Intelligent control system for automatic refining car

Abstract

The invention relates to the technical field of aluminum alloy liquid refining, in particular to an intelligent control system for an automatic refining car, which comprises a data acquisition module, an environment sensing module, a material identification module, a central processing unit and an execution terminal. According to the method, the three-dimensional semantic map is generated by analyzing the characteristics of the surrounding environment of the refining site in real time, and the optimal passing route is selected to control the mobile docking of the refining vehicle, so that the mobile flexibility of the refining vehicle is improved, and the docking efficiency is further improved; the three-dimensional impurity distribution diagram is formed by detecting the impurity distribution state in the aluminum alloy liquid, the refining mechanism is flexibly adjusted to conduct refining operation according to the impurity distribution state, the refining efficiency is improved while the refining effect is improved, the problem that refining operation of a traditional refining car is not uniform is solved, and especially under the condition that impurity distribution is not uniform, the refining efficiency is improved. And the refining effect can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of aluminum alloy liquid refining, and more particularly, to an intelligent control system for automatic refining of vehicles. Background Art

[0002] Before casting aluminum alloy, it is necessary to degas and refine the aluminum liquid to obtain high-quality aluminum liquid with less slag content and less gas content for casting. At present, the degassing and refining of aluminum liquid are mainly carried out by an automatic refining vehicle, which, as an automated device, is mainly used to refine metal solutions during the casting process to improve the quality and purity of the metal.

[0003] In the prior art, when the automatic refining vehicle is refining, it mainly executes the refining operation according to the established process set by the operator, and it is difficult to automatically adjust the refining process according to the impurity distribution in the aluminum alloy solution during the refining process, which affects the refining effect. At the same time, the docking efficiency between the automatic refining vehicle and the aluminum alloy liquid furnace is relatively low, and it requires the operator to control it manually, lacking flexibility.

[0004] Based on this, an intelligent control system for an automatic refining vehicle is proposed. Summary of the Invention

[0005] The main object of the present invention is to provide an intelligent control system for an automatic refining vehicle to overcome the problems mentioned in the above background art.

[0006] To achieve the above object, the present invention provides an intelligent control system for an automatic refining vehicle, including an environmental perception module, a material identification module, and a central processor. The central processor includes a first processing unit and a second processing unit:

[0007] The environmental perception module is used to dynamically capture the environmental characteristics around the refining site, perform target recognition on the environmental characteristics to obtain several target bodies and lock the docking body from them, comprehensively analyze the refining vehicle, several target bodies, and the docking body in combination with the environmental characteristic data to construct an activity area, and at the same time construct a three-dimensional semantic map based on the activity area and transmit it to the first processing unit of the central processor;

[0008] The material identification module is used for monitoring the physical parameters and analyzing the composition of the aluminum alloy liquid, including material state feedback and impurity distribution detection;

[0009] Perform state feedback on the aluminum alloy liquid, collect the physical parameters of the aluminum alloy liquid, perform normalization processing and then input them into a computer arithmetic processor for summation calculation to obtain a process evaluation index, and transmit it to the second processing unit of the central processor;

[0010] Detect the impurity distribution of the aluminum alloy liquid, input the internal space data of the docking body container into the computer graphics processor to construct a container space model, locate the impurities through the reflection signal of ultrasonic waves in the aluminum alloy liquid, substitute the impurity location distribution data into the container space model to construct a three-dimensional impurity distribution map, and transmit it to the second processing unit of the central processor;

[0011] The first processing unit of the central processor receives the three-dimensional semantic map transmitted by the environmental perception module, generates several moving passage paths for the refining vehicle to reach the docking body through the three-dimensional semantic map, comprehensively analyzes the several moving passage paths to select the first path, and generates corresponding control instructions;

[0012] The second processing unit of the central processor receives the material state feedback transmitted by the material recognition module and comprehensively analyzes the impurity distribution detection, and generates corresponding control instructions.

[0013] As a further improvement of the present invention, it further includes a data acquisition module and an execution terminal;

[0014] The data acquisition module includes a multimodal sensor assembly one, and the multimodal sensor assembly one is composed of a high-definition camera, a lidar and an ultrasonic radar;

[0015] The multimodal sensor assembly one acquires the environmental characteristics of the objects around the refining site, including position, shape, distance and semantic information data. The high-definition camera performs target recognition on the object shape to obtain several target bodies and locks the docking body from them. The lidar and ultrasonic radar obtain the object positioning points and positioning distance data;

[0016] The data acquisition module includes a multimodal sensor assembly two, and the multimodal sensor assembly two is composed of a high-temperature infrared sensor, a density sensor, a viscosity sensor and a sound wave probe;

[0017] The multimodal sensor assembly two acquires the physical parameters of the aluminum alloy liquid, including the temperature W collected by the infrared sensor, the density M collected by the density sensor and the dynamic viscosity N collected by the viscosity sensor;

[0018] The multimodal sensor assembly two acquires the impurity distribution data of the aluminum alloy liquid, including that the sound wave probe emits high-frequency sound waves to penetrate the aluminum alloy liquid, uses the acoustic impedance difference between the impurities and the matrix material to reflect the echo, and obtains the impurity location distribution data by analyzing the characteristics of the echo signal;

[0019] The specific analysis of the echo signal characteristics is as follows:

[0020] The echo signal characteristics include amplitude, time delay and frequency spectrum;

[0021] The echo amplitude is proportional to the acoustic impedance difference between the impurities and the matrix. The higher the amplitude, the larger the impurity volume. Through the formula Calculate the impurity volume T, where Zi , J i are type parameters representing the impurity amplitude and the matrix amplitude respectively;

[0022] The echo time delay calculates the impurity depth based on the acoustic wave propagation time difference. The impurity depth d is calculated through the formula where c represents the acoustic velocity parameter of the aluminum alloy liquid and t represents the time delay parameter;

[0023] The echo spectrum analysis converts the time-domain signal into the frequency-domain analysis through the fast Fourier transform;

[0024] For example:

[0025] Main frequency shift: Bubbles cause the frequency to shift to the low frequency (e.g., 10 MHz → 8.5 MHz), presenting a narrow frequency band;

[0026] Harmonic attenuation: Hard inclusions (such as SiC) cause the second harmonic to enhance, presenting a wide frequency band;

[0027] Taking the aluminum alloy liquid level in the docking body container as the matrix, through analysis, the impurity volume, depth position and type can be obtained, and the impurity localization distribution data of the aluminum alloy liquid in the docking body container can be obtained.

[0028] The execution terminal is responsible for receiving the control instructions transmitted by the central processing unit and performing corresponding control operations;

[0029] Based on the control instructions generated by the first processing unit of the central processing unit, control the starting of the moving mechanism of the refining vehicle to move automatically to the specified docking point of the docking body along the first path;

[0030] Based on the control instructions generated by the second processing unit of the central processing unit, control the starting of the refining mechanism of the refining vehicle to perform corresponding refining operations.

[0031] As a further improvement of the present invention, the specific process of the environmental perception module constructing the three-dimensional semantic map is as follows:

[0032] Set the refining vehicle and the docking body as the first coordinate point and the second coordinate point respectively, set several target bodies as the third coordinate point, and connect the first coordinate point and the second coordinate point with a straight line. Make a circle with the second coordinate point as the center and the straight line length as the radius to construct the activity area;

[0033] Construct a three-dimensional semantic map according to the activity area. Input the activity area into the computer graphics processor, construct a two-dimensional coordinate system with the center point of the activity area as the origin, and obtain the coordinate data of the first, second, and third coordinate points. Based on the two-dimensional coordinate system and the coordinate data, input into the computer SLAM algorithm model to construct a three-dimensional map of the activity area, and fill the semantic information data (traffic signs, road reference lines) collected by the data acquisition module into the three-dimensional map to obtain the three-dimensional semantic map.

[0034] As a further improvement of the present invention, the first processing unit of the central processing unit generates corresponding control instructions, and the specific process is as follows:

[0035] The three-dimensional semantic map includes the positions of each target object, the position of the refining vehicle, and the position of the docking object. Each target object is set as an obstacle, and the path without obstacles and marked with semantic information is extracted as the moving passage path. The passage distance and passage difficulty of each moving passage path are analyzed, and the path with a short passage distance and low passage difficulty is selected as the first path.

[0036] The specific analysis for selecting the first path is as follows:

[0037] Each moving passage path is numbered and marked. For example, moving passage path one, 2, 3... The passage distance of each passage path is calculated. For example:

[0038] The first coordinate (x1, y1) where the refining vehicle is located is set as calculation point one. If the current moving passage path one is a straight-line path directly reaching the docking object, then the second coordinate (x2, y2) where the docking object is located is set as calculation point two. The coordinate points of the two-dimensional coordinate system are input into the computer operation processor and substituted into the formula to obtain the passage distance JL1 of moving passage path one.

[0039] The first coordinate (x1, y1) where the refining vehicle is located is set as calculation point one. If the current moving passage path two is a zigzag path reaching the docking object in a zigzag manner, and if there is one turning point in moving passage path two, then the coordinate (x2, y2) of the turning point is set as calculation point two, and the second coordinate (x3, y3) where the docking object is located is set as calculation point three. The coordinate points of the two-dimensional coordinate system are input into the computer operation processor and substituted into the formula to obtain the passage distance JL2 of moving passage path two.

[0040] By analogy, the passage distances of each passage path are obtained. At the same time, according to the number of calculation points, the passage difficulty of each passage path is obtained. The passage distances are sorted in ascending order, and the passage difficulties are sorted in descending order. The passage paths with the same order are selected in sequence and set as the first path.

[0041] Generate control instructions based on the first path:

[0042] Step one: The refining vehicle starts the moving mechanism and moves along the first path towards the position of the docking object.

[0043] Step two: When the refining vehicle approaches the docking object, it stops moving, reaches the docking point, and controls the refining mechanism to dock with the docking object.

[0044] As a further improvement of the present invention, the material identification module performs status feedback, and the specific process is as follows:

[0045] Normalize the physical parameters of the aluminum alloy liquid and substitute them into the formula GY = W×β1 + M×β2 + N×β3 to obtain the process evaluation index GY, where β1, β2, and β3 represent the set weights, β1 + β2 + β3 = 1. Set the process evaluation index threshold GY', and construct the threshold interval (-GY', GY', +GY') by taking the maximum to the minimum within its influence range. Match GY with (-GY', GY', +GY'). If GY ∈ (-GY', GY', +GY'), the match is successful and the status feedback is normal. If the match fails and the status feedback is abnormal.

[0046] As a further improvement of the present invention, the second processing unit of the central processing unit generates corresponding control instructions. The specific process is as follows:

[0047] Based on the feedback of the material state of the aluminum alloy liquid, if the current status feedback is abnormal, the generated corresponding control instruction is to suspend the operation of the refining mechanism of the refining vehicle and give an early warning;

[0048] If the current status feedback is normal, combined with the three-dimensional impurity distribution map, the generated corresponding control instructions are:

[0049] Step 1: Start the refining mechanism of the refining vehicle and start to transport the refining agent for impurity refining;

[0050] Step 2: The refining mechanism of the refining vehicle obtains the impurity distribution state according to the three-dimensional impurity distribution map;

[0051] Step 3: According to the impurity distribution state, the refining mechanism moves horizontally and vertically. For the area with dense impurity distribution, extend the refining agent transportation time and increase the refining agent transportation power.

[0052] Advantages of the present invention:

[0053] The present invention can generate the optimal passing route according to the surrounding environment of the refining site, control the refining vehicle to move and dock towards the aluminum alloy liquid container, which is beneficial to improving the docking efficiency, reducing the control requirements for operators, and further improving the moving flexibility of the refining vehicle;

[0054] The present invention can detect the impurity distribution state in the aluminum alloy liquid, and thus flexibly adjust the refining operation according to the impurity distribution state, which is beneficial to improving the refining effect, reducing the impurity content in the aluminum alloy liquid, and at the same time reducing the problem of uneven refining in the refining operation performed according to the established process, and further improving the refining efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] The drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention. In the drawings:

[0056] Figure 1 is the system schematic diagram of the present invention;

[0057] Figure 2 is the impurity distribution data acquisition diagram of the present invention. Specific embodiments

[0058] It should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0059] In order to enable those skilled in the art of the present technology to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0060] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so as to describe the embodiments of the present invention here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0061] In order to make the purpose and advantages of the present invention more clear and understandable, the present invention will be further described below in conjunction with the embodiments; it should be understood that the specific embodiments described here are only used to explain the present invention and are not used to limit the present invention.

[0062] Please refer to Figure 1 - Figure 2 As shown, an automatic refining vehicle intelligent control system includes an environment perception module, a material identification module, and a central processor. The central processor includes a first processing unit and a second processing unit, and also includes a data acquisition module and an execution terminal.

[0063] The environment perception module is used to dynamically capture the environmental characteristics around the refining site, perform target recognition on the environmental characteristics to obtain several target bodies and lock the docking body therefrom, comprehensively analyze the refining vehicle, the several target bodies and the docking body in combination with the environmental characteristic data to construct an activity area, and at the same time construct a three-dimensional semantic map based on the activity area and transmit it to the first processing unit of the central processor;

[0064] The data acquisition module includes a multi-modal sensor assembly I, which is composed of a high-definition camera, a lidar, and an ultrasonic radar;

[0065] The multi-modal sensor assembly I collects environmental features of objects around the refining site, including position, shape, distance, and semantic information data. The high-definition camera performs target recognition on the object shape to obtain several target bodies and locks the docking body from them. The lidar and ultrasonic radar obtain object positioning point and positioning distance data;

[0066] The specific process of the environmental perception module for constructing a three-dimensional semantic map is as follows:

[0067] Set the refining vehicle and the docking body as the first coordinate point and the second coordinate point respectively, set several target bodies as the third coordinate points, and connect the first coordinate point and the second coordinate point with a straight line. Use the straight-line length as the radius and the second coordinate point as the center to construct a circular activity area;

[0068] Construct a three-dimensional semantic map according to the activity area. Input the activity area into the computer graphics processor, construct a two-dimensional coordinate system with the center point of the activity area as the origin, and obtain the coordinate data of the first, second, and third coordinate points. Based on the two-dimensional coordinate system and the coordinate data, input into the computer SLAM algorithm model to construct a three-dimensional map of the activity area. Fill the semantic information data (traffic signs, road reference lines) collected by the data acquisition module into the three-dimensional map to obtain a three-dimensional semantic map;

[0069] The first processing unit of the central processor receives the three-dimensional semantic map transmitted by the environmental perception module, generates several moving traffic paths for the refining vehicle to reach the docking body through the three-dimensional semantic map, comprehensively analyzes the several moving traffic paths, selects the first path, and generates corresponding control instructions;

[0070] The specific process of the first processing unit of the central processor for generating corresponding control instructions is as follows:

[0071] The three-dimensional semantic map includes the positions of each target body, the position of the refining vehicle, and the position of the docking body. Set each target body as an obstacle, extract the path without obstacles and marked with semantic information as the moving traffic path, analyze the traffic distance and traffic difficulty of each moving traffic path, and select the path with a short traffic distance and low traffic difficulty as the first path;

[0072] The specific analysis for selecting the first path is as follows:

[0073] Number and mark each moving traffic path, for example, traffic path 1, 2, 3... Calculate the traffic distance of each traffic path, for example:

[0074] Set the first coordinate (x1, y1) of the refining vehicle as calculation point one. If the current passing path one is a straight-line path directly reaching the docking body, then set the second coordinate (x2, y2) of the docking body as calculation point two, and input the coordinate points of the two-dimensional coordinate system into the computer operation processor and substitute them into the formula Obtain the passing distance JL1 of the passing path one;

[0075] Set the first coordinate (x1, y1) of the refining vehicle as calculation point one. If the current passing path two is a broken-line path reaching the docking body in a zigzag manner, and if the passing path two has one turning point, then set the coordinate (x2, y2) of the turning point as calculation point two, and set the second coordinate (x3, y3) of the docking body as calculation point three. Input the coordinate points of the two-dimensional coordinate system into the computer operation processor and substitute them into the formula Obtain the passing distance JL2 of the passing path two;

[0076] By analogy, obtain the passing distances of each passing path. At the same time, obtain the passing difficulty of each passing path according to the number of calculation points. Sort the passing distances in ascending order and the passing difficulties in descending order, and select the passing paths with the same sequence number in order as the first path;

[0077] Generate a control instruction based on the first path:

[0078] Step one: The refining vehicle starts the moving mechanism and moves along the first path towards the position close to the docking body;

[0079] Step two: When the refining vehicle approaches the docking body, it stops moving, reaches the docking point, and controls the refining mechanism to dock with the docking body.

[0080] The material identification module is used for monitoring the physical parameters and analyzing the composition of the aluminum alloy liquid, including material state feedback and impurity distribution detection;

[0081] Perform state feedback on the aluminum alloy liquid, collect the physical parameters of the aluminum alloy liquid, perform normalization processing, input them into the computer operation processor for summation calculation to obtain the process evaluation index, and transmit it to the second processing unit of the central processor;

[0082] The data acquisition module includes the multi-modal sensor component two, and the multi-modal sensor component two is composed of a high-temperature infrared sensor, a density sensor, a viscosity sensor, and an acoustic wave probe;

[0083] The multi-modal sensor component two collects the physical parameters of the aluminum alloy liquid, including the temperature W collected by the infrared sensor, the density M collected by the density sensor, and the dynamic viscosity N collected by the viscosity sensor;

[0084] The material identification module performs state feedback, and the specific process is as follows:

[0085] Normalize the physical parameters of the aluminum alloy liquid and substitute them into the weighted moving average calculation formula GY = W×β1 + M×β2 + N×β3 to obtain the process evaluation index GY, where β1, β2, and β3 represent the set weights, and β1 + β2 + β3 = 1. The weighted moving average calculation formula belongs to the prior art, and its purpose is to calculate the average value by assigning different weights to different data. The magnitude of the weight can be determined by technicians using subjective and objective weighting methods. Set the process evaluation index threshold GY', and construct a threshold interval (-GY', GY', +GY') by taking the maximum value to the minimum value within its influence range. Match GY with (-GY', GY', +GY'). If GY ∈ (-GY', GY', +GY'), the match is successful and the status feedback is normal. If the match fails and the status feedback is abnormal;

[0086] Detect the impurity distribution of the aluminum alloy liquid. Input the internal space data of the docking body container into the computer graphics processor to construct a container space model. Locate the impurities through the reflection signal of ultrasonic waves in the aluminum alloy liquid. Substitute the impurity location distribution data into the container space model to construct a three-dimensional impurity distribution map and transmit it to the second processing unit of the central processor;

[0087] The multimodal sensor component two collects the impurity distribution data of the aluminum alloy liquid, including the acoustic wave probe emitting high-frequency acoustic waves to penetrate the aluminum alloy liquid, using the acoustic impedance difference between the impurities and the matrix material to reflect the echo, and obtaining the impurity location distribution data by analyzing the characteristics of the echo signal;

[0088] The specific analysis of the echo signal characteristics is as follows:

[0089] The echo signal characteristics include amplitude, time delay, and frequency spectrum;

[0090] The echo amplitude is proportional to the acoustic impedance difference between the impurity and the matrix. The higher the amplitude, the larger the impurity volume. Calculate the impurity volume T through the formula where Z i and J i represent the type parameters of the impurity amplitude and the matrix amplitude respectively;

[0091] The echo time delay calculates the impurity depth based on the acoustic wave propagation time difference. Calculate the impurity depth d through the formula where c represents the acoustic velocity parameter of the aluminum alloy liquid and t is the time delay parameter;

[0092] The echo frequency spectrum analysis converts the time-domain signal into frequency-domain analysis through fast Fourier transform;

[0093] For example:

[0094] Main frequency shift: Bubbles cause the frequency to shift to the low frequency (e.g., 10 MHz → 8.5 MHz), showing a narrow frequency band;

[0095] Harmonic attenuation: Hard inclusions (such as SiC) cause the enhancement of second harmonic, showing a wide frequency band;

[0096] Taking the aluminum alloy liquid level in the docking body container as the matrix, through analysis, the impurity volume, depth position and type can be obtained, and the impurity localization distribution data of the aluminum alloy liquid in the docking body container can be obtained;

[0097] The second processing unit of the central processing unit receives the material state feedback transmitted by the material identification module and the impurity distribution detection for comprehensive analysis, and generates corresponding control instructions;

[0098] The second processing unit of the central processing unit generates corresponding control instructions, and the specific process is as follows:

[0099] Based on the feedback of the aluminum alloy liquid material state, if the current state feedback is abnormal, the corresponding control instruction generated is to suspend the operation of the refining mechanism of the refining vehicle and give an early warning;

[0100] If the current state feedback is normal, combined with the three-dimensional impurity distribution map, the corresponding control instructions generated are:

[0101] Step 1: Start the refining mechanism of the refining vehicle and start to transport the refining agent for impurity refining;

[0102] Step 2: The refining mechanism of the refining vehicle obtains the impurity distribution state according to the three-dimensional impurity distribution map;

[0103] Step 3: According to the impurity distribution state, the refining mechanism moves horizontally and vertically. For the area where the impurity distribution is dense, extend the refining agent delivery time and increase the refining agent delivery power.

[0104] The execution terminal is responsible for receiving the control instructions transmitted by the central processing unit and performing corresponding control operations;

[0105] Based on the control instructions generated by the first processing unit of the central processing unit, control the starting of the moving mechanism of the refining vehicle to move to the specified docking point of the docking body by itself according to the first path;

[0106] Based on the control instructions generated by the second processing unit of the central processing unit, control the starting of the refining mechanism of the refining vehicle to perform corresponding refining operations.

[0107] The above are only the embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. An automatic intelligent control system for vehicle refining, including an environment perception module, a material identification module and a central processor. The central processor includes a first processing unit and a second processing unit, and is characterized in that: The environment perception module is used to dynamically capture the environmental characteristics around the refining site, perform target recognition on the environmental characteristics to obtain several target bodies and lock the docking body from them, comprehensively analyze the environmental characteristic data of the refining vehicle, several target bodies and the docking body to construct an activity area, and at the same time construct a three-dimensional semantic map based on the activity area and transmit it to the first processing unit of the central processor; The material identification module is used for monitoring the physical parameters and component analysis of aluminum alloy liquid, including material state feedback and impurity distribution detection; Perform state feedback on the aluminum alloy liquid, collect the physical parameters of the aluminum alloy liquid, input them into a computer arithmetic processor for calculation after normalization to obtain a process evaluation index, and transmit it to the second processing unit of the central processor; Perform impurity distribution detection on the aluminum alloy liquid, input the internal space data of the docking body container into a computer graphics processor to construct a container space model, locate impurities through the reflection signal of ultrasonic waves in the aluminum alloy liquid, substitute the impurity location distribution data into the container space model to construct a three-dimensional impurity distribution map, and transmit it to the second processing unit of the central processor; The first processing unit of the central processor receives the three-dimensional semantic map transmitted by the environment perception module, generates several moving paths for the refining vehicle to reach the docking body through the three-dimensional semantic map, comprehensively analyzes the several moving paths to select the first path, and generates corresponding control instructions; The second processing unit of the central processor receives the material state feedback and impurity distribution detection transmitted by the material identification module for comprehensive analysis, and generates corresponding control instructions.

2. The automatic refining vehicle intelligent control system according to claim 1, characterized in that, It also includes a data acquisition module and an execution terminal; The data acquisition module includes a multi-modal sensor component one and a multi-modal sensor component two; Among them, the multi-modal sensor component one is composed of a high-definition camera, a lidar and an ultrasonic radar; the multi-modal sensor component one collects the environmental characteristics of the objects around the refining site, including position, shape, distance and semantic information data, the high-definition camera performs target recognition on the object shape to obtain several target bodies and locks the docking body from them, and the lidar and ultrasonic radar obtain the object positioning point and positioning distance data; The multi-modal sensor component two is composed of a high-temperature infrared sensor, a density sensor, a viscosity sensor and a sound wave probe; the multi-modal sensor component two collects the physical parameters of the aluminum alloy liquid, including the temperature W collected by the infrared sensor, the density M collected by the density sensor and the dynamic viscosity N collected by the viscosity sensor; the multi-modal sensor component two also collects the impurity distribution data of the aluminum alloy liquid, including the sound wave probe emitting high-frequency sound waves to penetrate the aluminum alloy liquid, using the acoustic impedance difference between the impurities and the matrix material to reflect the echo, and obtaining the impurity location distribution data by analyzing the characteristics of the echo signal; The execution terminal is responsible for receiving the control instructions transmitted by the central processor and performing corresponding control operations; Based on the control instructions generated by the first processing unit of the central processor, control the movement mechanism of the refining vehicle to start and move automatically to the specified docking point of the docking body according to the first path; Based on the control instructions generated by the second processing unit of the central processing unit, control the refining mechanism of the refining vehicle to start and execute the corresponding refining operations.

3. An automatic refining vehicle intelligent control system according to claim 1, characterized in that, The specific process for the environmental perception module to construct a three-dimensional semantic map is as follows: Set the refining vehicle and the docking body as the first coordinate point and the second coordinate point respectively, set several target bodies as the third coordinate points, connect the first coordinate point and the second coordinate point with a straight line, and construct an activity area by making a circle with the second coordinate point as the center and the straight-line length as the radius; Construct a three-dimensional semantic map according to the activity area. Input the activity area into the computer graphics processor, construct a two-dimensional coordinate system with the center point of the activity area as the origin, and obtain the coordinate data of the first, second, and third coordinate points. Based on the two-dimensional coordinate system and the coordinate data, input them into the computer SLAM algorithm model to construct a three-dimensional map of the activity area, and fill the semantic information data collected by the data acquisition module into the three-dimensional map correspondingly to obtain the three-dimensional semantic map.

4. An automatic refining vehicle intelligent control system according to claim 1, characterized in that, The first processing unit of the central processing unit generates corresponding control instructions. The specific process is as follows: The three-dimensional semantic map includes the positions of each target body, the position of the refining vehicle, and the position of the docking body. Set each target body as an obstacle, and extract the path without obstacles and marked with semantic information as the moving passage path. Analyze the passage distance and passage difficulty of each moving passage path, and select the path with a short passage distance and low passage difficulty as the first path; Generate control instructions based on the first path: Step 1: The refining vehicle starts the moving mechanism and moves along the first path towards the position of the docking body; Step 2: When the refining vehicle approaches the docking body, it stops moving, reaches the docking point, and controls the refining mechanism to dock with the docking body.

5. An automatic refining vehicle intelligent control system according to claim 1, characterized in that, The material identification module conducts status feedback. The specific process is as follows: Normalize the physical parameters of the aluminum alloy liquid and substitute them into the formula GY = W×β1 + M×β2 + N×β3 to obtain the process evaluation index GY, where β1, β2, and β3 represent the set weights, β1 + β2 + β3 = 1. Set the process evaluation index threshold GY', and construct the threshold interval (-GY', GY', +GY') by taking the maximum to the minimum within its influence range. If GY ∈ (-GY', GY', +GY'), the matching is successful and the status feedback is normal. If the matching fails and the status feedback is abnormal.

6. The automatic refining vehicle intelligent control system according to claim 5, characterized in that, The second processing unit of the central processing unit generates corresponding control instructions. The specific process is as follows: Based on the status of the aluminum alloy liquid material for feedback, if the current status feedback is abnormal, the corresponding control instruction generated is to suspend the operation of the refining mechanism of the refining vehicle and give an early warning; If the current status feedback is normal, combined with the three-dimensional impurity distribution map, the corresponding control instructions generated are: Step 1: The refining mechanism of the refining vehicle starts and begins to transport the refining agent for impurity refining; Step 2: The refining mechanism of the refining vehicle obtains the impurity distribution status according to the three-dimensional impurity distribution map; Step 3: According to the impurity distribution status, the refining mechanism moves horizontally and vertically. For the area with dense impurity distribution, extend the refining agent transportation time and increase the refining agent transportation power.

Images