An automatic operation system and method for a mine crusher based on artificial intelligence

Abstract

The present invention relates to the field of artificial intelligence, and particularly to an automatic operation system and method for a mine crusher based on artificial intelligence. It includes: converting the collected image data into a three-dimensional environmental model, and performing equally spaced grid segmentation on the three-dimensional environmental model by means of grid segmentation; introducing a multi-objective optimization control algorithm to calculate the comprehensive objective function value of each candidate striking point to obtain the optimal striking point; using an adaptive multi-objective path planning algorithm to calculate the cost function of the path to obtain the optimal path; using an adaptive striking force control algorithm to calculate the striking force; and using a control signal generation algorithm to generate the control signal of the crusher. It aims to solve the technical problem that the existing control signal generation method of the crusher is relatively simple, unable to handle the non-linear changes in complex working environments, and prone to cause unstable phenomena when the crushing hammer is operating at high speed.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to an automatic operation system and method for a mine crusher based on artificial intelligence. Background Art

[0002] In modern industry and infrastructure construction, crushers are widely used for the crushing and processing of various stones. With the progress of technology, many deficiencies have gradually emerged in the traditional crusher operation mode. The operation process of traditional crushers relies on manual control. Operators need to monitor and adjust the working state of crushers based on experience, which not only consumes a large amount of manpower and time but also is prone to misoperation, resulting in equipment damage or a decline in production efficiency. In addition, due to the large differences in the characteristics of stones such as hardness and particle size, the control parameters in the crushing process need to be frequently adjusted, and manual adjustment is often difficult to accurately control, thus affecting the stability of product quality.

[0003] In addition, the following technical problems also exist in the existing automatic operation methods of crushers: The selection of the striking point is based on a preset fixed path or simple surface analysis, making it difficult to accurately adapt to the complex conditions of the ore surface, resulting in inaccurate strikes and even damage to the crushing hammer; Due to the unstable striking angle of the crushing hammer, it is easy to cause uneven striking force or poor crushing effect; During the operation of the crusher, the path planning of the crushing hammer from the initial position to the striking point is inaccurate, easily leading to energy waste and a decline in control accuracy; The striking force adjustment mechanism is relatively lagging and cannot be dynamically adjusted according to the changes in the real-time operation environment, resulting in difficult accurate control of the striking force and affecting the crushing effect; The control signal generation method of the crusher is relatively simple and cannot handle the non-linear changes in a complex operation environment, easily causing unstable phenomena when the crushing hammer operates at high speed.

[0004] To solve the above problems, artificial intelligence technology has gradually been introduced into the automatic control of crushers. The development of artificial intelligence technology provides new ideas and methods for the intelligent and automatic control of crushers. By embedding artificial intelligence algorithms into the control system of crushers, real-time monitoring and dynamic adjustment of the crushing process can be achieved, thereby improving production efficiency, reducing equipment failure rates, and reducing manual intervention and potential safety risks. Summary of the Invention

[0005] The present invention provides an automatic operation system and method for a mine crusher based on artificial intelligence, aiming to solve the following technical problems: the selection of the impact point in the existing technology is based on a preset fixed path or simple surface analysis, which is difficult to accurately adapt to the complex conditions of the ore surface, resulting in inaccurate impacts and even damage to the crushing hammer; due to the unstable impact angle of the crushing hammer, it is easy to cause uneven impact force or poor crushing effect; during the operation of the crusher, the path planning of the crushing hammer from the initial position to the impact point is inaccurate, easily leading to energy waste and a decrease in control accuracy; the impact force adjustment mechanism is relatively lagging and cannot be dynamically adjusted according to the changes in the real-time operation environment, resulting in difficult accurate control of the impact force and affecting the crushing effect; the control signal generation method of the crusher is relatively simple and cannot handle the non-linear changes in the complex operation environment, easily causing unstable phenomena when the crushing hammer operates at high speed.

[0006] An automatic operation system and method for a mine crusher based on artificial intelligence according to the present invention specifically include the following technical solutions:

[0007] An automatic operation method for a mine crusher based on artificial intelligence includes the following steps:

[0008] S1: Obtain the hardness of the ore surface based on the collected image data, and obtain candidate impact points through the grid segmentation method; introduce a multi-objective optimization control algorithm, calculate the comprehensive objective function value of each candidate impact point, obtain the optimal impact point, and thus obtain the hardness of the ore surface at the optimal impact point.

[0009] S2: Based on the optimal impact point, use the adaptive multi-objective path planning algorithm to calculate the optimal path, and calculate the speed and acceleration of the crushing hammer at different times according to the optimal path.

[0010] S3: Based on the speed of the crushing hammer at different times, the acceleration of the crushing hammer, and the hardness of the ore surface at the optimal impact point, use the adaptive impact force control algorithm to calculate the impact force.

[0011] S4: Based on the impact force, the speed of the crushing hammer, the acceleration of the crushing hammer, and the hardness of the ore surface at the optimal impact point, use the control signal generation algorithm to generate the control signal of the crusher.

[0012] Preferably, the S1 specifically includes:

[0013] Convert the image data into a three-dimensional environment model to obtain the hardness of the ore surface; perform equally spaced grid segmentation on the three-dimensional environment model through the grid segmentation method, and the grid center point is the candidate impact point.

[0014] Preferably, the S1 specifically includes:

[0015] Based on the candidate striking points and the hardness of the ore surface, the comprehensive objective function value of each candidate striking point is obtained through a multi-objective optimization control algorithm, and the striking point with the largest comprehensive objective function value is the optimal striking point.

[0016] Preferably, the S2 specifically includes:

[0017] In the implementation process of the adaptive multi-objective path planning algorithm, the cost function of the path is calculated, and the specific formula is:

[0018]

[0019] where J(L(t)) represents the cost function of the path at time t; represents the integral of time from the starting time t 0 to the termination time t f ; a(t) represents the acceleration of the stone crusher at time t; represents the change rate of the motor power fluctuation at time t; represents the position of the stone crusher at time t; p o represents the optimal striking point.

[0020] Preferably, the S2 specifically includes:

[0021] The cost function of the path is minimized through iterative optimization to obtain the optimal path.

[0022] Preferably, the S3 specifically includes:

[0023] The adaptive striking force control algorithm controls the striking force through the speed and acceleration of the stone crusher, and dynamically adjusts the striking force in combination with the hardness information of the ore striking point at the optimal striking point.

[0024] Preferably, the S4 specifically includes:

[0025] The calculation formula for generating the control signal of the stone crusher is:

[0026]

[0027] where C(t) represents the control signal of the stone crusher at time t; γ represents the adjustment coefficient for adjusting the weight of the striking force in the generation of the control signal; F(t) represents the striking force of the stone crusher at time t; v(t) represents the speed of the stone crusher at time t; δ represents the adjustment coefficient for adjusting the influence of the acceleration and the hardness of the ore surface at the optimal striking point on the generation of the control signal; H(p o ) represents the hardness of the ore surface at the optimal striking point.

[0028] An automatic operation system for a mine stone crusher based on artificial intelligence includes the following parts:

[0029] Image data acquisition module, three-dimensional environment model generation module, candidate strike point selection module, multi-objective optimization control algorithm module, path planning and optimization module, adaptive strike force control module, control signal generation module;

[0030] Image data acquisition module: By installing a multi-camera machine vision camera in the working area of the crusher, image data from different angles is collected and output to the three-dimensional environment model generation module;

[0031] Three-dimensional environment model generation module: Based on the image data from the image data acquisition module, using a convolutional neural network and a three-dimensional reconstruction algorithm, the image data is converted into a three-dimensional environment model, and the hardness of the ore surface is obtained. The three-dimensional environment model is output to the candidate strike point selection module, and the hardness of the ore surface is output to the multi-objective optimization control algorithm module;

[0032] Candidate strike point selection module: Based on the three-dimensional environment model from the three-dimensional environment model generation module, the three-dimensional environment model is equally spaced grid-divided by the grid segmentation method, and the grid center point is the candidate strike point. The candidate strike point is output to the multi-objective optimization control algorithm module;

[0033] Multi-objective optimization control algorithm module: Based on the candidate strike points from the candidate strike point selection module and the hardness of the ore surface from the three-dimensional environment model generation module, the multi-objective optimization control algorithm is used to obtain the comprehensive objective function value of each candidate strike point. The strike point with the largest comprehensive objective function value is the optimal strike point; the hardness of the ore surface at the optimal strike point is obtained through the convolutional neural network and three-dimensional reconstruction algorithm of the three-dimensional environment model generation module; the optimal strike point is output to the path planning and optimization module, and the hardness of the ore surface at the optimal strike point is output to the adaptive strike force control module and the control signal generation module;

[0034] Path planning and optimization module: Based on the optimal strike point from the multi-objective optimization control algorithm module, the cost function of the path is calculated using the adaptive multi-objective path planning algorithm, and the gradient descent method is used for iterative optimization to minimize the cost function and obtain the optimal path. The speed and acceleration of the crusher hammer at different times are calculated according to the optimal path and output to the adaptive strike force control module and the control signal generation module;

[0035] Adaptive strike force control module: Based on the speed and acceleration of the crusher hammer at different times and the hardness of the ore surface at the optimal strike point, the adaptive strike force control algorithm is used to calculate the strike force in real time and output the strike force to the control signal generation module;

[0036] Control signal generation module: Based on the striking force of the adaptive striking force control module, the gravel hammer speeds, accelerations at different times of the path planning and optimization module, and the hardness of the ore surface at the optimal striking point of the multi-objective optimization control algorithm module, a control signal generation algorithm is used to generate the control signal for the crusher.

[0037] The beneficial effects of the technical solution of the present invention are:

[0038] 1. Through the multi-objective optimization control algorithm, a comprehensive evaluation of multiple indicators such as energy transfer efficiency, striking stability, and ore hardness adaptability is carried out on the candidate striking points, and the comprehensive objective function values of each candidate striking point are obtained, ensuring the effective transfer of energy, stable control of the striking angle, and adaptive processing of ores with different hardnesses during the operation of the crusher, thereby achieving the effects of reducing energy waste, extending the service life of the equipment, and ensuring the operation stability in complex ore environments. By maximizing the comprehensive objective function value, the optimal striking point is selected, realizing the optimal control of the crusher operation.

[0039] 2. The adaptive multi-objective path planning algorithm is adopted to optimize the moving path of the gravel hammer, and the cost function of the path is minimized by the gradient descent method, significantly reducing the acceleration fluctuation and motor power fluctuation, ensuring that the gravel hammer can stably move to the optimal striking point in a complex environment, avoiding unnecessary vibrations or errors, and ensuring the accuracy of the crusher operation.

[0040] 3. Through the adaptive striking force control algorithm, according to the speed of the gravel hammer, the acceleration of the gravel hammer, and the hardness information of the ore surface at the optimal striking point at different times, the striking force is precisely adjusted, further improving the striking accuracy and operation effect of the crusher.

[0041] 4. The control signal generation algorithm is used to generate the control signal for controlling the operation process of the crusher, ensuring the stability of the striking process of the crusher under different operating conditions. Especially when the gravel hammer is moving at high speed, through the square root and logarithmic processing of the speed and acceleration of the gravel hammer, the fluctuation of the control signal is further reduced, enabling the crusher to still operate stably in a complex operating environment, reducing the failure rate and maintenance cost of the equipment. Brief description of the drawings

[0042] Figure 1 It is a structure diagram of an automatic operation system of a mine crusher based on artificial intelligence according to the present invention;

[0043] Figure 2 It is a flowchart of an automatic operation method of a mine crusher based on artificial intelligence according to the present invention. Detailed implementation manners

[0044] To further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with 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 the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0046] The following specifically describes the specific solutions of an automatic operation system and method for a mine crusher based on artificial intelligence provided by the present invention in conjunction with the accompanying drawings.

[0047] Referring to the attached Figure 1 , which shows the structure diagram of an automatic operation system for a mine crusher based on artificial intelligence provided by an embodiment of the present invention. The system includes the following parts:

[0048] An image data acquisition module, a three-dimensional environment model generation module, a candidate strike point selection module, a multi-objective optimization control algorithm module, a path planning and optimization module, an adaptive strike force control module, and a control signal generation module;

[0049] The image data acquisition module: By installing a multi-camera machine vision camera covering the operation area of the crusher, image data at different angles is collected, and the image data is output to the three-dimensional environment model generation module;

[0050] The three-dimensional environment model generation module: Based on the image data from the image data acquisition module, using a convolutional neural network and a three-dimensional reconstruction algorithm, the image data is converted into a three-dimensional environment model, and the hardness of the ore surface is obtained. The three-dimensional environment model is output to the candidate strike point selection module, and the hardness of the ore surface is output to the multi-objective optimization control algorithm module;

[0051] The candidate strike point selection module: Based on the three-dimensional environment model of the three-dimensional environment model generation module, the three-dimensional environment model is equally spaced grid-divided by the grid segmentation method, and the grid center point is the candidate strike point. The candidate strike point is output to the multi-objective optimization control algorithm module;

[0052] Multi-objective optimization control algorithm module: Based on the candidate strike points of the candidate strike point selection module and the hardness of the ore surface of the three-dimensional environment model generation module, use the multi-objective optimization control algorithm to obtain the comprehensive objective function value of each candidate strike point. The strike point with the largest comprehensive objective function value is the optimal strike point; obtain the hardness of the ore surface at the optimal strike point through the convolutional neural network and three-dimensional reconstruction algorithm of the three-dimensional environment model generation module; output the optimal strike point to the path planning and optimization module, and output the hardness of the ore surface at the optimal strike point to the adaptive strike force control module and the control signal generation module;

[0053] Path planning and optimization module: Based on the optimal strike point of the multi-objective optimization control algorithm module, use the adaptive multi-objective path planning algorithm to calculate the cost function of the path, and use the gradient descent method for iterative optimization to minimize the cost function to obtain the optimal path. Calculate the speed and acceleration of the stone crusher at different times according to the optimal path, and output the speed and acceleration of the stone crusher at different times to the adaptive strike force control module and the control signal generation module;

[0054] Adaptive strike force control module: Based on the speed and acceleration of the stone crusher at different times and the hardness of the ore surface at the optimal strike point, use the adaptive strike force control algorithm to calculate the strike force in real time, and output the strike force to the control signal generation module;

[0055] Control signal generation module: Based on the strike force of the adaptive strike force control module, the speed and acceleration of the stone crusher at different times of the path planning and optimization module, and the hardness of the ore surface at the optimal strike point of the multi-objective optimization control algorithm module, use the control signal generation algorithm to generate the control signal of the stone crusher.

[0056] Refer to the appendix Figure 2 which shows the flowchart of an automatic operation method of a mine stone crusher based on artificial intelligence provided by an embodiment of the present invention. The method includes the following steps:

[0057] S1. Obtain the hardness of the ore surface based on the collected image data, and obtain candidate strike points through the grid segmentation method; introduce the multi-objective optimization control algorithm, calculate the comprehensive objective function value of each candidate strike point, and obtain the optimal strike point, so as to obtain the hardness of the ore surface at the optimal strike point;

[0058] By installing a multi-view machine vision camera covering the operation area of the stone crusher, collect image data I = {I 1 , I 2 ,..., I n}, I jis the j-th image data, and n is the number of image data; using a convolutional neural network and a three-dimensional reconstruction algorithm, the image data is converted into a three-dimensional environmental model M(X, Y, Z), and the hardness of the ore surface is obtained; the convolutional neural network and the three-dimensional reconstruction algorithm are well-known technical means in the art and will not be elaborated here;

[0059] Perform equally-spaced grid division on the three-dimensional environmental model M(X, Y, Z) by the grid division method, and the grid center point is the candidate strike point P = {p 1 , p 2 ,..., p N}, where p i (x i , y i , z i ) represents the i-th candidate strike point, and (x i , y i , z i ) represents the position coordinates of the i-th candidate strike point, and N is the number of candidate strike points;

[0060] To achieve the optimal operation control effect of the crusher, based on the candidate strike points and the hardness of the ore surface, use a multi-objective optimization control algorithm to obtain the comprehensive objective function value of each candidate strike point, ensure the energy transfer efficiency and strike stability of the strike point during the operation of the crusher, and adapt to the hardness of the ore surface. The strike point with the largest comprehensive objective function value is the optimal strike point;

[0061] The multi-objective optimization control algorithm comprehensively evaluates according to three aspects: the energy transfer efficiency, strike stability, and ore hardness adaptability of the candidate strike points, obtains the comprehensive objective function value of each candidate strike point, and the strike point with the largest comprehensive objective function value is the optimal strike point;

[0062] The core of the energy transfer efficiency is that when the crusher strikes the ore, the energy is transferred from the crushing hammer of the crusher to the candidate strike point. The distance between the candidate strike point and the crushing hammer directly affects the effective transfer of energy. Using the first adjustment parameter to non-linearly weight the distance from the candidate strike point to the crushing hammer can flexibly control the influence of the distance from the candidate strike point to the crushing hammer on the energy transfer efficiency, and combined with the energy loss factor, it reflects the energy loss caused by ore material factors during the energy transfer process;

[0063] The strike stability considers two aspects: the strike angle and the hardness of the ore surface. By adjusting the strike angle through the hyperbolic tangent function, the output range of the strike angle can be limited within Between them, so as to ensure that the change in the striking angle is acceptable within a certain range; the hardness of the surface of the candidate striking point directly affects the striking effect of the crusher, and the influence of the hardness of the ore surface at the candidate striking point on the striking stability is adjusted through the second adjustment parameter;

[0064] The adaptability of the ore hardness is processed in the form of a piecewise function to adapt to the dynamic change of the ore hardness. The purpose is to adapt to the change of the ore hardness. Different hardness ores need different treatment methods when choosing the striking point. Low-hardness ores are easy to break and are adjusted in an exponential decay manner. High-hardness ores are not easy to break and are adjusted with a logarithmic function to slow down the influence of the hardness increase and ensure that the high-hardness area will not be completely excluded from the selection range of the striking point due to excessive hardness;

[0065] The calculation formula of the comprehensive objective function value is:

[0066]

[0067] Among them, O(p i ) represents the comprehensive objective function value, that is, the priority of the i-th candidate striking point, which is used to select the optimal striking point; d(p i ) represents the distance between the i-th candidate striking point and the crusher hammer; k 1 represents the first adjustment parameter, which is used to adjust the influence of the distance between the i-th candidate striking point and the crusher hammer on the energy transfer efficiency; R(p i ) represents the energy loss factor, that is, the loss due to energy transfer at the i-th candidate striking point, which can be specifically set according to the specific implementation scenario and is not limited here; θ(p i ) represents the striking angle of the i-th candidate striking point, that is, the included angle between the crusher hammer and the i-th candidate striking point; tanh represents the hyperbolic tangent function, which is used to smooth the non-linear influence of the striking angle of the i-th candidate striking point; H(p i ) represents the hardness of the ore surface at the i-th candidate striking point; k 2 represents the second adjustment parameter, which is used to adjust the influence of the hardness of the ore surface at the i-th candidate striking point on the striking stability; represents the ore hardness adaptability function, which is used to dynamically adjust the priority of the candidate striking point according to the hardness of the ore surface, reflecting the adaptability of the i-th candidate striking point to the hardness of the ore surface, and avoiding unnecessary energy consumption or damage to the crusher due to excessive hardness of the surface of the candidate striking point. The specific calculation formula of the ore hardness adaptability function is:

[0068]

[0069] Among them, H 0 represents the hardness threshold, which is used to distinguish the high and low of the ore hardness; when H(pi ) ≤ H 0 When it is the case, the ore hardness adaptability is calculated by an exponential decay function, indicating that an increase in ore hardness will rapidly reduce the ore hardness adaptability; β 1 represents an adjustment parameter used to control the influence degree of the exponential decay function on ore hardness, adjusts the curve steepness of the ore hardness adaptability, and controls the decay rate of the exponential decay function; when H(p i ) > H 0 When it is the case, the ore hardness adaptability is calculated by a logarithmic function, gently adjusts the ore hardness adaptability, reduces the influence on the priority of adjusting candidate striking points, and prevents the candidate striking points from being wrongly selected first due to too high ore hardness; ∈ represents a minimum value to prevent the denominator from being zero;

[0070]

[0071] Among them, p o represents the optimal striking point, which is the maximum value point selected after optimizing the comprehensive objective function among all candidate striking points; arg max represents a mathematical operator, indicating to select the candidate striking point p i ;

[0072] Meanwhile, based on the optimal striking point, the hardness of the ore surface at the optimal striking point is obtained through a convolutional neural network and a three-dimensional reconstruction algorithm;

[0073] S2. Based on the optimal striking point, use an adaptive multi-objective path planning algorithm to calculate the optimal path, and calculate the speed and acceleration of the gravel hammer at different times according to the optimal path;

[0074] To ensure that the gravel hammer can accurately move to the optimal striking point, based on the optimal striking point, use an adaptive multi-objective path planning algorithm to calculate the cost function of the path;

[0075] The goal of the adaptive multi-objective path planning algorithm is to plan the best path of the gravel hammer from the initial position to the optimal striking point by optimizing acceleration, motor power fluctuation, and position error, and further ensure the energy efficiency, operation stability, and operation accuracy in a complex environment through multi-objective fusion and non-linear processing, such as square root and logarithmic function processing;

[0076] The calculation formula of the cost function of the path is:

[0077]

[0078] Among them, J(L(t)) represents the cost function of the path at time t; represents from the starting time t 0 to the termination time t fIntegrate over time to accumulate the costs at different time points throughout the process; a(t) represents the acceleration of the breaker hammer at time t; |a(t)| 1 / 2 represents controlling the energy consumption by the square root of the acceleration to keep the increase in energy consumption gentle, thereby reducing unnecessary energy consumption; represents the rate of change of the motor power fluctuation at time t, reflecting the motor power fluctuation of the motor system during the operation of the crusher; represents using a logarithmic function to smooth the motor power fluctuation to ensure the smooth operation of the motor system and reduce the control instability caused by the motor power fluctuation; represents the position of the breaker hammer at time t; p o represents the optimal impact point; represents the deviation between the position of the breaker hammer and the optimal impact point at time t, that is, the position error; represents the position error suppression term, ensuring that the breaker hammer accurately reaches the optimal impact point and can be adjusted in time when a position deviation occurs to avoid unnecessary vibrations or instabilities caused by overcorrection;

[0079] Through iterative optimization using the gradient descent method to minimize the cost function of the path, the optimal path is obtained, that is, the best movement trajectory of the breaker hammer from the initial position to the optimal impact point; calculate the speed and acceleration of the breaker hammer at different times according to the optimal path; the gradient descent method is a well-known technical means for those skilled in the art and will not be elaborated here;

[0080] During the path planning process, adjust the path according to the real-time updated position and speed of the breaker hammer, and the optimal path selection can be maintained in a changing operating environment;

[0081] S3. Based on the speed of the breaker hammer at different times, the acceleration of the breaker hammer, and the hardness of the ore surface at the optimal impact point, use an adaptive impact force control algorithm to calculate the impact force;

[0082] Based on the speed of the breaker hammer at different times, the acceleration of the breaker hammer, and the hardness of the ore surface at the optimal impact point, use an adaptive impact force control algorithm to calculate the impact force in real time; the adaptive impact force control algorithm precisely controls the impact force through the speed and acceleration of the breaker hammer, avoiding energy waste and impact errors caused by the lag adjustment of the impact force, and dynamically adjusts the impact force in combination with the hardness information of the ore impact point at the optimal impact point. In order to avoid unreasonable impact forces in the case of extremely low or even zero speed, a minimum value prevention mechanism is added to ensure the stability of the impact effect;

[0083] The calculation formula for the impact force is:

[0084]

[0085] Among them, F(t) represents the impact force at time t; m represents the mass of the gravel hammer; a(t) represents the acceleration of the gravel hammer at time t; H(p o ) represents the hardness of the ore surface at the optimal impact point, which determines the transmission effect of the impact force and the degree of fragmentation of the gravel; B represents the impact area of the gravel hammer, which is determined by the physical size of the gravel hammer; v(t) represents the speed of the gravel hammer at time t; ∈ represents a minimum value to prevent the denominator from being zero;

[0086] S4. Based on the impact force, the speed of the gravel hammer, the acceleration of the gravel hammer, and the hardness of the ore surface at the optimal impact point, use a control signal generation algorithm to generate a control signal for the crusher;

[0087] Based on the impact force, the speed of the gravel hammer, the acceleration of the gravel hammer, and the hardness of the ore surface at the optimal impact point, use a control signal generation algorithm to generate a control signal for the crusher; the control signal generation algorithm combines the impact force, the speed of the gravel hammer, the acceleration of the gravel hammer, and the hardness of the ore surface at the optimal impact point to generate a control signal for controlling the impact process of the crusher. It combines non-linear function processing and an adaptive adjustment mechanism to ensure the stability of the impact process under different operating conditions of the crusher;

[0088] The square root and logarithmic processing of the speed of the gravel hammer effectively reduce the signal fluctuation of the gravel hammer during high-speed movement; the square term of the hardness of the ore surface is used to reduce the excessive interference of extremely high-hardness ores on the control signal and ensure the stable output of the control signal;

[0089] The calculation formula for generating the control signal of the crusher is:

[0090]

[0091] Among them, C(t) represents the control signal of the crusher at time t, which is used to control the impact process of the gravel hammer; γ represents an adjustment coefficient used to adjust the weight of the impact force in the generation of the control signal; F(t) represents the impact force of the gravel hammer at time t; ln(1 + |v(t)| 1 / 2 ) represents the smoothing processing function of the speed of the gravel hammer at time t, which adopts a form combining a logarithmic function and a square root. The logarithmic function can smooth the high-speed movement state of the gravel hammer, and the square root further slows down the change rate; v(t) represents the speed of the gravel hammer at time t; δ represents an adjustment coefficient used to adjust the influence of the acceleration and the hardness of the ore surface at the optimal impact point on the generation of the control signal; a(t) represents the acceleration of the gravel hammer at time t; represents the square root processing of the acceleration to ensure the stability of the control signal; H(p o ) represents the hardness of the ore surface at the optimal impact point; H(po ) 2 represents the square term of the hardness of the ore surface, which is used to reduce the excessive interference of extremely high-hardness ores on the control signal and ensure the stable output of the control signal;

[0092] According to the impact force, the speed of the gravel hammer, the acceleration of the gravel hammer, and the hardness of the ore surface at the optimal impact point, a control signal generation algorithm is used to generate the control signal of the crusher to precisely control the impact action of the crusher.

[0093] In summary, an automatic operation system and method for a mine crusher based on artificial intelligence are completed.

[0094] The sequence of the invention embodiments is only for description and does not represent the advantages or disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0095] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. The key point of each embodiment is to illustrate the differences from other embodiments.

[0096] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments or perform equivalent replacements for some of the technical features. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of each embodiment of the present invention and should all be included within the protection scope of the present invention.

Claims

1. An automatic operation method of a mine crusher based on artificial intelligence, characterized in that: The following steps are involved: S1: Obtain the hardness of the ore surface based on the collected image data, and obtain the candidate striking points through the grid segmentation method; introduce the multi-objective optimization control algorithm, calculate the comprehensive objective function value of each candidate striking point, and obtain the optimal striking point, thereby obtaining the hardness of the ore surface at the optimal striking point; S2: Based on the optimal strike point, an adaptive multi-objective path planning algorithm is introduced. In the implementation process of the adaptive multi-objective path planning algorithm, the cost function of the path is calculated. The specific formula is: Where J(L(t)) represents the cost function of the path at time t; Represents from the start time t0 to the end time t f Integrate over time; a(t) represents the acceleration of the hammer at time t; It represents the rate of change of motor power fluctuation at time t; represents the position of the stone breaker at time t; p O Indicates the optimal striking point; The cost function of the path is minimized through iterative optimization to obtain the optimal path, and the speed and acceleration of the stone breaker at different times are calculated based on the optimal path; S3: Calculate the striking force using an adaptive striking force control algorithm based on the speed of the rock breaker at different times, the acceleration of the rock breaker and the hardness of the ore surface at the optimal striking point; S4: Generate a control signal for the rock crusher using a control signal generation algorithm based on the impact force, the rock crusher speed, the rock crusher acceleration and the hardness of the ore surface at the optimal impact point.

2. The method for automatic operation of a mine crusher based on artificial intelligence according to claim 1, characterized in that: The S1 specifically includes: The image data is converted into a three-dimensional environmental model to obtain the hardness of the ore surface; the three-dimensional environmental model is segmented into equally spaced grids using the grid segmentation method, and the center point of the grid is the candidate impact point.

3. The method for automatic operation of a mine crusher based on artificial intelligence according to claim 2 is characterized in that: The S1 specifically includes: Based on the candidate striking points and the hardness of the ore surface, the comprehensive objective function value of each candidate striking point is obtained through a multi-objective optimization control algorithm, and the striking point with the largest comprehensive objective function value is the optimal striking point.

4. The method for automatic operation of a mine crusher based on artificial intelligence according to claim 1, characterized in that: The S3 specifically includes: The adaptive striking force control algorithm controls the striking force through the speed and acceleration of the stone breaker, and dynamically adjusts the striking force in combination with the hardness information of the ore striking point at the optimal striking point.

5. The method for automatic operation of a mine crusher based on artificial intelligence according to claim 1, characterized in that: The S4 specifically includes: The calculation formula for generating the control signal of the lithotripsy is: Wherein, C(t) represents the control signal of the crusher at time t; γ represents the adjustment coefficient used to adjust the weight of the impact force in the control signal generation; F(t) represents the impact force of the crushing hammer at time t; v(t) represents the speed of the crushing hammer at time t; δ represents the adjustment coefficient used to adjust the influence of acceleration and the hardness of the ore surface at the optimal impact point on the control signal generation; H(p O ) represents the hardness of the ore surface at the optimal striking point.

6. An artificial intelligence-based automatic operation system for a mine crusher, applied to the artificial intelligence-based automatic operation method for a mine crusher according to claim 1, characterized in that: Includes the following parts: Image data acquisition module, three-dimensional environment model generation module, candidate strike point selection module, multi-objective optimization control algorithm module, path planning and optimization module, adaptive strike force control module, control signal generation module; Image data acquisition module: collects image data and outputs the image data to the three-dimensional environment model generation module; 3D environment model generation module: converts image data into a 3D environment model, obtains the hardness of the ore surface, outputs the 3D environment model to the candidate impact point selection module, and outputs the hardness of the ore surface to the multi-objective optimization control algorithm module; Candidate strike point selection module: The three-dimensional environment model is segmented into equally spaced grids by the grid segmentation method. The center point of the grid is the candidate strike point, and the candidate strike point is output to the multi-objective optimization control algorithm module; Multi-objective optimization control algorithm module: Based on the candidate striking points and the hardness of the ore surface, the multi-objective optimization control algorithm is used to obtain the optimal striking point; the hardness of the ore surface at the optimal striking point is obtained through the three-dimensional environment model generation module; the optimal striking point is output to the path planning and optimization module, and the hardness of the ore surface at the optimal striking point is output to the adaptive striking force control module and the control signal generation module; Path planning and optimization module: Based on the optimal striking point, the optimal path is obtained using an adaptive multi-objective path planning algorithm. The speed and acceleration of the stone breaker at different times are calculated according to the optimal path, and the speed and acceleration of the stone breaker at different times are output to the adaptive striking force control module and the control signal generation module; Adaptive impact force control module: Based on the speed and acceleration of the stone breaker at different times and the hardness of the ore surface at the optimal impact point, the impact force is calculated using an adaptive impact force control algorithm and output to the control signal generation module; Control signal generation module: Based on the impact force, the speed of the crushing hammer at different times, the acceleration and the hardness of the ore surface at the optimal impact point, the control signal generation algorithm is used to generate the control signal of the crusher.

Images