A control method and system for bucket-wheel reclaimer based on black-backed jackal optimization algorithm

Abstract

This invention discloses a control method and system for a bucket wheel reclaimer based on the Blackback Jackal optimization algorithm, belonging to the field of automatic control technology for bucket wheel reclaimers. The method collects actual trajectory data and operating status data after the bucket wheel reclaimer completes the k-th reclaiming cycle. The actual trajectory data includes the trolley travel position, boom slewing angle, and boom pitch angle; the operating status data includes the bucket wheel motor current and reclaiming flow rate. The method calculates the trajectory tracking error based on the actual trajectory data and the desired trajectory data, and constructs a comprehensive fitness function by combining the control quantity change and the actuator operating constraints. The learning gain of the iterative learning controller is used as the optimization variable for the Blackback Jackal optimization algorithm. This method can adaptively correct the control quantity for the next reclaiming cycle based on the trajectory error and operating status of the previous reclaiming cycle, improving the trajectory tracking stability of the bucket wheel reclaimer under complex stockpile conditions and reducing the risk of bucket wheel motor overload and excessive reclaiming flow rate.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for bulk material handling equipment, specifically to a control method and system for a bucket wheel reclaimer based on the Blackback Jackal optimization algorithm. Background Technology

[0002] Bucket wheel reclaimers are crucial equipment in bulk material handling, stockpile transfer, and continuous reclaiming systems. They typically complete reclaiming operations through a combination of actions including trolley travel, boom rotation, boom pitch, and bucket wheel rotation. During reclaiming, the trajectory of the bucket wheel relative to the material pile directly affects reclaiming efficiency, uniformity of material collection, equipment load fluctuations, and the stability of the conveying system. Therefore, improving the trajectory tracking capability of bucket wheel reclaimers under complex operating conditions is a significant challenge in their automated control.

[0003] In actual operation, material piles are not ideally regular in shape. Influenced by the stacking method, material particle size, moisture content, density, natural collapse, and preceding material handling operations, the height, slope, density, and local resistance of the pile will change with time and spatial location. During the process of the bucket wheel cutting into, steadily handling, and exiting the pile, it is easily affected by factors such as sudden load changes, fluctuations in material handling flow, and the inertia of the actuator, causing the actual operating trajectory to deviate from the preset trajectory. This can lead to problems such as uneven material handling, localized over-cutting, empty handling, increased bucket wheel motor current, or impact on the actuator.

[0004] Among existing control methods for bucket wheel reclaimers, the most common are PID control based on fixed parameters, preset trajectory control, or speed regulation control based on empirical rules. These methods are simple to implement, have a good engineering application foundation, and can meet basic control requirements under conditions of relatively stable material pile shape and small load changes. However, when the material pile shape, density, or reclaiming resistance changes, fixed control parameters are difficult to adapt to changes in operating conditions in a timely manner, which can easily lead to a decrease in trajectory tracking accuracy. If the control parameters are increased to improve the response speed, it may cause sudden changes in control quantity, actuator vibration, or bucket wheel overload; if the parameter settings are too conservative, it will reduce the reclaiming efficiency.

[0005] Existing technologies also utilize methods such as 3D scanning, stockpile modeling, bucket wheel current detection, and belt flow detection to achieve automatic material handling control. For example, the material handling entry point, rotation angle, pitch angle, or material handling boundary can be determined by acquiring a 3D model of the stockpile; or the rotation speed can be adjusted based on bucket wheel motor current, belt scale, or flow detection data to improve the stability of the material handling flow rate. In related existing technologies, for example, patent document CN113003149B discloses a control method for automatic material handling in a boom-type bucket wheel stacker-reclaimer, which adjusts the material handling process through operating parameters such as bucket wheel current and material handling flow rate to improve the automatic material handling effect. Such solutions can improve the automation level of the material handling process and reduce manual intervention to some extent, but their control focus is usually on material handling path planning, entry posture determination, start and end angle judgment, or constant flow rate adjustment. They often lack mechanisms to further correct the control quantity for the next material handling cycle for trajectory errors that have already occurred in the previous material handling cycle, and they do not involve adaptive optimization of controller parameters based on trajectory errors between cycles.

[0006] In addition, some control methods employ adaptive control, sliding mode control, and other approaches to improve the position or attitude control performance of bucket wheel reclaimers. For example, patent document CN115771773B discloses a method for tracking control of the bucket wheel position by establishing a three-degree-of-freedom dynamic model of the bucket wheel reclaimer and designing a sliding mode control law based on the error between the given position and the actual position. This type of method can enhance the system's robustness to disturbances and parameter changes to some extent, but it usually requires establishing a relatively accurate equipment dynamic model or relying on specific control law designs. Due to the characteristics of bucket wheel reclaimers, such as large inertia, multiple coupled actuators, and significant load changes with material state, deviations can easily exist between model parameters and actual operating conditions, and the control effect may still fluctuate under different material pile conditions. Furthermore, this type of scheme mainly focuses on position or attitude tracking control during a single operation, failing to fully utilize the trajectory error generated in the previous reclaiming cycle to correct the control quantity for the next reclaiming cycle, and lacking a mechanism to uniformly incorporate trajectory error, control quantity changes, and safety constraints such as bucket wheel motor current and reclaiming flow rate into the optimization evaluation.

[0007] Iterative learning control is suitable for controlled objects with repetitive operation characteristics. Bucket wheel reclaimers exhibit a certain degree of cyclical repetition during layered, segmented, or reciprocating material reclaiming processes. Therefore, theoretically, error data from previous reclaiming cycles can be used to correct the control input for subsequent reclaiming cycles. However, the actual effectiveness of iterative learning control is closely related to the selection of the learning gain. Too small a learning gain leads to slow error convergence, making it difficult to promptly correct trajectory deviations; too large a learning gain may cause control oscillations, overshoot, or actuator impacts. Traditional methods of tuning the learning gain using manual experience or trial and error are labor-intensive and difficult to adapt to continuously changing field conditions involving varying stockpile shapes, material densities, and equipment loads.

[0008] Therefore, existing technologies still have the following shortcomings: First, fixed parameter control is difficult to adapt to changes in stockpile conditions, and trajectory tracking performance is prone to decline; second, some automatic material handling schemes mainly focus on path planning, flow regulation, or boundary judgment, and do not make sufficient use of errors between material handling cycles; third, adaptive control, sliding mode control, and other schemes usually rely on equipment dynamics models or specific control laws, and the control effect is prone to fluctuation when there are deviations between model parameters and actual operating conditions; fourth, the learning gain of ordinary iterative learning control is difficult to automatically adjust according to the actual material handling status; fifth, if only the reduction of trajectory error is pursued, it may bring safety risks such as sudden changes in control quantity, overload of bucket wheel motor, or excessive material handling flow. Based on the above problems, it is necessary to provide a bucket wheel reclaimer control method and system based on the black-backed jackal optimization algorithm to improve the trajectory tracking stability and operational safety of the bucket wheel reclaimer under complex stockpile conditions. Summary of the Invention

[0009] The technical problem this invention aims to solve is to overcome the shortcomings of existing bucket wheel reclaimers, where fixed parameter control or preset trajectory control is difficult to adaptively adjust when material pile shape, density, and reclaiming resistance change. This leads to decreased trajectory tracking accuracy, uneven material reclaiming, sudden changes in control quantities, and increased risk of equipment overload. The invention provides a bucket wheel reclaimer control method and system based on the Blackback Jackal optimization algorithm. By collecting actual trajectory data and operating status data during the reclaiming cycle, the Blackback Jackal optimization algorithm is used to optimize the learning gain of the iterative learning controller, and multi-channel control quantities for the next reclaiming cycle are generated under safety constraints, thereby effectively solving the problems in the background technology.

[0010] To achieve the above objectives, the present invention provides the following technical solution: a control method for a bucket wheel reclaimer based on the black-backed jackal optimization algorithm, comprising the following steps:

[0011] After the bucket wheel reclaimer completes the kth material reclaiming cycle, the actual trajectory data and operating status data of the kth material reclaiming cycle are collected. The actual trajectory data includes the trolley travel position, boom slewing angle and boom pitch angle, and the operating status data includes the bucket wheel motor current and material reclaiming flow rate.

[0012] Obtain the expected trajectory data corresponding to the k-th material handling cycle, and calculate the trajectory tracking error based on the expected trajectory data and the actual trajectory data:

[0013]

[0014] in, The trajectory tracking error of the i-th sampling point in the k-th material handling cycle is... For the expected trajectory data of the i-th sampling point, This represents the actual trajectory data of the i-th sampling point in the k-th material handling cycle;

[0015] A comprehensive fitness function is constructed based on trajectory tracking error, control quantity change, and actuator operation constraints.

[0016] The learning gain of the iterative learning controller is used as the variable to be optimized in the black-backed jackal optimization algorithm, and the fitness corresponding to different learning gains is evaluated by the comprehensive fitness function to obtain the optimal learning gain corresponding to the k-th material feeding cycle.

[0017] Substitute the optimal learning gain into the iterative learning control law to generate the control quantity for the (k+1)th material feeding cycle.

[0018] The control quantity of the (k+1)th material handling cycle is limited and verified. If the control quantity meets the operating constraints of the actuator, the control quantity is converted into drive commands for the trolley traveling mechanism, boom slewing mechanism and boom pitching mechanism.

[0019] After the (k+1)th material handling cycle is completed, the control quantity for the next material handling cycle is updated based on the actual trajectory data and operating status data of that cycle, so as to achieve inter-cycle adaptive correction of the material handling trajectory of the bucket wheel reclaimer.

[0020] Furthermore, the actual trajectory data is represented as follows:

[0021]

[0022] in, Let i be the position of the trolley at the i-th sampling point in the k-th material handling cycle. For boom slewing angle, The boom pitch angle;

[0023] The control quantity is expressed as:

[0024]

[0025] in, This is the command for the travel speed of the large vehicle. For boom slewing angular velocity command, This is the boom pitch rate command.

[0026] Furthermore, the comprehensive fitness function includes at least a trajectory error term, a control smoothing term, and a safety constraint term; the trajectory error term is used to characterize the deviation between the actual trajectory data and the expected trajectory data; the control smoothing term is used to characterize the magnitude of the change in control quantity between adjacent material handling cycles; and the safety constraint term is used to characterize whether at least one of the bucket wheel motor current, material handling flow rate, and actuator control quantity exceeds the corresponding safety threshold.

[0027] The overall fitness function satisfies:

[0028]

[0029] in, Let L be the overall fitness value corresponding to the learning gain to be optimized. For trajectory error term, To control the smoothing term, For safety constraint terms, λ is the weight coefficient of the control smoothing term, and μ is the weight coefficient of the safety constraint term;

[0030] The trajectory error term is determined based on the trajectory tracking error of multiple sampling points within a material handling cycle. The control smoothing term is determined based on the change between the candidate control quantity of the (k+1)th material handling cycle and the control quantity of the kth material handling cycle. The safety constraint term is determined based on the exceedance of the bucket wheel motor current, material handling flow rate, and actuator control quantity relative to the corresponding safety threshold.

[0031] Furthermore, the learning gain includes a proportional learning gain and a differential learning gain corresponding to the trolley travel channel, the boom slewing channel, and the boom pitch channel, respectively, and the learning gain is expressed as:

[0032]

[0033] in, These represent the proportional learning gain and the differential learning gain of the vehicle's travel lane, respectively. These are the proportional learning gain and differential learning gain of the boom slewing channel, respectively. These are the proportional learning gain and differential learning gain of the boom pitch channel, respectively.

[0034] Furthermore, optimizing the learning gain using the black-backed jackal optimization algorithm includes:

[0035] Multiple black-backed jackals are generated within a preset learning gain search range, with each black-backed jackal representing a set of candidate learning gains.

[0036] Calculate the fitness value of each individual black-backed jackal based on the comprehensive fitness function.

[0037] Dominant individuals are determined based on their fitness values, and the dominant individuals include at least one black-backed jackal individual with the best current fitness value.

[0038] Based on the positional relationship between the dominant individual and the other black-backed jackal individuals, update the candidate learning gain corresponding to each black-backed jackal individual;

[0039] Boundary verification is performed on the updated candidate learning gains to ensure that the updated candidate learning gains remain within the preset learning gain search range;

[0040] Repeat the fitness evaluation and candidate learning gain update until the preset number of iterations is reached or the fitness value meets the convergence condition, and output the candidate learning gain with the best fitness value as the optimal learning gain.

[0041] Furthermore, the iterative learning control law is a PD-type iterative learning control law, and the control quantity of the (k+1)th material picking cycle is generated based on the control quantity of the kth material picking cycle, the trajectory tracking error, the change in trajectory tracking error, and the optimal learning gain.

[0042] The change in trajectory tracking error is determined by the trajectory tracking error of adjacent sampling points within the k-th material feeding cycle.

[0043] Furthermore, the PD-type iterative learning control law satisfies:

[0044]

[0045] in, This is the control quantity for the i-th sampling point in the (k+1)-th material feeding cycle. This is the control quantity for the i-th sampling point in the k-th material feeding cycle. To achieve the optimal ratio of learning gain, The optimal differential learning gain is given by Δt, where Δt is the sampling period.

[0046] When the control quantity is a multi-channel control quantity These are the optimal proportional learning gain matrix and the optimal differential learning gain matrix corresponding to the trolley travel channel, boom slewing channel, and boom pitch channel, respectively.

[0047] Furthermore, a bucket wheel reclaimer control system based on the Blackjack algorithm is provided. The system includes a sensor acquisition unit, a data processing controller, a desired trajectory storage unit, a Blackjack optimization controller, an iterative learning controller, a safety constraint controller, a PLC control unit, and an actuator drive unit. The signal output of the sensor acquisition unit is connected to the signal input of the data processing controller. The data output of the desired trajectory storage unit is connected to the data input of the data processing controller. The data output of the data processing controller is connected to the data inputs of both the Blackjack optimization controller and the iterative learning controller. The learning gain output of the Blackjack optimization controller is connected to the learning gain input of the iterative learning controller. The control output of the iterative learning controller is connected to the control input of the safety constraint controller. The control output of the safety constraint controller is connected to the instruction input of the PLC control unit. The control output of the PLC control unit is connected to the control input of the actuator drive unit. The actuator drive unit is connected to the trolley traveling mechanism, the boom slewing mechanism, and the boom pitching mechanism.

[0048] The sensor acquisition unit includes a trolley position sensor, a boom slewing angle sensor, a boom pitch angle sensor, a bucket wheel motor current sensor, and a material handling flow sensor; the trolley position sensor, boom slewing angle sensor, boom pitch angle sensor, bucket wheel motor current sensor, and material handling flow sensor are respectively connected to the data processing controller;

[0049] The data processing controller includes a sampling synchronization module, an error calculation module, and a fitness construction module. The input of the sampling synchronization module is connected to the sensor acquisition unit, and the output of the sampling synchronization module is connected to the input of the error calculation module. The error calculation module is also connected to the desired trajectory storage unit, and the output of the error calculation module is connected to the input of the fitness construction module. The output of the fitness construction module is connected to the Blackback Jackal optimization controller.

[0050] Furthermore, the safety constraint controller includes a control quantity limiting module, a previous valid control quantity storage module, and an attenuation processing module; the control quantity output terminal of the iterative learning controller is connected to the input terminals of the control quantity limiting module and the attenuation processing module, respectively; the control quantity limiting module is bidirectionally connected to the previous valid control quantity storage module; the output terminal of the control quantity limiting module is connected to the instruction input terminal of the PLC control unit.

[0051] When the control quantity limiting module determines that the control quantity of the (k+1)th material feeding cycle exceeds the allowable range of the actuator, or the bucket wheel motor current or the material feeding flow exceeds the preset safety threshold, the control quantity limiting module calls the previous valid control quantity stored in the previous valid control quantity storage module, or receives the attenuation control quantity output by the attenuation processing module, and sends the previous valid control quantity or attenuation control quantity to the PLC control unit.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] 1. This invention utilizes the layered, segmented, and cyclical material reclaiming characteristics of a bucket wheel reclaimer. It compares the actual trajectory in the previous reclaiming cycle with the expected trajectory to obtain the trajectory tracking error, and uses this error to correct the control quantity in the next reclaiming cycle. Compared with control methods that rely solely on fixed parameters or preset trajectories, this invention can transform the trajectory deviation that has occurred into a basis for subsequent control. This allows the bucket wheel reclaimer to gradually correct the reclaiming trajectory even when the pile shape, material density, and reclaiming resistance change, thereby improving the trajectory tracking stability during repeated reclaiming processes.

[0054] 2. This invention uses the learning gain of the iterative learning controller as the optimization object of the black-backed jackal optimization algorithm, and evaluates the control effect of different learning gains under the current material handling conditions through a comprehensive fitness function. This method avoids the problem of repeatedly trying to adjust the learning gain based on human experience, and enables the learning gain to be adjusted according to the trajectory error and operating state changes in the material handling cycle, thereby improving the problems of slow convergence, easy oscillation or insufficient adaptability of fixed learning gain under complex conditions.

[0055] 3. When constructing the comprehensive fitness function, this invention simultaneously considers trajectory error, control quantity changes, and actuator operation constraints. Therefore, the optimization process does not simply pursue the minimum trajectory error, but rather, while improving trajectory tracking accuracy, it constrains situations such as sudden changes in control quantity, excessive current of the bucket wheel motor, and excessive material reclaiming flow rate. This reduces the risk of actuator impact and equipment overload, making the optimized control quantity more suitable for the on-site operation requirements of the bucket wheel reclaimer.

[0056] 4. This invention implements control quantities into multiple execution channels such as trolley travel, boom rotation, and boom pitch, and performs amplitude limiting verification before outputting the control quantities. When a candidate control quantity does not meet the operational constraints, the previous effective control quantity or the attenuated control quantity is used as a substitute. Through the above-mentioned multi-channel collaborative control and safety backoff mechanism, this invention can improve the material reclaiming trajectory correction capability while ensuring that the control commands match the actual operating capabilities of the actuators, thereby enhancing the continuity and safety of the bucket wheel reclaimer under complex material pile conditions. Attached Figure Description

[0057] Figure 1 This is an overall flowchart of the control method of the present invention;

[0058] Figure 2 This is a block diagram of the control system structure of the present invention;

[0059] Figure 3 This is a schematic diagram of the multi-channel control object of the bucket wheel reclaimer of the present invention;

[0060] Figure 4 This is a schematic diagram illustrating the construction of the comprehensive fitness function of this invention;

[0061] Figure 5 This is a flowchart of the black-backed jackal optimization learning gain process of the present invention;

[0062] Figure 6 This is a block diagram of the PD-type iterative learning control closed loop of the present invention;

[0063] Figure 7 This is a flowchart of the safety constraints and rollback process of the present invention. Detailed Implementation

[0064] The embodiments of the present invention will be further described below with reference to the accompanying drawings. It should be understood that the following embodiments are used to illustrate the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention. All equivalent substitutions or conventional parameter adjustments made by those skilled in the art based on the model of the bucket wheel reclaimer, the drive method, the sensor configuration, the control system platform, and the on-site working conditions without departing from the technical concept of the present invention should be understood as falling within the scope of implementation of the present invention.

[0065] This embodiment provides a control method and system for a bucket wheel reclaimer based on the Blackback Jackal optimization algorithm. This method is applicable to bucket wheel reclaimers with repetitive reclaiming cycles, especially suitable for layered, segmented, or reciprocating reclaiming scenarios. In each reclaiming cycle, the bucket wheel reclaimer completes the cutting and reclaiming of the material pile through actions such as trolley travel, boom rotation, boom pitch, and bucket wheel rotation. Because the actual shape, density, particle size, moisture content, and local collapse of the material pile are constantly changing, the actual running trajectory of the bucket wheel is prone to deviating from the expected trajectory. This embodiment utilizes the trajectory error and running status data from the previous reclaiming cycle to optimize the learning gain of the iterative learning controller, and uses the optimized learning gain to generate the control quantity for the next reclaiming cycle, thereby achieving adaptive correction of the reclaiming trajectory between cycles.

[0066] I. Controlled Objects and System Composition

[0067] like Figure 2 As shown, the bucket wheel reclaimer control system in this embodiment includes a sensor acquisition unit, a data processing controller, a desired trajectory storage unit, a black-backed jackal optimization controller, an iterative learning controller, a safety constraint controller, a PLC control unit, and an actuator drive unit.

[0068] The sensor acquisition unit is used to collect the actual trajectory data and operating status data of the bucket wheel reclaimer during the material reclaiming cycle. The sensor acquisition unit may include a trolley position sensor, a boom slewing angle sensor, a boom pitch angle sensor, a bucket wheel motor current sensor, and a material reclaiming flow sensor. Among them, the trolley position sensor is used to detect the trolley's traveling position along the track direction, the boom slewing angle sensor is used to detect the boom's horizontal slewing angle, the boom pitch angle sensor is used to detect the boom's pitch angle, the bucket wheel motor current sensor is used to detect the bucket wheel motor's operating current, and the material reclaiming flow sensor is used to detect the material flow rate entering the belt conveyor system.

[0069] The signal output terminals of each sensor are connected to the signal input terminals of the data processing controller. The data processing controller is used to perform sampling synchronization, filtering, outlier removal, and trajectory error calculation on the data collected by each sensor. The data output terminal of the desired trajectory storage unit is connected to the data input terminal of the data processing controller and is used to provide the data processing controller with the desired trajectory data corresponding to the current material handling task.

[0070] The data output terminal of the data processing controller is connected to the data input terminals of the Blackjack optimization controller and the iterative learning controller, respectively. The Blackjack optimization controller is used to optimize the learning gain of the iterative learning controller based on the trajectory tracking error, the change in control quantity, and the operating constraints of the actuator. The learning gain output terminal of the Blackjack optimization controller is connected to the learning gain input terminal of the iterative learning controller. The iterative learning controller generates the control quantity for the next material handling cycle based on the optimal learning gain and the trajectory tracking error.

[0071] The control output terminal of the iterative learning controller is connected to the control input terminal of the safety constraint controller; the safety constraint controller is used to perform amplitude limiting verification on the candidate control quantity, and to perform backoff processing or attenuation processing when the candidate control quantity does not meet the actuator operation constraints; the control output terminal of the safety constraint controller is connected to the instruction input terminal of the PLC control unit; the control output terminal of the PLC control unit is connected to the actuator drive unit; the actuator drive unit drives the trolley traveling mechanism, boom slewing mechanism and boom pitching mechanism respectively.

[0072] In actual deployment, the Blackback Jackal optimization controller, iterative learning controller, and safety constraint controller can be integrated into the bucket wheel reclaimer's electrical control box, or they can be installed in a PLC control cabinet, industrial computer, edge computing controller, or upper control system. The above controllers can be implemented as independent hardware or deployed as control software modules on the same industrial control platform.

[0073] II. Material Recycling Cycle Data Collection

[0074] like Figure 1As shown, after the bucket wheel reclaimer completes the kth reclaiming cycle, the control system collects the actual trajectory data and operating status data of that reclaiming cycle; M sampling points are set within one reclaiming cycle, and each sampling point is denoted as:

[0075]

[0076] The sampling period between adjacent sampling points is:

[0077]

[0078] Where i = 2, 3, ..., M;

[0079] In the k-th material handling cycle, the actual trajectory data of the i-th sampling point is represented as follows:

[0080]

[0081] in, This indicates the travel position of the large vehicle. For boom slewing angle, The boom pitch angle; the operating status data includes at least the bucket wheel motor current and the material handling flow rate, expressed as follows:

[0082]

[0083] in, Let represent the bucket wheel motor current at the ith sampling point in the kth material handling cycle. This represents the material flow rate at the corresponding sampling point.

[0084] In one alternative implementation, the operating status data may further include hydraulic pressure, boom vibration, bucket wheel speed, belt scale data, or stockpile scanning data; the aforementioned additional data may be used to expand safety constraints or assist in judging abnormal material handling conditions.

[0085] To ensure that data from different channels correspond to the same material feeding time, the data processing controller timestamps the data from each sensor. For short-term pulse interference, moving average filtering, median filtering, or low-pass filtering can be used for processing. For data that significantly exceeds the sensor's range, it can be discarded as an anomaly or replaced with the previous valid sample value.

[0086] III. Calculation of Trajectory Error and Error Change

[0087] The data processing controller calculates the trajectory tracking error based on the actual trajectory data and the expected trajectory data of the k-th material handling cycle:

[0088]

[0089] When expanded, it appears as follows:

[0090]

[0091] in, This is due to the trajectory error of the large vehicle traveling in the lane. For the boom slewing path trajectory error, This refers to the trajectory error of the boom pitch channel.

[0092] The variation in trajectory error is determined by the trajectory error of adjacent sampling points within the same feeding cycle:

[0093]

[0094] in, Let be the rate of change of the trajectory error at the i-th sampling point in the k-th material handling cycle; for the first sampling point t1, we can let:

[0095]

[0096] Alternatively, the trajectory error of the sampling point at the end of the previous material handling cycle can be used for calculation.

[0097] In one alternative implementation, to improve the lead time of the control quantity generation, the prediction error can be obtained by linear extrapolation based on the trajectory error variation trend of adjacent sampling points; the prediction error can be expressed as:

[0098]

[0099] When the sampling period is constant, we have:

[0100]

[0101] Right now:

[0102]

[0103] in, The error of the next sampling point is predicted based on the error change trend of adjacent sampling points in the k-th material feeding cycle. This prediction error can be used as the error input in the PD-type iterative learning control law, or weighted with the current error and used as the basis for control quantity correction. Δt is the sampling period.

[0104] IV. Definition of Control Quantities

[0105] like Figure 3 As shown, the multi-channel control objects of the bucket wheel reclaimer include the trolley traveling mechanism, the boom slewing mechanism, and the boom pitching mechanism. The bucket wheel is located at the front end of the boom and forms a material reclaiming interface with the material pile. The trolley traveling direction is denoted as x, the boom slewing angle is denoted as θ, and the boom pitching angle is denoted as φ.

[0106] In this embodiment, the control quantity of the bucket wheel reclaimer is a multi-channel control quantity, including the trolley travel speed command, the boom slewing angular velocity command, and the boom pitch angular velocity command; the control quantity of the i-th sampling point in the k-th reclaiming cycle is expressed as:

[0107]

[0108] in, This is the command for the travel speed of the large vehicle. For boom slewing angular velocity command, This is the boom pitch rate command.

[0109] After being converted by the PLC control unit, the above control quantities can be converted into frequency converter commands, servo drive commands, hydraulic valve opening commands, or other control signals suitable for field actuators. For trolley traveling mechanisms and boom slewing mechanisms using frequency converter drives, the trolley traveling speed commands and boom slewing angular velocity commands can be converted into corresponding frequency converter setpoints. For boom pitching mechanisms using hydraulic drives, the boom pitching angular velocity commands can be converted into hydraulic proportional valve control quantities.

[0110] V. Learning Gain Settings

[0111] This implementation adopts PD-type iterative learning control; the learning gain includes the proportional learning gain and differential learning gain corresponding to the trolley travel channel, boom slewing channel, and boom pitch channel, respectively, and the learning gain vector is expressed as:

[0112]

[0113] in, These represent the proportional learning gain and the differential learning gain of the vehicle's travel lane, respectively. These are the proportional learning gain and differential learning gain of the boom slewing channel, respectively. These are the proportional learning gain and differential learning gain of the boom pitch channel, respectively.

[0114] To facilitate multi-channel matrix operations, the proportional learning gain and the differential learning gain can also be written as diagonal matrices:

[0115]

[0116] Each learning gain has a set search range; for example:

[0117]

[0118] The search range can be determined based on the rated speed of the bucket wheel reclaimer, boom inertia, actuator response capability, allowable acceleration, and on-site commissioning experience. The search range should be set to ensure that the control quantity corresponding to the candidate learning gain does not significantly exceed the acceptable range of the actuator.

[0119] VI. Construction of the Comprehensive Fitness Function

[0120] like Figure 4 As shown, the Blackback Jackal optimized controller evaluates the control effect of each candidate learning gain through a comprehensive fitness function; the comprehensive fitness function simultaneously considers trajectory error, control quantity variation, and safety constraints, and is expressed as:

[0121]

[0122] in, The overall fitness value corresponding to the candidate learning gain L. For trajectory error term, To control the smoothing term, λ is the safety constraint term, λ is the weight coefficient of the control smoothing term, and μ is the weight coefficient of the safety constraint term.

[0123] The trajectory error term characterizes the deviation between the actual trajectory and the desired trajectory within a material handling cycle; in one implementation, the trajectory error term is:

[0124]

[0125] in, This is the trajectory error weight matrix, used to set the error weights for the trolley travel path, boom slewing path, and boom pitch path.

[0126] The trajectory error term can also be written in norm form:

[0127]

[0128] in:

[0129]

[0130] The control smoothing term characterizes the change in the control quantity of the (k+1)th fetch cycle relative to the control quantity of the kth fetch cycle generated by the candidate learning gain; for the candidate learning gain L, the candidate control quantity is obtained according to the iterative learning control law. Then the smoothing control term is:

[0131]

[0132] Where is the control quantity change weight matrix.

[0133] Safety constraints are used to characterize whether the bucket wheel motor current, material handling flow rate, and actuator control quantity exceed the corresponding safety threshold; in one embodiment, the safety constraint is:

[0134]

[0135] in, This refers to the upper limit of the allowable current for the bucket wheel motor. To allow a maximum material handling capacity, The upper limit of the control quantity allowed by the actuator; when When it is a vector, it corresponds to the upper limit of the trolley travel speed, the upper limit of the boom slewing angular velocity, and the upper limit of the boom pitch angular velocity, respectively.

[0136] To reduce the impact of different physical quantities on fitness evaluation, the current, flow, and control terms can be normalized; for example, the current over-limit term can be written as:

[0137]

[0138] The traffic limit exceeding item can be written as:

[0139]

[0140] In another implementation, when the field control system only needs to use trajectory error as the primary optimization objective, the weights of the control smoothing term and safety constraint term can be reduced, or λ and μ can be set to smaller values; in this case, the comprehensive fitness function can degenerate into the trajectory error integral performance index:

[0141]

[0142] or:

[0143]

[0144] This degradation mode is suitable for working conditions where the actuator has a large margin, the material flow rate fluctuates little, or the trajectory accuracy needs to be adjusted first. In actual operation, it is preferable to consider trajectory error, control smoothness, and safety constraints at the same time to ensure that the control quantity is more suitable for the on-site operation of large bucket wheel reclaimers.

[0145] VII. Black-backed Jackal Optimization Learning Gain

[0146] like Figure 5 As shown, the black-backed jackal optimization controller uses the learning gain of the iterative learning controller as the variable to be optimized; the position of each black-backed jackal corresponds to a set of candidate learning gains:

[0147]

[0148] Among them, X j This represents the position of the j-th black-backed jackal individual, j=1,2,…,N, where N is the population size;

[0149] Initialize the population within the preset learning gain search range:

[0150]

[0151] in, Let be the initial position of the j-th individual. To learn the lower bound vector of gain, To learn the upper bound vector of gain, It is a random vector between 0 and 1.

[0152] For each individual black-backed jackal, its corresponding candidate learning gain is substituted into the iterative learning control law to obtain the candidate control quantity. Then, the fitness value is calculated through the comprehensive fitness function; the smaller the fitness value, the better the overall effect of the candidate learning gain in terms of trajectory error, control smoothness and safety constraints.

[0153] In each iteration, dominant individuals are determined based on their fitness values. In one specific implementation, the individuals with the best, second-best, and third-best fitness values ​​are designated as the first dominant individual, the second dominant individual, and the third dominant individual, respectively, and their positions are represented as follows:

[0154]

[0155] Where g represents the current iteration number.

[0156] The remaining individuals are updated based on the position of the dominant individual; the individual distance vector can be represented as:

[0157]

[0158] in, This indicates the position of the j-th black-backed jackal individual in generation g. For random disturbance coefficients.

[0159] In one specific implementation, the individual location is updated as follows:

[0160]

[0161] in, Let represent the position of the j-th black-backed jackal individual in the (g+1)th generation, and represent the convergence adjustment coefficient.

[0162] The convergence adjustment coefficient and the random disturbance coefficient can satisfy:

[0163]

[0164] Where r1 and r2 are random numbers between 0 and 1, and a is an adjustment factor that decreases with the number of iterations; the adjustment factor can decrease in the following way:

[0165]

[0166] Among them, G max This represents the maximum number of iterations.

[0167] After the individual position is updated, boundary checks are performed on each candidate learning gain; for the m-th learning gain component... If it is less than the corresponding lower limit, then let:

[0168]

[0169] If it is greater than the corresponding upper limit, then let:

[0170]

[0171] Repeat the fitness evaluation, dominant individual determination, individual position update, and boundary verification processes until the maximum number of iterations is reached, or the change in optimal fitness across several consecutive generations is less than a preset convergence threshold; finally, the individual with the best fitness value is output as the optimal learning gain.

[0172]

[0173] It should be noted that the above individual update method is a specific implementation of the black-backed jackal optimization algorithm. In practical applications, the population size, number of dominant individuals, random perturbation method, boundary treatment method and convergence condition can be adjusted without changing the technical idea of ​​"using the learning gain of the iterative learning controller as the individual position and optimizing based on the comprehensive fitness function".

[0174] 8. Generation of PD-type iterative learning control variables

[0175] like Figure 6 As shown, after the Blackback Jackal optimization controller outputs the optimal learning gain, the iterative learning controller substitutes the optimal learning gain into the PD-type iterative learning control law to generate the control quantity for the (k+1)th material feeding cycle.

[0176] In one implementation, the PD-type iterative learning control law is:

[0177]

[0178] in, This is the control quantity for the i-th sampling point in the (k+1)-th material feeding cycle. This is the control quantity for the i-th sampling point in the k-th material feeding cycle. To achieve the optimal ratio of learning gain, This represents the optimal differential learning gain.

[0179] When using multi-channel control, it can be written as follows:

[0180]

[0181] Therefore, the multi-channel control quantity for the (k+1)th material feeding cycle can be expressed as:

[0182]

[0183] in, These represent the error variation rates of the trolley travel path, boom slewing path, and boom pitch path, respectively.

[0184] In another implementation, the prediction error obtained through linear extrapolation can be used to update the control input; in this case, the control law can be written as:

[0185]

[0186] in, The error is predicted based on the error change trend of adjacent sampling points; this method enables the control quantity correction to have a certain predictability and is suitable for material feeding conditions where the error change trend is obvious.

[0187] In another implementation, the current error can be weighted with the prediction error:

[0188]

[0189] Where γ is a weighting coefficient between 0 and 1; in this case, the control law can be written as:

[0190]

[0191] The above linear extrapolation method is an optional implementation method. In practical applications, it can be selected whether to enable it based on the computing power of the field controller and the system response speed.

[0192] To further avoid abrupt changes in control values ​​between material handling cycles, the generated control values ​​can also be smoothed:

[0193]

[0194] in, The smoothed control variable is η, which is a smoothing coefficient between 0 and 1.

[0195] IX. Safety Limiting, Downtime, and Attenuation Handling

[0196] like Figure 7 As shown, after the iterative learning controller generates the candidate control quantity for the (k+1)th material feeding cycle, the safety constraint controller performs a limit verification on the control quantity; the limit verification includes at least the following:

[0197] Determine whether the vehicle's travel speed command is within the permissible range:

[0198]

[0199] Determine if the boom slewing angular velocity command is within the allowable range:

[0200]

[0201] Determine if the boom pitch rate command is within the allowable range:

[0202]

[0203] Determine whether the change in control quantity between adjacent sampling points exceeds the allowable change amount:

[0204]

[0205] Determine if the bucket wheel motor current is within the safe threshold range:

[0206]

[0207] Determine if the material handling flow rate is within the safe threshold range:

[0208]

[0209] If all the above conditions are met, the safety constraint controller will output the control quantity of the (k+1)th material handling cycle to the PLC control unit, which will then convert it into the corresponding actuator drive instruction.

[0210] If the candidate control quantity of the (k+1)th material handling cycle does not meet the actuator operation constraints, or if the comprehensive fitness value calculated based on the candidate control quantity is greater than the comprehensive fitness value corresponding to the kth material handling cycle, then the safety constraint controller stops using the candidate control quantity and performs rollback or attenuation processing.

[0211] Rollback processing refers to using the previous valid control variable as the control variable for the (k+1)th material handling cycle:

[0212]

[0213] in, This is the effective control quantity that satisfies the safety constraints of the previous set.

[0214] Attenuation processing refers to proportionally reducing the change in the candidate control variable relative to the control variable in the previous cycle.

[0215]

[0216] in, ρ is the original candidate control quantity calculated by the iterative learning control law, and ρ is the attenuation coefficient that is greater than 0 and less than 1.

[0217] In one implementation, when the bucket wheel motor current or the material handling flow exceeds the safety threshold, the previous effective control quantity is used first; when only the change in the control quantity exceeds the threshold, the control quantity after attenuation can be used; the above strategy can be set according to the equipment safety level and on-site control requirements.

[0218] 10. Specific Operation Process

[0219] Before the bucket wheel reclaimer starts operation, the operator or the upper-level dispatching system sets the expected trajectory corresponding to the current material reclaiming task, and sets the upper limit of bucket wheel motor current, upper limit of material reclaiming flow rate, upper limit of trolley travel speed, upper limit of boom slewing angular velocity, upper limit of boom pitch angular velocity, and upper limit of control quantity change.

[0220] After the bucket wheel reclaimer completes the kth reclaiming cycle, the sensor acquisition unit sends the trolley travel position, boom slewing angle, boom pitch angle, bucket wheel motor current, and reclaiming flow rate of the reclaiming cycle to the data processing controller. The data processing controller synchronously processes the acquired data and calculates the trajectory tracking error for each sampling point.

[0221] Subsequently, the Blackjackal optimization controller generates several sets of candidate learning gains within a preset learning gain search range; each set of candidate learning gains corresponds to a Blackjackal individual; the Blackjackal optimization controller substitutes the candidate learning gains into the iterative learning control law to generate the corresponding candidate control quantity, and calculates the comprehensive fitness value based on the trajectory error term, control smoothing term, and safety constraint term.

[0222] After multiple rounds of fitness evaluation and candidate learning gain updates, the Blackback Jackal optimization controller outputs the learning gain with the optimal fitness value; the iterative learning controller substitutes the optimal learning gain into the PD-type iterative learning control law to generate the trolley travel speed command, boom slewing angular velocity command, and boom pitch angular velocity command for the k+1th material handling cycle.

[0223] The safety constraint controller performs amplitude limit verification on the above control quantity; if the control quantity meets the safety requirements, it is sent to the PLC control unit; if the control quantity does not meet the safety requirements, the previous valid control quantity or the control quantity after attenuation processing is called; the PLC control unit converts the safety control quantity into actuator drive instructions, so that the trolley traveling mechanism, boom slewing mechanism and boom pitching mechanism execute the k+1th material handling cycle according to the corrected control quantity.

[0224] After the (k+1)th material reclaiming cycle is completed, the system collects the actual trajectory data and operating status data again, and enters the next cycle correction process. Through the above process, the control system can continuously use the trajectory error and operating status data of the previous material reclaiming cycle to correct the control quantity of the next material reclaiming cycle, thereby improving the trajectory tracking stability of the bucket wheel reclaimer under complex material pile conditions.

[0225] XI. Optional Implementation Methods

[0226] In one alternative implementation, the actual trajectory data may further include the spatial coordinates of the bucket wheel center; the spatial coordinates of the bucket wheel center can be calculated based on the trolley travel position, boom slewing angle, boom pitch angle and boom length; in this case, the trajectory tracking error can be determined by the deviation between the spatial coordinates of the bucket wheel center and the desired spatial coordinates of the bucket wheel center.

[0227] In one alternative implementation, the operating status data may further include hydraulic pressure and boom vibration signals; in this case, the safety constraints may be further supplemented with hydraulic pressure over-limit penalties and boom vibration over-limit penalties to improve the ability to identify overload, impact, or collapse conditions.

[0228] In one alternative implementation, the Blackback Jackal optimization controller can run after each material handling cycle or once every few material handling cycles. When the material pile conditions change slowly, the optimization frequency can be reduced to reduce the computational load on the control system. When a significant increase in trajectory error, a significant increase in bucket wheel motor current fluctuation, or a significant increase in material handling flow rate fluctuation is detected, the optimization frequency can be increased.

[0229] In one alternative implementation, the desired trajectory data can come from a manually preset trajectory table, or from a three-dimensional stockpile model, a material handling task planning system, or a host scheduling system. This invention does not limit the specific source of the desired trajectory data, as long as the desired trajectory data corresponds to the actual trajectory data at the sampling points.

[0230] In one alternative implementation, the control system can be integrated into the bucket wheel reclaimer's electrical control box, or it can be implemented by a PLC located in the electrical control box and optimized control software located in an industrial computer or edge computing controller. For existing bucket wheel reclaimer retrofit projects, the original PLC and actuator drive unit can be retained, and only the software function modules of the data processing controller, the Blackback Jackal optimization controller, and the iterative learning controller can be added.

[0231] The above implementation methods can be used in combination with each other; as long as they still adopt the technical route of "collecting actual trajectory data and running status data of the material picking cycle, calculating trajectory tracking error, optimizing the learning gain of the iterative learning controller with the black-backed jackal optimization algorithm, generating multi-channel control quantities for the next material picking cycle, and performing safety limiting or rollback processing", they fall within the scope of the technical solution disclosed in this invention.

[0232] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Various changes and modifications can be made to the present invention without departing from the spirit and scope thereof, and all such changes and modifications fall within the scope of the present invention as claimed.

Claims

1. A control method for a bucket wheel reclaimer based on the black-backed jackal optimization algorithm, characterized in that, Includes the following steps: After the bucket wheel reclaimer completes the kth material reclaiming cycle, the actual trajectory data and operating status data of the kth material reclaiming cycle are collected. The actual trajectory data includes the trolley travel position, boom slewing angle and boom pitch angle, and the operating status data includes the bucket wheel motor current and material reclaiming flow rate. Obtain the expected trajectory data corresponding to the k-th material handling cycle, and calculate the trajectory tracking error based on the expected trajectory data and the actual trajectory data: in, The trajectory tracking error of the i-th sampling point in the k-th material handling cycle is... For the expected trajectory data of the i-th sampling point, This represents the actual trajectory data of the i-th sampling point in the k-th material handling cycle; A comprehensive fitness function is constructed based on trajectory tracking error, control quantity change, and actuator operation constraints. The learning gain of the iterative learning controller is used as the variable to be optimized in the black-backed jackal optimization algorithm, and the fitness corresponding to different learning gains is evaluated by the comprehensive fitness function to obtain the optimal learning gain corresponding to the k-th material feeding cycle. Substitute the optimal learning gain into the iterative learning control law to generate the control quantity for the (k+1)th material feeding cycle. The control quantity of the (k+1)th material handling cycle is limited and verified. If the control quantity meets the operating constraints of the actuator, the control quantity is converted into drive commands for the trolley traveling mechanism, boom slewing mechanism and boom pitching mechanism. After the (k+1)th material handling cycle is completed, the control quantity for the next material handling cycle is updated based on the actual trajectory data and operating status data of that cycle, so as to achieve inter-cycle adaptive correction of the material handling trajectory of the bucket wheel reclaimer.

2. The bucket wheel reclaimer control method based on the black-backed jackal optimization algorithm according to claim 1, characterized in that: The actual trajectory data is represented as follows: in, This represents the travel position of the trolley at the i-th sampling point in the k-th material handling cycle. For boom slewing angle, The boom pitch angle; The control quantity is expressed as: in, This is the command for the travel speed of the large vehicle. For boom slewing angular velocity command, This is the boom pitch rate command.

3. The bucket wheel reclaimer control method based on the black-backed jackal optimization algorithm according to claim 1, characterized in that: The comprehensive fitness function includes at least a trajectory error term, a control smoothing term, and a safety constraint term. The trajectory error term is used to characterize the deviation between the actual trajectory data and the expected trajectory data. The control smoothing term is used to characterize the magnitude of the change in control quantity between adjacent material handling cycles. The safety constraint term is used to characterize whether at least one of the bucket wheel motor current, material handling flow rate, and actuator control quantity exceeds the corresponding safety threshold. The overall fitness function satisfies: in, Let L be the overall fitness value corresponding to the learning gain to be optimized. For trajectory error term, To control the smoothing term, For safety constraint terms, λ is the weight coefficient of the control smoothing term, and μ is the weight coefficient of the safety constraint term; The trajectory error term is determined based on the trajectory tracking error of multiple sampling points within a material handling cycle. The control smoothing term is determined based on the change between the candidate control quantity of the (k+1)th material handling cycle and the control quantity of the kth material handling cycle. The safety constraint term is determined based on the exceedance of the bucket wheel motor current, material handling flow rate, and actuator control quantity relative to the corresponding safety threshold.

4. The bucket wheel reclaimer control method based on the black-backed jackal optimization algorithm according to claim 1, characterized in that: The learning gain includes a proportional learning gain and a differential learning gain corresponding to the trolley travel path, the boom slewing path, and the boom pitch path, respectively. The learning gain is expressed as: in, These represent the proportional learning gain and the differential learning gain of the vehicle's travel lane, respectively. These are the proportional learning gain and differential learning gain of the boom slewing channel, respectively. These are the proportional learning gain and differential learning gain of the boom pitch channel, respectively.

5. The bucket wheel reclaimer control method based on the black-backed jackal optimization algorithm according to claim 1, characterized in that: The optimization of the learning gain using the black-backed jackal optimization algorithm includes: Multiple black-backed jackals are generated within a preset learning gain search range, with each black-backed jackal representing a set of candidate learning gains. Calculate the fitness value of each individual black-backed jackal based on the comprehensive fitness function. Dominant individuals are determined based on their fitness values, and the dominant individuals include at least one black-backed jackal individual with the best current fitness value. Based on the positional relationship between the dominant individual and the other black-backed jackal individuals, update the candidate learning gain corresponding to each black-backed jackal individual; Boundary verification is performed on the updated candidate learning gains to ensure that the updated candidate learning gains remain within the preset learning gain search range; Repeat the fitness evaluation and candidate learning gain update until the preset number of iterations is reached or the fitness value meets the convergence condition, and output the candidate learning gain with the best fitness value as the optimal learning gain.

6. The bucket wheel reclaimer control method based on the black-backed jackal optimization algorithm according to claim 1, characterized in that: The iterative learning control law is a PD-type iterative learning control law. The control quantity of the (k+1)th material picking cycle is generated based on the control quantity of the kth material picking cycle, the trajectory tracking error, the change in the trajectory tracking error, and the optimal learning gain. The change in trajectory tracking error is determined by the trajectory tracking error of adjacent sampling points within the k-th material feeding cycle.

7. A bucket wheel reclaimer control method based on the black-backed jackal optimization algorithm according to claim 6, characterized in that: The PD-type iterative learning control law satisfies: in, This is the control quantity for the i-th sampling point in the (k+1)-th material feeding cycle. This is the control quantity for the i-th sampling point in the k-th material feeding cycle. To achieve the optimal ratio of learning gain, The optimal differential learning gain is given by Δt, where Δt is the sampling period. When the control quantity is a multi-channel control quantity These are the optimal proportional learning gain matrix and the optimal differential learning gain matrix corresponding to the trolley travel channel, boom slewing channel, and boom pitch channel, respectively.

8. The control method for a bucket wheel reclaimer based on the black-backed jackal optimization algorithm according to claim 1, characterized in that: The limit verification of the control quantity for the (k+1)th material feeding cycle includes: Determine whether the trolley travel speed command, boom slewing speed command, and boom pitch speed command are within the allowable control range of the corresponding actuator; Determine whether the change in control quantity between adjacent sampling points exceeds a preset change threshold. Determine whether the bucket wheel motor current and material handling flow rate are within the corresponding safety threshold range; When the control quantity of the (k+1)th material handling cycle does not meet the operating constraints of the actuator, or when the comprehensive fitness value calculated based on the control quantity is greater than the comprehensive fitness value corresponding to the kth material handling cycle, the control quantity is stopped, and the previous effective control quantity or the control quantity after attenuation is used to drive the bucket wheel reclaimer to execute the next material handling cycle. The attenuated control quantity satisfies: in, ρ is the original control quantity calculated by the iterative learning control law, and ρ is the attenuation coefficient that is greater than 0 and less than 1.

9. A bucket wheel reclaimer control system based on a black-backed jackal optimization algorithm, characterized in that, The system is used to execute the bucket wheel reclaimer control method based on the black-backed jackal optimization algorithm according to any one of claims 1 to 8. The system includes a sensor acquisition unit, a data processing controller, a desired trajectory storage unit, a black-backed jackal optimization controller, an iterative learning controller, a safety constraint controller, a PLC control unit, and an actuator drive unit. The signal output terminal of the sensor acquisition unit is connected to the signal input terminal of the data processing controller. The data output terminal of the desired trajectory storage unit is connected to the data input terminal of the data processing controller. The data output terminal of the data processing controller is connected to the data input terminals of both the black-backed jackal optimization controller and the iterative learning controller. The learning gain output terminal of the black-backed jackal optimization controller is connected to the learning gain input terminal of the iterative learning controller. The control output terminal of the iterative learning controller is connected to the control input terminal of the safety constraint controller. The control output terminal of the safety constraint controller is connected to the instruction input terminal of the PLC control unit. The control output terminal of the PLC control unit is connected to the control input terminal of the actuator drive unit. The actuator drive unit is connected to the trolley traveling mechanism, the boom slewing mechanism, and the boom pitching mechanism. The sensor acquisition unit includes a trolley position sensor, a boom slewing angle sensor, a boom pitch angle sensor, a bucket wheel motor current sensor, and a material handling flow sensor; the trolley position sensor, boom slewing angle sensor, boom pitch angle sensor, bucket wheel motor current sensor, and material handling flow sensor are respectively connected to the data processing controller; The data processing controller includes a sampling synchronization module, an error calculation module, and a fitness construction module. The input of the sampling synchronization module is connected to the sensor acquisition unit, and the output of the sampling synchronization module is connected to the input of the error calculation module. The error calculation module is also connected to the desired trajectory storage unit, and the output of the error calculation module is connected to the input of the fitness construction module. The output of the fitness construction module is connected to the Blackback Jackal optimization controller.

10. The bucket wheel reclaimer control system based on the black-backed jackal optimization algorithm according to claim 9, characterized in that: The safety constraint controller includes a control quantity limiting module, a previous valid control quantity storage module, and an attenuation processing module; the control quantity output terminal of the iterative learning controller is connected to the input terminals of the control quantity limiting module and the attenuation processing module, respectively; the control quantity limiting module is bidirectionally connected to the previous valid control quantity storage module; the output terminal of the control quantity limiting module is connected to the instruction input terminal of the PLC control unit. When the control quantity limiting module determines that the control quantity of the (k+1)th material feeding cycle exceeds the allowable range of the actuator, or the bucket wheel motor current or the material feeding flow exceeds the preset safety threshold, the control quantity limiting module calls the previous valid control quantity stored in the previous valid control quantity storage module, or receives the attenuation control quantity output by the attenuation processing module, and sends the previous valid control quantity or attenuation control quantity to the PLC control unit.

Images