Self-adaptive electromagnetic control system for magnetic attraction steel plate material transportation based on machine vision driving

Abstract

The invention discloses a self-adaptive electromagnetic control system based on machine vision traveling and magnetic attraction steel plate conveying. The self-adaptive electromagnetic control system comprises a moving trolley (6), a connecting rod (7), a camera (8) and a magnetic force control system. The magnetic force control system is used for controlling the magnetic force of the travelling crane grabbing mechanism (1); on the basis of the AI algorithm, the hoisting task processing efficiency is improved, the labor cost is reduced, various plate hoisting application scenes are supported, and unmanned and intelligent development of travelling cranes is promoted.

Description

Technical Field

[0001] The invention relates to the technical field of cranes, and in particular to an adaptive electromagnetic control system for magnetically sucking steel plates for crane transportation based on machine vision. Background Art

[0002] With the rapid development of intelligent manufacturing technology, the steel industry has an increasingly urgent need for efficient, accurate and safe steel plate handling and sorting technology. Traditional steel plate handling methods have problems such as low efficiency, high labor costs and great safety hazards. Although the existing permanent magnetic lifting technology has improved the handling efficiency to a certain extent, it is slightly insufficient in terms of intelligence.

[0003] In the process of transporting steel plates, in order to avoid damage to the steel plates, a crane equipped with an electromagnet is often used for adsorption and lifting operations. During the current manual operation, a safety operator is required to stand under the lifting operation to observe the bending degree of the steel plate to determine whether the lifting operation can be performed. The labor cost is high, and failures may occur in the middle of the process, which threatens the lives of the observers and is prone to accidents.

[0004] During the permanent magnet lifting process of the crane, the crane operator currently increases the current to the maximum limit to maintain the maximum magnetic force, and relies on the safety personnel to observe the bending degree of the steel plate and adjust the current appropriately to complete the electromagnet steel plate adsorption operation. The entire lifting process often relies on the work experience of the crane operator, and the degree of intelligence is relatively low. Summary of the invention

[0005] The purpose of the present invention is to provide an adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision, including a mobile trolley, a connecting rod, a camera, and a magnetic control system;

[0006] The magnetic force control system is used to control the magnetic force of the crane grabbing mechanism;

[0007] The crane grabbing mechanism is used to lift one or more steel plates;

[0008] The mobile trolley is used to fix and transport the camera so that the camera is located below the bottom steel plate of the lifting target;

[0009] The camera captures the real-time image of the steel plate lifted by the crane and transmits it to the magnetic control system;

[0010] The magnetic control system recognizes the real-time image of the steel plate lifted by the crane to obtain the edge position of the bottom steel plate;

[0011] The magnetic control system calculates the bending curvature of the bottom steel plate based on the edge position of the bottom steel plate, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold;

[0012] After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the lifting crane, causing the bottom steel plate to fall off until the bending curvature of the bottom steel plate is less than the preset bending threshold.

[0013] Further, the connecting rod comprises a first connecting rod and a second connecting rod connected in an L-shape;

[0014] One end of the first connecting rod is fixed on the moving trolley, and the other end is connected to the second connecting rod;

[0015] The first connecting rod is height-adjustable in the vertical direction;

[0016] The length of the second connecting rod is adjustable in the horizontal direction.

[0017] Further, the vehicle grabbing mechanism includes a fixed plate and a plurality of electromagnetic disks connected to the bottom;

[0018] The magnetic force control system is used to control the magnetic force of the electromagnetic disk;

[0019] The electromagnetic disk is used for adsorbing the steel plate.

[0020] Furthermore, the lifting crane also includes auxiliary cameras distributed in the four corners of the crane.

[0021] The auxiliary camera compensates and corrects the blind area of ​​the camera.

[0022] Further, the steps of controlling the magnetic force of the crane grabbing mechanism by the magnetic control system include:

[0023] 1) Determine whether the top steel plate is the target lifting plate, if so, proceed to step 2), otherwise, proceed to step 3);

[0024] 2) The magnetic control system calculates the electromagnetic force F required for lifting 实 , thereby controlling the crane grabbing mechanism to lift, ending the control, and entering step 4);

[0025] 3) The magnetic control system sets the electromagnetic force to the maximum value, controls the crane grabbing mechanism to lift, lifts the non-target lifting plate away, and returns to step 1);

[0026] 4) Use the camera to capture the real-time image of the steel plate lifted by the crane and transmit it to the magnetic control system;

[0027] The magnetic control system calculates the bending curvature of the bottom steel plate based on the real-time image of the steel plate lifted by the crane, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold;

[0028] 5) After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the lifting crane to make the bottom steel plate fall off, and returns to step 4) until the bending curvature of the bottom steel plate is less than the preset bending threshold.

[0029] Furthermore, the electromagnetic force F required for lifting 实 As shown below:

[0030] F 实 =αF 磁 +β

[0031] Among them, α is the error coefficient, β is the electromagnetic force bias; F 磁 is the theoretical value of electromagnetic force.

[0032] Furthermore, the theoretical value of electromagnetic force F 磁 As shown below:

[0033]

[0034] Where B is the magnetic induction intensity, S is the magnetic pole area, μ 0 is the magnetic permeability.

[0035] Furthermore, the error coefficient α and the electromagnetic force bias β are obtained by fitting historical data through an AI model.

[0036] Furthermore, the preset bending threshold is determined by the material and size of the steel plate.

[0037] The technical effect of the present invention is undoubted, and the beneficial effects of the present invention are as follows:

[0038] 1) Improve the efficiency of lifting task processing

[0039] The present invention mainly uses artificial intelligence models, takes the plate model, hoisting quantity, ambient temperature and humidity as input, and uses reinforcement learning and other methods. On the basis of deriving the magnetic theoretical value based on electromagnetic theory, the AI ​​model is used to solve the electromagnetic force bias, and the plate hoisting arc is identified through machine vision to achieve adaptive magnetic correction. This invention will realize the automation and intelligence of the crane steel plate hoisting process and improve the efficiency of hoisting task processing.

[0040] 2) Reduce labor costs

[0041] The present invention uses a small car carrying a camera to replace the observation safety officer under the plate, which reduces the labor cost. Based on the support of the algorithm, it can realize fully automatic unmanned driving in the future, further reducing the labor cost.

[0042] 3) Support various plate lifting application scenarios

[0043] The present invention is applicable to various plate hoisting scenarios. Various plates are not limited to volume differences such as length, width and height, but also include material differences. The bending arc threshold can be determined by trial hoisting, thereby realizing multi-scenario hoisting applications.

[0044] In summary, the present invention is based on the AI ​​algorithm, improves the efficiency of lifting task processing, reduces labor costs, supports a variety of plate lifting application scenarios, and promotes the development of unmanned and intelligent driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is the schematic diagram of the driving electromagnet system;

[0046] Figure 2 It is the flow chart of electromagnet adaptive magnetic adsorption;

[0047] In the figure, there are lifting plate a, steel plate b, crane lifting device 1, crane fixing plate 2, crane adsorption electromagnet 3, main camera connecting rod 4, main camera 5, unmanned vehicle 6, auxiliary camera connecting rod 7, and auxiliary camera 8. DETAILED DESCRIPTION

[0048] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.

[0049] Embodiment 1:

[0050] See also Figure 1 to Figure 2 , an adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision, including a mobile car 6, a connecting rod 7, a camera 8, and a magnetic control system;

[0051] The magnetic force control system is used to control the magnetic force of the vehicle grabbing mechanism 1;

[0052] The crane grabbing mechanism 1 is used to lift one or more steel plates;

[0053] The mobile trolley 6 is used to fix and transport the camera 8 so that the camera 8 is located below the bottom steel plate of the lifting target;

[0054] The camera 8 captures the real-time image of the steel plate lifted by the crane and transmits it to the magnetic control system;

[0055] The magnetic control system recognizes the real-time image of the steel plate lifted by the crane to obtain the edge position of the bottom steel plate;

[0056] The magnetic control system calculates the bending curvature of the bottom steel plate based on the edge position of the bottom steel plate, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold;

[0057] After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the lifting crane, causing the bottom steel plate to fall off until the bending curvature of the bottom steel plate is less than the preset bending threshold.

[0058] The connecting rod 7 includes a first connecting rod and a second connecting rod connected in an L-shaped manner;

[0059] One end of the first connecting rod is fixed on the moving trolley 6, and the other end is connected to the second connecting rod;

[0060] The first connecting rod is height-adjustable in the vertical direction;

[0061] The length of the second connecting rod is adjustable in the horizontal direction.

[0062] The vehicle grabbing mechanism 1 comprises a fixing plate 2 and a plurality of electromagnetic disks 3 connected to the bottom;

[0063] The magnetic force control system is used to control the magnetic force of the electromagnetic disk 3;

[0064] The electromagnetic disk 3 is used for adsorbing the steel plate.

[0065] The lifting crane also includes auxiliary cameras 5 distributed in the four corners of the crane.

[0066] The auxiliary camera 5 compensates for the blind spot of the camera 8 .

[0067] The steps of the magnetic control system controlling the magnetic force of the crane grabbing mechanism 1 include:

[0068] 1) Determine whether the top steel plate is the target lifting plate, if so, proceed to step 2), otherwise, proceed to step 3);

[0069] 2) The magnetic control system calculates the electromagnetic force F required for lifting 实 , thereby controlling the crane grabbing mechanism 1 to lift, and proceeding to step 4);

[0070] 3) The magnetic control system sets the electromagnetic force to the maximum value, controls the crane grabbing mechanism 1 to lift, lifts the non-target lifting plate away, and returns to step 1);

[0071] 4) Using the camera 8 to capture the real-time image of the steel plate lifted by the crane, and transmit it to the magnetic control system;

[0072] The magnetic control system calculates the bending curvature of the bottom steel plate based on the real-time image of the steel plate lifted by the crane, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold;

[0073] 5) After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the lifting crane to make the bottom steel plate fall off, and returns to step 4 until the bending curvature of the bottom steel plate is less than the preset bending threshold.

[0074] Electromagnetic force F required for lifting 实 As shown below:

[0075] F 实 =αF 磁 +β

[0076] Among them, α is the error coefficient, β is the electromagnetic force bias; F 磁 is the theoretical value of electromagnetic force.

[0077] Theoretical value of electromagnetic force F 磁 As shown below:

[0078]

[0079] Where B is the magnetic induction intensity, S is the magnetic pole area, μ 0 is the magnetic permeability.

[0080] The error coefficient α and electromagnetic force bias β are obtained by fitting historical data through an AI model.

[0081] The preset bending threshold is determined by the material and size of the steel plate.

[0082] Embodiment 2:

[0083] An adaptive electromagnetic control system for magnetically sucking steel plates for transporting materials based on machine vision, comprising a mobile trolley 6, a connecting rod 7, a camera 8, and a magnetic control system;

[0084] The magnetic force control system is used to control the magnetic force of the vehicle grabbing mechanism 1;

[0085] The crane grabbing mechanism 1 is used to lift one or more steel plates;

[0086] The mobile trolley 6 is used to fix and transport the camera 8 so that the camera 8 is located below the bottom steel plate of the lifting target;

[0087] The camera 8 captures the real-time image of the steel plate lifted by the crane and transmits it to the magnetic control system;

[0088] The magnetic control system recognizes the real-time image of the steel plate lifted by the crane to obtain the edge position of the bottom steel plate;

[0089] The magnetic control system calculates the bending curvature of the bottom steel plate based on the edge position of the bottom steel plate, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold;

[0090] After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the lifting crane, causing the bottom steel plate to fall off until the bending curvature of the bottom steel plate is less than the preset bending threshold.

[0091] Embodiment 3:

[0092] An adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision, the technical content is the same as that of embodiment 2, further, the connecting rod 7 includes a first connecting rod and a second connecting rod connected in an L-shape;

[0093] One end of the first connecting rod is fixed on the moving trolley 6, and the other end is connected to the second connecting rod;

[0094] The first connecting rod is height-adjustable in the vertical direction;

[0095] The length of the second connecting rod is adjustable in the horizontal direction.

[0096] Embodiment 4:

[0097] An adaptive electromagnetic control system for magnetically sucking steel plates for transporting materials based on machine vision, the technical content of which is the same as any one of Embodiments 2-3, and further, the crane grabbing mechanism 1 includes a fixed plate 2 and a plurality of electromagnetic disks 3 connected to the bottom;

[0098] The magnetic force control system is used to control the magnetic force of the electromagnetic disk 3;

[0099] The electromagnetic disk 3 is used for adsorbing the steel plate.

[0100] Embodiment 5:

[0101] An adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision crane, the technical content is the same as any one of embodiments 2-4, further, the lifting crane also includes auxiliary cameras 5 distributed in the four corners of the crane. The auxiliary camera 5 compensates and corrects the blind spot of the camera 8.

[0102] Embodiment 6:

[0103] An adaptive electromagnetic control system for magnetically sucking steel plates for transporting materials based on machine vision, the technical content of which is the same as any one of Embodiments 2-5, and further, the steps of controlling the magnetic force of the magnetic control system for the crane grabbing mechanism 1 include:

[0104] 1) Determine whether the top steel plate is the target lifting plate, if so, proceed to step 2, otherwise, proceed to step 3);

[0105] 2) The magnetic control system calculates the electromagnetic force F required for lifting 实 , thereby controlling the crane grabbing mechanism 1 to lift, and proceeding to step 4);

[0106] 3) The magnetic control system sets the electromagnetic force to the maximum value, controls the crane grabbing mechanism 1 to lift, lifts the non-target lifting plate away, and returns to step 1);

[0107] 4) Using the camera 8 to capture the real-time image of the steel plate lifted by the crane, and transmit it to the magnetic control system;

[0108] The magnetic control system calculates the bending curvature of the bottom steel plate based on the real-time image of the steel plate lifted by the crane, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold;

[0109] 5 After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the lifting crane to make the bottom steel plate fall off, and returns to step 4 until the bending curvature of the bottom steel plate is less than the preset bending threshold.

[0110] Embodiment 7:

[0111] An adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision, the technical content is the same as any one of embodiments 2-6, further, the electromagnetic force F required for lifting 实 As shown below:

[0112] F 实 =αF 磁 +β

[0113] Among them, α is the error coefficient, β is the electromagnetic force bias; F 磁 is the theoretical value of electromagnetic force.

[0114] Embodiment 8:

[0115] An adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision, the technical content is the same as any one of embodiments 2-7, further, the electromagnetic force theoretical value F 磁 As shown below:

[0116]

[0117] Where B is the magnetic induction intensity, S is the magnetic pole area, μ 0 is the magnetic permeability.

[0118] Embodiment 9:

[0119] An adaptive electromagnetic control system for magnetically attracted steel plate transport based on machine vision, the technical content of which is the same as any one of Examples 2-8, and further, the error coefficient α and electromagnetic force bias β are obtained by fitting historical data through an AI model.

[0120] Embodiment 10:

[0121] An adaptive electromagnetic control system for magnetically attracted steel plate transportation based on machine vision, the technical content of which is the same as any one of Examples 2-9, and further, the preset bending threshold is determined by the material and size of the steel plate.

[0122] Embodiment 11:

[0123] The invention discloses an adaptive electromagnetic control system for magnetically attracted steel plate transport based on machine vision, comprising a driving mechanical device with magnetically adjustable force and an unmanned vehicle with a telescopic connecting rod and a camera.

[0124] During the lifting and transportation of steel plates, the main camera 8 carried on the unmanned vehicle is mainly used to detect the number of steel plates and identify the bending degree of the lifting target steel plates.

[0125] The auxiliary camera 5 is mainly installed at the four corners of the vehicle to compensate for the vacant position of the main camera, to further confirm the number and bending degree of the lifted steel plates, and to send the curvature value as feedback to the control system.

[0126] After receiving the bending arc value, the control system calls up the internal database to determine whether it meets the lifting safety standards. If it does not meet the lifting standards, the current value is reduced to allow the last plate to fall, thereby completing the safe lifting operation task.

[0127] During the lifting and transportation of the plates, the unmanned vehicle follows the vehicle and monitors the bending radius at all times to avoid the danger of the plates falling during transportation.

[0128] During the lifting process, the present invention is based on the physics formula of electromagnetic force. On this basis, combined with the artificial intelligence algorithm model, the model of the plate, the surrounding temperature and humidity, and the magnetic bias are fitted to give a more accurate magnetic force size.

[0129] 1) According to electromagnetism, the magnetic force of an electromagnet is:

[0130]

[0131] Where F is the electromagnetic attraction, B is the magnetic induction intensity, S is the magnetic pole area, μ 0 is the magnetic permeability.

[0132] 2) During the actual lifting process, the plate stacking may result in a complex situation where multiple types of plates are mixed. This embodiment only discusses the situation where the target plate is at the top of the plate stack, and the other various situations are supplemented in Embodiment 3.

[0133] The material and model of the plate can be obtained from the prior knowledge of the target task, and the theoretical electromagnetic force requirement can be calculated through the physical model. However, due to factors such as oxidation and moisture of the plate, the actual electromagnetic force required for lifting is different from the theoretical value. The formula is:

[0134] F 实 =αF 磁 +β

[0135] Where α is the error coefficient, which includes the electromagnetic conduction loss between target plates due to environmental factors (humidity, etc.), target property changes (steel surface oxidation), and different plate stacking methods. F 磁 The electromagnet theory calculated by the formula method should provide the magnetic force, β is the electromagnetic force bias, and the difference between the actual electromagnetic force required for safe lifting and the theoretical value.

[0136] However, since the error coefficient α has too many influencing factors and the influence of factors such as environmental humidity is difficult to confirm, the theoretical value of electromagnetic force F 磁 As a benchmark, AI models such as reinforcement learning are used to fit the electromagnetic force bias β and build a crane hoisting electromagnetic force estimation model.

[0137] In actual lifting conditions, the target plate is often not at the top of the plate stack, and other types of plates need to be moved away to achieve the lifting operation of the target plate.

[0138] In this embodiment, there is a lack of prior knowledge about other types of plates, and it is impossible to accurately obtain the model and material of the plates. Therefore, the present invention relies on the feedback signal of the bending curvature of the plate from the main camera on the trolley to simulate the manual operation process.

[0139] First, increase the current to the maximum, determine and identify the number of plates to be hoisted, and wait for the main camera to identify the arc value and confirm that it is within the safe hoisting threshold range before implementing the next step of transportation operation.

[0140] If the bending degree is too large and exceeds the lifting threshold range, the magnetic force will be reduced, allowing the last steel plate to fall freely onto the plate stack, and then the plate transportation and lifting operations can be carried out.

Claims

1. An adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision, characterized by: It includes a mobile car (6), a connecting rod (7), a camera (8), and a magnetic control system; The magnetic force control system is used to control the magnetic force of the crane grabbing mechanism (1); The crane grabbing mechanism (1) is used to lift one or more steel plates; The mobile trolley (6) is used to fix and transport the camera (8) so that the camera (8) is located below the bottom steel plate of the lifting target object; The camera (8) captures a real-time image of the steel plate lifted by the crane and transmits it to the magnetic control system; The magnetic control system recognizes the real-time image of the steel plate lifted by the crane to obtain the edge position of the bottom steel plate; The magnetic control system calculates the bending curvature of the bottom steel plate based on the edge position of the bottom steel plate, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold; After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the crane grabbing mechanism (1) to cause the bottom steel plate to fall off until the bending curvature of the bottom steel plate is less than a preset bending threshold.

2. According to claim 1, the adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision is characterized by: The connecting rod (7) comprises a first connecting rod and a second connecting rod connected in an L-shape; One end of the first connecting rod is fixed on the moving trolley (6), and the other end is connected to the second connecting rod; The first connecting rod is height-adjustable in the vertical direction; The length of the second connecting rod is adjustable in the horizontal direction.

3. According to the adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision driving according to claim 1, it is characterized by: The vehicle grabbing mechanism (1) comprises a fixing plate (2) and a plurality of electromagnetic disks (3) connected to the bottom; The magnetic force control system is used to control the magnetic force of the electromagnetic disk (3); The electromagnetic disk (3) is used for adsorbing the steel plate.

4. According to claim 3, the adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision is characterized in that: The lifting crane also includes auxiliary cameras (5) distributed in the four corners of the crane; The auxiliary camera (5) compensates and corrects the blind spot of the camera (8).

5. According to the adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision driving according to claim 1, it is characterized in that: The steps of controlling the magnetic force of the crane grabbing mechanism (1) by the magnetic control system include: 1) Determine whether the top steel plate is the target lifting plate, if so, proceed to step 2), otherwise, proceed to step 3); 2) The magnetic control system calculates the electromagnetic force F required for lifting 实 , thereby controlling the crane grabbing mechanism (1) to lift and proceed to step 4); 3) The magnetic control system sets the electromagnetic force to the maximum value, controls the crane grabbing mechanism (1) to lift, lifts the non-target lifting plate away, and returns to step 1); 4) Using the camera (8) to capture the real-time image of the steel plate lifted by the crane and transmit it to the magnetic control system; The magnetic control system calculates the bending curvature of the bottom steel plate based on the real-time image of the steel plate lifted by the crane, and sends an overload signal to the magnetic control system if the bending curvature of the bottom steel plate is greater than or equal to a preset bending threshold; 5) After receiving the overload signal, the magnetic control system reduces the electromagnetic force of the lifting crane to make the bottom steel plate fall off, and returns to step 4) until the bending curvature of the bottom steel plate is less than the preset bending threshold.

6. The adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision according to claim 1 is characterized in that: Electromagnetic force F required for lifting 实 As shown below: F 实 =αF 磁 +b Among them, α is the error coefficient, β is the electromagnetic force bias; F 磁 is the theoretical value of electromagnetic force.

7. The adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision according to claim 6 is characterized in that: Theoretical value of electromagnetic force F 磁 As shown below: Among them, B is the magnetic induction intensity, S is the magnetic pole area, and μ0 is the magnetic permeability.

8. The adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision according to claim 6 is characterized in that: The error coefficient α and electromagnetic force bias β are obtained by fitting historical data through an AI model.

9. The adaptive electromagnetic control system for magnetic steel plate transportation based on machine vision according to claim 1 is characterized in that: The preset bending threshold is determined by the material and size of the steel plate.

Images