehicle metering means that after the driver drives the vehicle to the automatic weighing equipment, he can swipe his ID card through the window to identify the vehicle identity without getting off the vehicle. At the same time, the camera at the metering area (hair-passing and skin-passing) captures the vehicle image, and the preset AI image recognition model identifies the driver's seat information of the vehicle, that is, whether there are people in the main driver's seat and the co-driver's seat and records the number of people in the vehicle, as well as identifies the vehicle information, that is, whether the vehicle type corresponds to the license plate number. The preset metering rules of the on-vehicle metering method include the correspondence between the vehicle type and the license plate number. If the vehicle type and the license plate number in the vehicle information correspond, the automatic weighing The measurement data is obtained from the weighing equipment; the under-vehicle measurement means that after the driver drives the vehicle to the automatic weighing equipment, he needs to get off the vehicle and swipe his ID card to identify the vehicle identity. At the same time, the camera at the measuring place (hair passing place and skin passing place) takes the vehicle image, and the preset AI image recognition model identifies the driving personnel information of the vehicle, that is, whether there are people in the main driving seat and the co-pilot seat, and identifies the vehicle information, that is, whether the vehicle type corresponds to the license plate number. The preset measurement rules of the under-vehicle measurement method include that there is no one in the main driving seat and the co-pilot seat of the vehicle, and the vehicle type corresponds to the license plate number. If there is someone in the main driving seat or the co-pilot seat of the vehicle, the voice guidance device will remind the people in the vehicle to get off. If the vehicle type corresponds to the license plate number and there is no one in the vehicle, the measurement data is obtained from the automatic weighing equipment. In the above process, the voice guidance device can remind the driver to weigh according to the steps, and no other staff needs to operate on-site throughout the process (when the data is abnormal, the staff can remotely direct the driver to handle it), which improves the efficiency of the weighing process and reduces personnel waste.

[0064] Step S105, factory management: verify the vehicle's waybill completion status based on the measurement data and generate exception information.

[0065] When the vehicle arrives at the exit of the coking plant, it is also necessary to verify the identity of the vehicle. The verification method is the same as above (for example: swiping the ID card for verification or through the preset AI image recognition model for verification). It will not be repeated here. After the vehicle verifies its identity, the difference in the measurement data between the hair and the skin is calculated, and the difference is compared with the net weight in the material information when registering at the factory. When the difference is the same as the net weight, the waybill completion status of the vehicle is completed, there is no abnormal information, and the confirmation of leaving the factory information is generated, and the vehicle can leave the factory normally; when the difference is different from the net weight, the waybill completion status of the vehicle is incomplete, there is abnormal information, the vehicle cannot leave the factory at this time, and needs to wait for further processing by the staff, which effectively reduces the situation where the goods are not unloaded cleanly. At the same time, the verification calculation does not require human participation, which not only improves the calculation efficiency, thereby improving the driver's factory efficiency, but also reduces the consumption of human resources in the factory.

[0066] It is worth noting that, when transporting materials, the order of the above steps S101-S105 is fixed, but the order of steps S106-S112 is not fixed, and any one or several of the steps can be completed in any order as needed.

[0067] Step S106, inventory management: collect inventory information and formulate a purchase plan based on the inventory information.

[0068] In this embodiment, inventory information of various materials is counted regularly, and inventory information is counted irregularly according to the needs of staff. At the same time, when materials are put into or taken out of the warehouse, records are made (i.e., inventory in and out information is recorded) to complete the inventory measurement process.

[0069] When counting inventory, a variety of statistical methods can be used. The statistical method can be to update the inventory by reweighing the materials (this method is suitable for situations where the inventory quantity is small), or to summarize and calculate the historical in and out records and historical inventory to count the inventory. A circle of tracks can also be installed above the coal pit, and the coal pan moves along the track to shoot video data of the coal pit. The video data is identified by a preset AI image recognition model to obtain the inventory quantity of each coal pit. The preset AI image recognition model can determine the volume of the inventory in the coal pit based on the video data of the coal pit. The inventory quantity in the coal pit = the volume of the inventory / coal pit capacity × the total inventory quantity when the coal pit is full of materials. Any two of the above methods can also be used for statistics at the same time, so as to continuously train the AI ​​image recognition model and reduce errors.

[0070] like Figure 6 As shown, as an optional implementation of this embodiment, when purchasing, a purchase plan is determined based on inventory information and material consumption information, including (steps S21 to S27): Step S21, determining the inventory quantity of each material based on the inventory information; Step S22: determining the current material consumption rate based on the material consumption information; Step S23, obtaining the historical material consumption rate of the current cycle; Step S24, calculating an estimated material consumption rate based on the historical material consumption rate, the current material consumption rate and a preset weight; Step S25: determining the estimated inventory quantity of various materials at various times based on the inventory quantity and the estimated material consumption rate; Step S26, determining the time when the estimated inventory quantity is less than the preset inventory threshold as the latest purchase time; Step S27: Determine a procurement plan based on the latest procurement time and the corresponding material type.

[0071] In this embodiment, the inventory quantity of each material is found from the inventory information, the current material consumption speed is found from the material consumption information, and the historical material consumption speed of the current cycle is obtained from the database (if one month is a cycle and the current month is November, the historical material consumption speed of November each year is obtained), the estimated material consumption speed = the current material consumption speed × the first preset weight + the historical material consumption speed × the second preset weight, the estimated inventory quantity of various materials in the future time = the current inventory quantity of the material - the estimated material consumption speed × (future time - current time); the time when the estimated inventory quantity is less than the preset inventory threshold (pre-set, not specifically limited here) is determined as the latest purchase time; the purchase plan includes a preset time before the latest purchase time (the preset time can be preset according to the actual situation, for example: it can be one week, two weeks, or one month, which is not specifically limited here) to remind the staff to purchase the corresponding material type, wherein the first preset weight and the second preset weight are both preset, and are not specifically limited here.

[0072] When calculating the estimated material consumption rate, you can also use the AI ​​estimated consumption model (same as above, no longer repeated here) to analyze the inventory consumption rate in different time periods, so as to obtain the consumable time at each moment based on the inventory quantity and the inventory consumption rate, and determine the moment when the consumable time is less than the preset consumable time as the latest procurement time.

[0073] The procurement plan also includes a recommended procurement time, which is earlier than the latest procurement time. The recommended procurement time is determined based on the market coal price trend, including: collecting the current price data and supply and demand information of various coal trading markets in real time through API interface or data crawler technology, and collecting historical coal price information within a preset historical period (for example, within three months, half a year, and one year, etc.), training machine learning algorithms (for example, random forests, support vector machines, etc.) through historical coal price information to obtain a price prediction model, inputting the current price data and supply and demand information of various coal trading markets into the price prediction model, and obtaining the predicted coal price at various times, and determining the time when the predicted coal price is lower than the preset price and earlier than the latest procurement time as the recommended procurement time, so that the staff can purchase materials according to the recommended procurement time, thereby reducing procurement costs.

[0074] Step S107, coal blending management: determine the coal blending plan based on the AI ​​calculation model.

[0075] Among them, the coal blending plan includes the coal pit number and the quantity of coal to be blended.

[0076] When production is needed in the coking plant, coal blending demand will be generated. The coal blending demand includes the coal quantity and coal blending demand parameters (for example, moisture, dry ash content Ad, volatile matter Vdaf, etc.). The coal blending plan is determined by the material parameters corresponding to various material types in the current inventory and the AI ​​calculation model, thereby improving the coal blending efficiency.

[0077] like Figure 7 As shown, as an optional implementation of this embodiment, the coal blending scheme is determined based on the AI ​​calculation model, including (steps S31 to S36): Step S31, obtaining the coal blending quantity, coal blending requirement parameters and material parameters of various material types; Step S32: determining the coal blending ratio of various material types based on the AI ​​calculation model; Step S33: determining the required quantity of various material types based on the coal blending ratio and coal blending quantity; Step S34, obtaining the coal pit number corresponding to each material type and the quantity of material in each coal pit; Step S35, determining the coal quantity for each coal pit number based on the material quantity and the required quantity; Step S36: determine the coal blending plan based on the coal blending ratio and coal blending quantity.

[0078] In this embodiment, the coal blending quantity, coal blending demand parameters and material parameters of various material types are obtained from the database, and the coal blending quantity, coal blending demand parameters and material parameters of various material types are input into the AI ​​calculation model to obtain the coal blending ratio of various material types; the demand quantity of various material types is calculated according to the coal blending ratio and the coal blending quantity.

[0079] Obtain the coal pit numbers corresponding to various material types and the material quantities in each coal pit from the database, calculate the inventory differences in each coal pit corresponding to the same material type, if there is an inventory difference that is greater than the demand quantity of this material type, then the coal pit number with the largest inventory quantity is determined as the coal blending coal pit number, and the coal blending quantity of this coal blending coal pit number is the demand quantity of this material type; if there is no inventory difference that is greater than the demand quantity of this material type, then the coal blending quantity in each coal pit corresponding to this material type = the inventory quantity in the coal pit number / the total inventory quantity of this material type × the demand quantity of this material type, and the coal blending plan includes the coal blending ratio of various material types and the coal blending quantity of each coal pit number.

[0080] like Figure 8 As shown, as an optional implementation of this embodiment, before determining the coal blending scheme based on the AI ​​calculation model, the method further includes (steps S41 to S43): Step S41, obtaining historical coal blending information; Historical coal blending information includes historical demand parameters, historical material parameters and historical coal blending ratios; Step S42: dividing the historical coal blending information according to the types of historical demand parameters to obtain multiple historical coal blending information combinations; Step S43: training the coal blending algorithm based on multiple historical coal blending information combinations to obtain an AI calculation model.

[0081] In this embodiment, historical coal blending information is obtained from a database, and historical coal blending information with the same type of historical demand parameters is divided into a historical coal blending information combination to obtain multiple historical coal blending information combinations. The coal blending algorithms are trained using the multiple historical coal blending information combinations to obtain an AI computing model, wherein the coal blending algorithm can be a linear programming algorithm, a genetic algorithm, a support vector machine algorithm, and a neural network algorithm, and the type of algorithm is not specifically limited here.

[0082] Step S108, data management: perform statistical analysis on the data and generate data reports.

[0083] In this embodiment, user needs are obtained, and user needs include irregular needs and regular needs. Various data in the database are statistically analyzed based on user needs to generate analysis results, and the analysis results are visualized. The analysis results can be various data reports. When performing data analysis, the data can be analyzed through preset tools, that is, AI technology or data analysis tools, to find abnormalities in the data. If there are abnormalities in the analysis results, early warning information is generated. For example: within a period of time, the material consumption rate exceeds the preset consumption rate, an abnormal warning will be issued, so that the staff can further verify and find the cause of the abnormality based on the early warning information, which is helpful to timely discover and handle the abnormality. Among them, the data analysis tool can be Excel, Python, or SQL, which is not specifically limited here.

[0084] Step S109, logistics monitoring: monitoring the location of the vehicle and estimating the arrival time of the vehicle.

[0085] The historical transportation data of the vehicle is obtained from the database. The historical transportation data includes but is not limited to the vehicle type, transportation cargo information, weather and traffic data, etc. The AI ​​computing algorithm is trained with the historical transportation data to obtain an arrival time prediction model. The AI ​​computing algorithm can be a neural network algorithm or a support vector machine algorithm, which is not specifically limited here.

[0086] Obtain vehicle transportation data from the vehicle, including the vehicle location and destination. Collect weather data and traffic data at the vehicle location at the same time. Input the vehicle location, weather data, traffic data and destination at each moment into the arrival time prediction model to estimate the vehicle's arrival time so that staff can prepare for receiving the goods in advance. By monitoring the vehicle location, it is also easier to query the distribution of vehicles and the transportation status of waybills.

[0087] Step S110, light guidance: in response to the vehicle entering the coal bunker, a guide path is drawn and the coal unloading position is indicated by laser lights on the travel route in the coal bunker.

[0088] In this embodiment, when the vehicle enters the coal bunker at the unloading site, the route inside the coal bunker is found from the driving route, and the lighting equipment is controlled to project a guide path on the route through laser lights, so that the driver can determine the route more clearly, and a sign (such as the word "unloading") is projected above the coal unloading position through laser lights to indicate the coal unloading position.

[0089] Step S111, material sampling: generating sampling instructions based on sampling points and sampling times, and generating packaging instructions based on packaging parameters.

[0090] It should be noted that step S111 corresponding to material sampling and step S104 corresponding to unmanned measurement have different sequences according to different application scenarios or material types, which are not specifically limited here.

[0091] Material sampling includes manual sampling and automatic sampling. Manual sampling means that in some application scenarios, staff manually operate the sampling equipment to take samples. Automatic sampling means that after the driver arrives at the designated location and swipes his ID card, the sampling equipment automatically takes samples according to the generated sampling instructions. The samples obtained by the sampling equipment will be transported to the designated location through funnels, conveyor belts and other components. The designated location includes but is not limited to small coke ovens (for example: determining the quality of coal samples through experiments in small coke ovens) and laboratories (samples are tested through various testing equipment in the laboratory).

[0092] As an optional implementation of this embodiment, Fig. 9 As shown, when automatic sampling is performed, a sampling instruction is generated based on the sampling point and the sampling times, and a packing instruction is generated based on the packing parameters, including (steps S51 to S57): Step S51, obtaining vehicle monitoring data, material monitoring data and vehicle quantity; The number of vehicles is the number of transport vehicles corresponding to one waybill information; Step S52, determining the sampling times for each vehicle based on the number of vehicles; Step S53: performing image recognition on the vehicle monitoring data to determine the sampling range; The sampling range is the range where the material is located; Step S54, performing image recognition on the material monitoring data, and determining the sampling point based on the recognition result, the sampling number and the sampling range; Step S55, generating a sampling instruction based on the sampling point and the sampling times; Step S56, obtaining packaging parameters; Step S57: Generate a packing instruction based on the packing parameters.

[0093] In this embodiment, sampling and testing are required at the sampling location. Video surveillance equipment, such as a binocular camera, is installed at the sampling location to capture the vehicle and materials on the vehicle at the sampling location. Vehicle monitoring data and material monitoring data are obtained from the video surveillance equipment. The number of vehicles in the waybill information corresponding to the current vehicle is obtained from the database. The total number of samples of the material type in each waybill information is preset. The number of samples for each vehicle = the total number of samples of the material type of the vehicle / the number of vehicles corresponding to the waybill information. The preset target recognition model is used to identify whether there is a transport vehicle (such as a truck) in the vehicle monitoring data. If there is a transport vehicle, the preset semantic segmentation model is used to identify the key position of the bottom of the transport vehicle to obtain the height of the bottom of the vehicle, thereby obtaining the height of the bottom material, that is, the depth to which the sampling equipment can descend, so that the possibility of damaging the vehicle is reduced while the bottom sample can be obtained. The height of the carriage is calculated by the preset 3D vision algorithm in the binocular camera, the range of the materials is obtained according to the height of the bottom of the car and the height of the carriage, and the range of the materials is determined as the sampling range. The material monitoring data is image recognized by the AI ​​image recognition model to obtain the recognition result, which includes but is not limited to the moisture content of the materials at each position. The sampling number and sampling position are selected according to the preset rules, and the coordinates corresponding to the sampling positions in the sampling range are determined as the sampling points, and the sampling instructions are generated. The sampling instructions include the sampling number and the sampling points of each sampling, so that the sampling equipment can automatically complete the sampling process according to the sampling instructions without human intervention, obtain the packaging parameters from the staff or the database, and generate the packaging instructions according to the packaging parameters, so that the packaging equipment can automatically complete the packaging process according to the packaging instructions without human intervention, wherein the AI ​​image recognition model is a pre-trained machine learning model.

[0094] Step S112, material testing: obtaining sample testing data, analyzing the sample testing data, and generating abnormal warning information.

[0095] In this embodiment, material testing includes manual testing and unmanned testing. Manual testing includes but is not limited to coding, sample preparation, weighing, testing, system entry, submission for review, archiving, and report generation. Each link needs to form a paper document. In the coding link, the sample is bagged first and then encoded. A part of the encoded sample is retained to facilitate future re-inspection, and the other part is used for subsequent sample preparation. The coding rules can be customized according to actual conditions, such as: year, month, day, serial number, project name, etc.; the sample preparation link grinds the sample and packs it into multiple small bags of samples; the weighing link puts each small bag of sample into a balance for weighing, and the testing link puts the sample into various testing equipment for testing; the system entry link enters the data of the above links into the software; the submission for review link submits the review of each data through the software; the archiving and report generation links archive the data of the above links and generate various reports.

[0096] The above-mentioned weighing link and each subsequent link can also be completed through unmanned steps, that is, unmanned testing through robots, for example: the weighing links through unmanned testing include but are not limited to sample entry, weighing sample picking, sample placement, opening the balance, weighing samples, reading and recovering samples, wherein the sample entry process scans the sample QR code through the camera installed on the robot to enter the sample information; in the weighing sample picking link, the robot picks up the sample in the sample preparation area to transfer the sample; in the sample placement link, the robot puts the picked sample into the preset weighing preparation area; a protective cover is provided on the balance, and in the opening balance link, the robot operates to open the entrance of the protective cover; in the weighing sample link, the robot picks up the sample and puts the sample into the balance; in the reading link, the camera installed on the robot scans the balance data to obtain the weight of the sample; when all samples are weighed, the sample recovery link is carried out, that is, the robot recovers the sample and places it in the completion area.

[0097] When the robot automatically completes each link of the unmanned test, the camera installed on the robot collects image data, and the image data is processed by computer vision algorithms (such as SIFT, SURF, ORB, etc.) to extract key features in the image. The key features include but are not limited to shape, color, texture, etc. The extracted features are compared with a pre-stored database through machine learning algorithms (such as support vector machines SVM, decision trees, random forests, etc.) or deep learning algorithms (such as convolutional neural networks CNN) to confirm the identity and position of samples, scales, etc. After determining the identity and position of the target object in each link, the robot's movement path is planned according to the robot's current position, the position of the target object and the collision detection algorithm (such as space segmentation algorithm, GJK and EPA algorithm, etc.), and the robot movement is controlled by a preset path planning algorithm (such as A* algorithm, Dijkstra algorithm, etc.) to complete the unmanned test steps.

[0098] During the testing process, the electronic equipment can directly obtain sample testing data from various testing equipment. If there is a testing equipment that cannot directly transmit data to the electronic equipment, the robot can also automatically complete the scanning and recording process of the sample testing data, and store the sample testing data in the database. When the electronic equipment collects the sample testing data, it determines whether there is any abnormality in the sample testing data. When there is an abnormality in the sample testing result, it will generate abnormal warning information and warn the staff. The abnormal warning information includes sample information and abnormal sample testing data. The sample information includes transport vehicle information and sample type.

[0099] Fig.10 A structural block diagram of an electronic device 200 provided in an embodiment of the present application.

[0100] like Fig.10 As shown, the electronic device 200 includes a processor 201 and a memory 202 , and may further include an information input / information output (I / O) interface 203 , one or more of a communication component 204 , and a communication bus 205 .

[0101] Among them, the processor 201 is used to control the overall operation of the electronic device 200 to complete all or part of the steps of the above-mentioned AI material control method; the memory 202 is used to store various types of data to support the operation of the electronic device 200, and these data may include, for example, instructions for any application or method used to operate on the electronic device 200, and application-related data. The memory 202 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM), programmable read-only memory (Programmable Read-Only Memory, PROM), read-only memory (Read-Only Memory, ROM), magnetic memory, flash memory, disk or optical disk. One or more.

[0102] The I / O interface 203 provides an interface between the processor 201 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 204 is used for wired or wireless communication between the electronic device 200 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 204 can include: Wi-Fi components, Bluetooth components, NFC components.

[0103] The electronic device 200 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the AI ​​material control method given in the above embodiment.

[0104] The communication bus 205 may include a path to transmit information between the above components. The communication bus 205 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 205 may be divided into an address bus, a data bus, a control bus, etc.

[0105] The electronic device 200 may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc., and may also be servers, etc.

[0106] The present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned AI material management and control method are implemented.

[0107] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0108] The terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0109] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of application involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the aforementioned application concept. For example, the above features are replaced with (but not limited to) technical features with similar functions applied in the present application.

Claims

1. An AI material control method, characterized in that: include: Driver management: obtaining driver information, identifying the driver identity and driver status based on the driver information, and matching the vehicle identity and vehicle status based on the driver identity and driver status; Factory entry management: determining the order of vehicles entering the factory based on the information of vehicles to be entered into the factory, wherein the information of vehicles to be entered into the factory refers to the information of vehicles whose status is waiting to enter the factory; Route guidance: planning the driving routes for vehicles entering the factory; Unmanned weighing: guide the vehicle to be weighed and obtain the vehicle's measurement data; Factory management: Verify the vehicle's waybill completion status based on measurement data and generate exception information; The method further comprises: Inventory management: collect inventory information and formulate purchasing plans based on the inventory information; Coal blending management: Determine the coal blending plan based on the AI ​​calculation model, which includes the coal pit number and the amount of coal to be blended; Data management: perform statistical analysis on data and generate data reports; Logistics monitoring: monitor the location of vehicles and estimate their arrival time; Light guidance: In response to vehicles entering the coal bunker, laser lights are used to indicate the approach and coal unloading location on the route inside the coal bunker; Material sampling: Generate sampling instructions based on sampling points and sampling times, and generate packaging instructions based on packaging parameters; Material testing: Obtain sample testing data, analyze the sample testing data, and generate abnormal warning information.

2. The method according to claim 1, characterized in that: The determining of the order of vehicles entering the factory based on the information of vehicles to be entered into the factory comprises: Determine the material type and waiting time of each vehicle based on the information of the vehicles to be transported into the factory; Obtain inventory information and material consumption information; Sorting the material types based on the inventory information and the material consumption information to obtain a first sorting result; Determine the order of vehicles entering the factory based on the first sorting result and the waiting time; The sorting of the material types based on the inventory information and the material consumption information to obtain a first sorting result includes: Determine a current material consumption rate based on the material consumption information; Determine the inventory quantity of each material based on the inventory information; Determine the consumable time based on the inventory quantity and the current material consumption speed; The material types are sorted based on the consumable time to obtain a first sorting result.

3. The method according to claim 1, characterized in that The planning of the driving route of the vehicle entering the factory includes: Obtain the material type corresponding to the vehicle entering the factory; Determine candidate locations for each step based on the material type and the preset process; Obtaining current queue information of each of the candidate positions; Calculate the distance between each two candidate positions in two adjacent steps; Determine the target position of each step based on the current queue information and the distance; Sorting the target locations according to the preset process to obtain a driving route; The guiding vehicle to be weighed and obtaining the vehicle's measurement data includes: Obtaining the vehicle identity based on a measurement method, wherein the measurement method includes on-vehicle measurement and off-vehicle measurement; Acquire vehicle images; Identify the vehicle image based on a preset AI image recognition model to obtain driver information and vehicle information; If the driver's seat information and the vehicle information both meet the preset measurement rules, then obtain the vehicle's measurement data; The verification of the vehicle's waybill completion status based on the metering data and the generation of abnormal information include: verifying the identity of the vehicle; Calculating the difference of the measurement data; comparing the difference with the net weight of the material; If the difference is the same as the net weight, then a confirmation factory information is generated; If the difference is not the same as the net weight, an exception message is generated.

4. The method according to claim 1, characterized in that: The collecting inventory information and formulating a purchase plan based on the inventory information includes: Inventory measurement: count inventory information and record inventory in and out information; Procurement: Determine the procurement plan based on the inventory information and material consumption information; The determining of the procurement plan based on the inventory information and the material consumption information includes: Determine the inventory quantity of each material based on the inventory information; Determine a current material consumption rate based on the material consumption information; Get the historical material consumption rate of the current cycle; Calculate an estimated material consumption rate based on the historical material consumption rate, the current material consumption rate and a preset weight; Determine the estimated inventory quantity of various materials at various times based on the inventory quantity and the estimated material consumption rate; Determine the time when the estimated inventory quantity is less than a preset inventory threshold as the latest purchase time; A procurement plan is determined based on the latest procurement time and the corresponding material type.

5. The method according to claim 1, characterized in that The coal blending scheme is determined based on the AI ​​calculation model, including: Obtain the coal blending quantity, coal blending requirement parameters and material parameters of various material types; Determine the coal blending ratio of various material types based on the AI ​​calculation model; Determine the required quantity of various types of materials based on the coal blending ratio and the coal blending quantity; Get the coal pit number corresponding to each material type and the quantity of materials in each coal pit; Determine the coal quantity for each coal pit number based on the material quantity and the required quantity; Determine a coal blending scheme based on the coal blending ratio and the coal blending quantity; Before determining the coal blending scheme based on the AI ​​computing model, the method further includes: Acquire historical coal blending information, wherein the historical coal blending information includes historical demand parameters, historical material parameters, and historical coal blending ratios; Dividing the historical coal blending information according to the types of the historical demand parameters to obtain a plurality of historical coal blending information combinations; The coal blending algorithm is trained based on multiple combinations of the historical coal blending information to obtain the AI ​​calculation model.

6. The method according to claim 1, characterized in that The statistical analysis of the data and generation of data reports include: Obtaining user needs, wherein the user needs include irregular needs and regular needs; Analyze the data based on the user needs and preset tools, and visualize the analysis results, which include the data report; If there is an abnormality in the analysis result, a warning message is generated; The monitoring of the vehicle's location and estimating the vehicle's arrival time include: Obtain historical transportation data of vehicles; The AI ​​calculation algorithm is trained based on the historical transportation data to obtain an arrival time prediction model; Acquiring vehicle transportation data, weather data, and traffic data, wherein the vehicle transportation data includes a vehicle location and a destination; Inputting the vehicle location, the weather data, the traffic data, and the destination into the arrival time prediction model to estimate the arrival time of the vehicle; In response to a vehicle entering a coal bunker, a guide path is drawn on the route in the coal bunker by means of laser lights and a coal unloading position is indicated, including: In response to the vehicle entering the coal bunker, determining a travel route within the coal bunker based on the travel route; An approach road is marked on the travel route by means of laser lights, and the coal unloading position is indicated by means of laser lights.

7. The method according to claim 1, characterized in that The step of generating a sampling instruction based on a sampling point and a sampling number, and generating a packing instruction based on a packing parameter, comprises: Obtain vehicle monitoring data, material monitoring data and vehicle quantity, where the vehicle quantity is the number of transport vehicles corresponding to one waybill information; determining a sampling number for each vehicle based on the number of vehicles; Performing image recognition on the vehicle monitoring data to determine a sampling range, where the sampling range is the range where the material is located; Performing image recognition on the material monitoring data, and determining a sampling point based on the recognition result, the sampling times and the sampling range; Generate a sampling instruction based on the sampling point and the sampling number; Get packaging parameters; Generate a packing instruction based on the packing parameters; The obtaining of sample test data, analyzing the sample test data, and generating abnormal warning information includes: Acquire and store sample analysis data; Determine whether the sample test data is abnormal; If there is an abnormality in the sample test data, abnormality warning information is generated.

8. An AI material control system, characterized in that: include: A camera, used to collect image data or video data at each location; ID card recognition equipment, used to recognize ID card information; Voice guidance equipment, used to play preset audio when the vehicle reaches a specified location or completes a specified step; Automatic weighing equipment, used to weigh vehicles or materials and obtain measurement data; Sampling equipment for sampling materials on the vehicle; Packaging equipment, used to pack samples; Testing equipment, used to test materials according to preset requirements and obtain sample testing data; Coal pan meter, used to collect images of materials in the coal bunker, Lighting equipment, used to emit laser lights to guide the vehicle's route and indicate the coal unloading location; Vehicle monitoring equipment, used to monitor the status of vehicles and materials in real time and obtain monitoring data, the vehicle monitoring equipment includes GPS equipment and vehicle-mounted sensors; Handheld terminal devices are used by drivers or staff to view and operate information; The electronic device is used to obtain the image data, the video data, the measurement data, the sample test data, the material image, the monitoring data and the operation information, and is also used to control the voice guidance device to play audio, control the sampling device to sample, control the packaging device to package, and control the lighting device to emit laser light.

9. An electronic device, characterized in that: comprising a processor coupled to a memory; The processor is configured to execute a computer program stored in the memory, so that the electronic device executes the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The method comprises a computer program or an instruction, which, when executed on a computer, causes the computer to execute the method according to any one of claims 1 to 7.

Images