A method and system for controlling the rotation speed of a rotary work equipment

Abstract

This invention relates to the field of rotary work equipment control, and discloses a method and system for controlling the rotation speed of rotary work equipment. It is based on real-time acquisition of the operating status, posture, and rotation speed of the rotary work equipment. The operating status includes the equipment's operating speed, vehicle tilt angle, boom vibration amplitude, and frequency; the posture includes the boom's position, attitude, and load. Multiple rotary work equipment parameters are used as fusion inputs, and weighting coefficients are set for each parameter in different terrains. A deep learning model is used to predict and output the current permissible rotation speed of the rotary work equipment. This permissible rotation speed is then used to control the rotary work process, maximizing operational efficiency and intelligence while ensuring safety.

Description

Technical Field

[0001] This invention relates to the field of rotary work equipment control, and in particular to a method and system for controlling the rotation speed of rotary work equipment. Background Technology

[0002] Rotary work equipment, such as engineering vehicles and excavators, relies heavily on its rotation function for efficient operation. It can achieve 360-degree unlimited rotation via hydraulic systems or electric motors. Currently, most rotary work equipment is operated by the driver controlling the rotation speed through a slewing joystick. This control method heavily depends on the driver's experience. However, when operating in complex terrain such as construction sites, forests, and gardens with numerous obstacles, due to slopes, uneven ground, or many obstacles, rotation is sometimes necessary while the boom is in motion or under load. Inappropriate control of the boom speed can lead to rollovers or collisions for inexperienced drivers.

[0003] Existing technical solutions for controlling the rotation speed of rotating work equipment are relatively simple. For example, the invention patent with publication number CN102071716A proposes a control system and method for the rotation speed of an excavator. It uses a pressure sensor to collect the pressure of the hydraulic oil in the main oil circuit and an engine speed sensor to collect the engine speed. The controller determines the working condition of the excavator based on the received pressure and speed signals, and outputs a corresponding proportional current according to the working condition. This causes the proportional hydraulic pressure regulating device to adjust the pressure of the pilot hydraulic oil leading to the valve core of the main swing valve in the excavator's hydraulic system according to the proportional current. This allows the valve core of the main swing valve to have different openings, thereby adjusting the rotation speed. However, this control method does not address the rotation speed control in complex terrain and also fails to solve the problem of rollover hazards caused by inexperienced drivers during operation.

[0004] In addition, since the rotating operation equipment rotates infinitely in 360 degrees, the operator's field of vision cannot cover the entire rotation range of the boom. This can lead to the inability to detect obstacles in time during the rotation, resulting in a collision hazard. If the rotation is controlled at low speed throughout the process, it will result in low operating efficiency and affect the progress of the operation.

[0005] Therefore, the key to improving the intelligence and even unmanned operation of rotary work equipment is to achieve automatic control of its rotation speed in complex terrain. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a method and system for controlling the rotation speed of rotating work equipment. This method is based on real-time acquisition of the operating status, posture, and rotation speed of the rotating work equipment. The operating status includes the equipment's operating speed, vehicle tilt angle, and boom vibration amplitude and frequency. The equipment posture includes the boom's position, orientation, and load. Multiple rotating work equipment parameters are used as fusion inputs, and weighting coefficients are set for each parameter in different terrains. A deep learning model is used to predict and output the current permissible rotation speed of the rotating work equipment. This permissible rotation speed is then used to control the rotating work process, maximizing operational efficiency and intelligence while ensuring safety.

[0007] This invention provides a method for controlling the rotation speed of a rotary work equipment, the method specifically including the following steps:

[0008] S1. Construct and train a deep learning model based on the pre-collected operating status, posture, and rotation speed of the rotating equipment;

[0009] S2. Based on the set sensor modules, collect the current operating status and current posture of the rotating work equipment, and use the deep learning model to output the current permissible rotation speed of the rotating work equipment;

[0010] S3. The controller parses the target position from the task and calculates precise hydraulic valve or motor control commands based on the current permissible rotation speed output using a PID algorithm, thereby controlling the rotating equipment to complete the rotating task.

[0011] Preferably, in step S1, constructing and training a deep learning model based on the pre-collected operating status, posture, and rotation speed of the rotating equipment specifically includes:

[0012] The operating status of the rotating work equipment includes the equipment operating speed, vehicle tilt angle, boom vibration amplitude and frequency;

[0013] The equipment attitude includes the position, attitude, and load of the boom;

[0014] Based on the pre-collected operating status and attitude parameters of the rotating equipment, a training parameter set is constructed according to the operation time series. Combined with the upper limit of the allowable rotation speed of the rotating equipment itself, an initial deep learning model is established using a model predictive control algorithm. The output of the initial deep learning model is the allowable rotation speed corresponding to each operation time series training parameter set of the rotating equipment.

[0015] The actual rotation speed corresponding to the pre-collected time series training parameter sets of each operation is compared with the allowable rotation speed output by the initial deep learning model. The initial deep learning model is optimized according to the comparison result until the error between the two comparison results is within the expected range. The training process is then stopped, and the trained deep learning model that meets the requirements is deployed in the controller.

[0016] Preferably, the training parameter set is in the following form:

[0017]

[0018] The permissible rotation speed is ;

[0019] The The training parameter set is constructed for the parameters of the rotating equipment collected at time i; For the equipment operating speed, the The weighted value for equipment operating speed; For the vehicle body tilt angle, the stated The vehicle body tilt angle is the weighted value; A is the boom vibration amplitude. Here, f is the weighted value for the boom vibration amplitude, and f is the boom vibration frequency. The weighted value for the boom vibration frequency; S is the boom position; J is the boom attitude. The weight value for the boom attitude; L is the load capacity of the boom, and the... This is the weighted value for the boom load;

[0020] The Set an upper limit on the permissible rotational speed of the rotating work equipment; the The output value for the deep learning model is in percentage form, which is the percentage value that sets the upper limit of the allowable rotation speed of the rotating equipment itself.

[0021] Preferably, step S2 specifically includes:

[0022] The rotating work equipment is equipped with a sensor module to collect data in real time on the current operating speed, tilt angle, boom vibration amplitude and frequency, as well as the boom position, attitude and load.

[0023] The collected real-time parameters of the rotating equipment are used to construct a current parameter set according to the training parameter set format. Then, the current parameter set is input into the trained deep learning model, and the deep learning model outputs the current allowable rotation speed of the rotating equipment, that is, the maximum allowable rotation speed in the current state.

[0024] Preferably, step S3 specifically includes:

[0025] The controller parses the target position from the current task, calculates the difference between the current position of the boom and the target position, and calculates the control quantity of the hydraulic valve or motor based on the current allowable rotation speed output by the PID algorithm. The control quantity is then converted into a corresponding control command and sent to the hydraulic valve or motor controller to complete the current task.

[0026] Preferably, step S3 further includes:

[0027] High-definition cameras are also installed around the boom and body of the rotating work equipment to capture high-definition images of the boom's rotation range in real time. The controller processes the captured high-definition images to detect whether there are obstacles within the boom's rotation range. If there are, the speed and direction of the obstacle are extracted. Based on the boom's rotation direction and speed, the permissible rotation speed of the rotating work equipment when there are obstacles is recalculated. Based on the recalculated permissible rotation speed, precise hydraulic valve or motor control commands are calculated to control the rotating work equipment to complete the rotating work task.

[0028] Preferably, the recalculation of the permissible rotational speed of the rotating work equipment when there are obstacles further includes:

[0029] If there are obstacles, extract the speed and direction of the obstacles:

[0030] If the obstacle is a static obstacle, the position difference between the current position of the boom and the static obstacle is calculated. If the position difference is greater than the first preset threshold, the rotation operation is still performed at the current allowable rotation speed output by the deep learning model until the position difference is equal to the first preset threshold.

[0031] If the position difference is less than the first preset threshold and greater than the second preset threshold, then the rotation operation is performed at 50% of the current allowable rotation speed until the position difference is equal to the second preset threshold.

[0032] If the position difference is less than the second preset threshold, deceleration control is performed based on the position error.

[0033] Preferably, if there is an obstacle, extracting the obstacle's speed and direction further includes:

[0034] If the obstacle is dynamic, extract its speed and direction.

[0035] If the obstacle moves in a direction opposite to the boom's direction of movement, the rotation operation will continue at the current permissible rotation speed output by the deep learning model, while continuously monitoring whether the dynamic obstacle is moving away from the rotating equipment. If it is not moving away, the position difference between the current position of the boom and the dynamic obstacle will be continuously monitored. If the position difference is greater than the third preset threshold, the rotation operation will continue at the current permissible rotation speed output by the deep learning model until the position difference equals the third preset threshold. If the position difference is less than the third preset threshold but greater than the fourth preset threshold, the rotation operation will continue at 30% of the current permissible rotation speed until the position difference equals the fourth preset threshold. If the position difference is less than the fourth preset threshold, the boom will be stopped rotating, and the rotating equipment will be controlled to move to avoid the dynamic obstacle.

[0036] If the obstacle's movement direction is opposite to the boom's movement direction, the position difference between the current boom position and the dynamic obstacle is calculated. If the position difference is greater than the third preset threshold, the rotation operation is still performed at the current allowable rotation speed output by the deep learning model until the position difference equals the third preset threshold. If the position difference is less than the third preset threshold, the boom is controlled to stop rotating, and the rotating operation equipment is controlled to move to avoid the dynamic obstacle.

[0037] Corresponding to the aforementioned method for controlling the rotation speed of a rotating work equipment, the present invention also provides a control system for the rotation speed of a rotating work equipment, the system comprising a building module, a sensor module, a controller, and an execution module;

[0038] The building module constructs and trains a deep learning model based on the pre-collected operating status, posture, and rotation speed of the rotating equipment, and deploys it within the controller.

[0039] The sensor module is used to collect the current operating status and current posture of the rotating equipment, and transmit the collected data to the deep learning model in the controller;

[0040] The controller uses the deep learning model to output the current permissible rotation speed of the rotating work equipment; it parses the target position from the work task, and calculates precise hydraulic valve or motor control commands based on the output current permissible rotation speed using a PID algorithm.

[0041] The execution module includes a hydraulic valve or a motor, which controls the rotating work equipment to complete the rotating work task according to the received control command.

[0042] Preferably, the construction and training of the deep learning model specifically includes:

[0043] The operating status of the rotating work equipment includes the equipment operating speed, vehicle tilt angle, boom vibration amplitude and frequency;

[0044] The equipment attitude includes the position, attitude, and load of the boom;

[0045] Based on the pre-collected operating status and attitude parameters of the rotating equipment, a training parameter set is constructed according to the operation time series. Combined with the upper limit of the allowable rotation speed of the rotating equipment itself, an initial deep learning model is established using a model predictive control algorithm. The output of the initial deep learning model is the allowable rotation speed corresponding to each operation time series training parameter set of the rotating equipment.

[0046] The actual rotation speed corresponding to the pre-collected time series training parameter sets of each operation is compared with the allowable rotation speed output by the initial deep learning model. The initial deep learning model is optimized according to the comparison result until the error between the two comparison results is within the expected range. The training process is then stopped, and the trained deep learning model that meets the requirements is deployed in the controller.

[0047] The beneficial effects of this invention are as follows: 1. Based on the operating status, posture, and rotation speed of the rotating work equipment, the operating status includes the equipment operating speed, vehicle body tilt angle, boom vibration amplitude and frequency; the posture includes the boom position, posture, and load; multiple rotating work equipment parameters are used as fusion inputs, and combined with the weighting coefficients set for each parameter in different terrains, a deep learning model is used to predict and output the current permissible rotation speed of the rotating work equipment. The upper limit of the safe rotation speed can be calculated in real time, and rapid rotation that may cause rollover can be actively suppressed in dangerous working conditions such as slopes; combined with the predictive control of vehicle body tilt angle and rotation, the rotation speed can be actively limited before danger or severe vibration occurs, protecting the key components of the rotating work equipment;

[0048] 2. The upper limit of rotation speed can be automatically predicted and matched according to the overall status of the rotation operation, which can reduce cycle time, improve work efficiency, automatically reproduce expert-level operation, reduce dependence on driver's operating experience and physical strength, and reduce losses caused by improper human operation.

[0049] 3. Based on the automatic control of the rotation speed of the rotating work equipment, collision detection is also implemented to predict the collision risk in advance, and further intervention is made on the rotation speed to maximize the improvement of work efficiency while ensuring safety.

[0050] 4. Based on the rotation speed control method of rotating work equipment, intelligent automatic driving can be realized, enabling rotating work equipment to operate safely and continuously in dangerous conditions, such as gardens, forests, landslides, mines and other high-risk areas where personnel are not suitable to enter, thereby improving the application scenarios of rotating work equipment. Attached Figure Description

[0051] Figure 1 A flowchart illustrating the steps of a method for controlling the rotational speed of rotating work equipment;

[0052] Figure 2 This is a structural diagram of the rotation speed control system for a rotating work equipment. Detailed Implementation

[0053] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0054] This invention provides a method for controlling the rotation speed of a rotary work equipment, such as... Figure 1 As shown, the method specifically includes the following steps:

[0055] S1. Construct and train a deep learning model based on the pre-collected operating status, posture, and rotation speed of the rotating equipment;

[0056] S2. Based on the set sensor modules, collect the current operating status and current posture of the rotating work equipment, and use the deep learning model to output the current permissible rotation speed of the rotating work equipment;

[0057] S3. The controller parses the target position from the task and calculates precise hydraulic valve or motor control commands based on the current permissible rotation speed output using a PID algorithm, thereby controlling the rotating equipment to complete the rotating task.

[0058] Preferably, in step S1, constructing and training a deep learning model based on the pre-collected operating status, posture, and rotation speed of the rotating equipment specifically includes:

[0059] The pre-collected operating status, posture, and rotation speed of the rotating equipment can be historical operating data of the rotating equipment stored in the database;

[0060] The operating status of the rotating work equipment includes the equipment operating speed, vehicle tilt angle, boom vibration amplitude and frequency;

[0061] The equipment attitude includes the position, attitude, and load of the boom;

[0062] A training parameter set is constructed based on the pre-collected operating status and attitude parameters of the rotating equipment according to the operation time series; the operation time series refers to the data collection or storage time of the rotating equipment's operating data, and the data in the training parameter set is the operating data obtained at the corresponding collection or storage time; combined with the upper limit of the allowable rotation speed of the rotating equipment itself, an initial deep learning model is established using a model predictive control algorithm, and the output of the initial deep learning model is the allowable rotation speed corresponding to each operation time series training parameter set of the rotating equipment;

[0063] The actual rotation speed corresponding to each pre-collected operation time series training parameter set is compared with the allowable rotation speed output by the initial deep learning model. The initial deep learning model is optimized based on the comparison result until the error between the two comparison results is within the expected range. The training process is then stopped, and the trained deep learning model that meets the requirements is deployed in the controller. The actual rotation speed corresponding to each operation time series training parameter set is the actual rotation speed obtained at the corresponding collection or storage time of the data in the training parameter set.

[0064] Specifically, the initial deep learning model established using the model predictive control algorithm describes the mapping relationship between the parameter set and the allowable rotation speed. The allowable rotation speed is the maximum allowable rotation speed of the boom corresponding to the parameter state in the current parameter set, and the maximum allowable rotation speed does not exceed the upper limit of the allowable rotation speed set by the rotating work equipment itself.

[0065] Preferably, the training parameter set is in the following form:

[0066]

[0067] The permissible rotation speed is ;

[0068] The The training parameter set is constructed for the parameters of the rotating equipment collected at time i; For the equipment operating speed, the The weighted value for equipment operating speed; For the vehicle body tilt angle, the stated The vehicle body tilt angle is the weighted value; A is the boom vibration amplitude. Here, f is the weighted value for the boom vibration amplitude, and f is the boom vibration frequency. The weighted value for the boom vibration frequency; S is the boom position; J is the boom attitude. The weight value for the boom attitude; L is the load capacity of the boom, and the... This is the weighted value for the boom load;

[0069] The Set an upper limit on the permissible rotational speed of the rotating work equipment; the The output value for the deep learning model is in percentage form, which is the percentage value that sets the upper limit of the allowable rotation speed of the rotating equipment itself.

[0070] Specifically, the mapping relationship between the parameter set and the permissible rotational speed can be set as follows: the weight value of the device operating speed. Different weight values ​​are applied based on preset first, second, and third speed thresholds, where the first speed threshold < the second speed threshold < the third speed threshold. When When the speed is less than the first speed threshold, set to , When it is between the first speed threshold and the second speed threshold, set to , When it is between the second speed threshold and the third speed threshold, set to , When the speed is greater than the third speed threshold, set to , ;when A higher value indicates a higher vehicle speed and a higher probability of rollover caused by that speed. This corresponds to the output value of the deep learning model. It should be relatively smaller; for example, when When the speed is less than the first speed threshold, The value is 90%; When the speed is between the first speed threshold and the second speed threshold, The value is 70%; When the speed is between the second and third speed thresholds, The value is 50%; When the speed is greater than the third speed threshold, The value is 30%.

[0071] The vehicle body tilt angle weight value The weighted value of the boom vibration amplitude The weighted value of the boom vibration frequency Similarly, at least three threshold ranges should be set, and the setting method should be related to the weight value of the device operating speed. The setup method is the same;

[0072] Specifically, the vehicle body tilt angle Different weight values ​​are applied based on preset first, second, and third tilt angle thresholds, where the first tilt angle threshold < the second tilt angle threshold < the third tilt angle threshold. When When less than the first tilt angle threshold, set to , When it is between the first tilt angle threshold and the second tilt angle threshold, set to , When it is between the second and third tilt angle thresholds, set to , When the tilt angle is greater than the third tilt angle threshold, set to ,in ;when The larger the value, the greater the tilt angle, and the higher the probability of the rotating equipment tipping over due to the tilt angle. The output value of the deep learning model will be higher in this case. The smaller it should be;

[0073] For example, when When the tilt angle is less than the first tilt angle threshold, The value is 80%; when Located between the first tilt angle threshold and the second tilt angle threshold, The value is 60%; when When it is between the second and third tilt angle thresholds, The value is 40%; when When the tilt angle is greater than the third tilt angle threshold, The value is 20%; when operating in mountainous or sloping environments, the greater the vehicle tilt angle, the more prone the boom is to tipping over during rotation. Different limits are applied to the rotation speed based on the vehicle tilt angle, until the tilt angle exceeds a certain angle, such as 75 degrees, at which point the deep learning model outputs a value. When the value is 0, rotation is strictly prohibited.

[0074] Specifically, since the input parameters of a deep learning model include multiple parameters, the output value of the deep learning model... The output value is the result of comprehensive mapping and processing of multiple parameters. For example, if the input only contains the equipment operating speed and vehicle tilt angle: when Located between the first tilt angle threshold and the second tilt angle threshold, When the speed is between the first speed threshold and the second speed threshold, The value is 55%; When the speed is between the second and third speed thresholds, The value is 50%;

[0075] when When it is between the second and third tilt angle thresholds, When the speed is between the first speed threshold and the second speed threshold, The value is 35%; When the speed is between the second and third speed thresholds, The value is 30%;

[0076] The above examples are only used to explain the principle of the mapping relationship of the deep learning model in this application, and are not exhaustive examples. The comprehensive mapping processing of multiple parameters such as equipment operating speed, vehicle body tilt angle, boom vibration amplitude and frequency, as well as boom position, attitude and load is similar.

[0077] The weight value of the boom posture and the weight value of the boom load Reference equipment operating speed weight value The setting method requires at least two threshold ranges;

[0078] Specifically, the more threshold values ​​set within a reasonable range, the more refined the control precision. However, considering the different processor capabilities and actual control precision requirements, the threshold value range can be selected according to the actual operating scenario.

[0079] Preferably, step S2 specifically includes:

[0080] The rotating work equipment is equipped with a sensor module to collect data in real time on the current operating speed, tilt angle, boom vibration amplitude and frequency, as well as the boom position, attitude and load.

[0081] The collected real-time parameters of the rotating equipment are used to construct a current parameter set according to the training parameter set format. Then, the current parameter set is input into the trained deep learning model, and the deep learning model outputs the current allowable rotation speed of the rotating equipment, that is, the maximum allowable rotation speed in the current state.

[0082] Preferably, step S3 specifically includes:

[0083] The controller parses the target position from the current task, calculates the difference between the current position of the boom and the target position, and calculates the control quantity of the hydraulic valve or motor based on the current allowable rotation speed output by the PID algorithm. The control quantity is then converted into a corresponding control command and sent to the hydraulic valve or motor controller to complete the current task.

[0084] Preferably, step S3 further includes:

[0085] High-definition cameras are also installed around the boom and body of the rotating work equipment to capture high-definition images of the boom's rotation range in real time. The controller processes the captured high-definition images to detect whether there are obstacles within the boom's rotation range. If there are, the speed and direction of the obstacle are extracted. Based on the boom's rotation direction and speed, the permissible rotation speed of the rotating work equipment when there are obstacles is recalculated. Based on the recalculated permissible rotation speed, precise hydraulic valve or motor control commands are calculated to control the rotating work equipment to complete the rotating work task.

[0086] Preferably, the recalculation of the permissible rotational speed of the rotating work equipment when there are obstacles further includes:

[0087] If there are obstacles, extract the speed and direction of the obstacles:

[0088] If the obstacle is a static obstacle, the position difference between the current position of the boom and the static obstacle is calculated. If the position difference is greater than the first preset threshold, the rotation operation is still performed at the current allowable rotation speed output by the deep learning model until the position difference is equal to the first preset threshold.

[0089] If the position difference is less than the first preset threshold and greater than the second preset threshold, then the rotation operation is performed at 50% of the current allowable rotation speed until the position difference is equal to the second preset threshold.

[0090] If the position difference is less than the second preset threshold, deceleration control is performed based on the position error.

[0091] Preferably, if there is an obstacle, extracting the obstacle's speed and direction further includes:

[0092] If the obstacle is dynamic, extract its speed and direction.

[0093] If the obstacle moves in a direction opposite to the boom's direction of movement, the rotation operation will continue at the current permissible rotation speed output by the deep learning model, while continuously monitoring whether the dynamic obstacle is moving away from the rotating equipment. If it is not moving away, the position difference between the current position of the boom and the dynamic obstacle will be continuously monitored. If the position difference is greater than the third preset threshold, the rotation operation will continue at the current permissible rotation speed output by the deep learning model until the position difference equals the third preset threshold. If the position difference is less than the third preset threshold but greater than the fourth preset threshold, the rotation operation will continue at 30% of the current permissible rotation speed until the position difference equals the fourth preset threshold. If the position difference is less than the fourth preset threshold, the boom will be stopped rotating, and the rotating equipment will be controlled to move to avoid the dynamic obstacle.

[0094] If the obstacle's movement direction is opposite to the boom's movement direction, the position difference between the current boom position and the dynamic obstacle is calculated. If the position difference is greater than the third preset threshold, the rotation operation is still performed at the current allowable rotation speed output by the deep learning model until the position difference equals the third preset threshold. If the position difference is less than the third preset threshold, the boom is controlled to stop rotating, and the rotating operation equipment is controlled to move to avoid the dynamic obstacle.

[0095] Corresponding to the aforementioned method for controlling the rotational speed of a rotating work equipment, the present invention also provides a control system for the rotational speed of a rotating work equipment, such as... Figure 2 As shown, the system includes a building module, a sensor module, a controller, and an execution module;

[0096] The building module constructs and trains a deep learning model based on the pre-collected operating status, posture, and rotation speed of the rotating equipment, and deploys it within the controller.

[0097] The sensor module is used to collect the current operating status and current posture of the rotating equipment, and transmit the collected data to the deep learning model in the controller;

[0098] The controller uses the deep learning model to output the current permissible rotation speed of the rotating work equipment; it parses the target position from the work task, and calculates precise hydraulic valve or motor control commands based on the output current permissible rotation speed using a PID algorithm.

[0099] The execution module includes a hydraulic valve or a motor, which controls the rotating work equipment to complete the rotating work task according to the received control command.

[0100] Preferably, the construction and training of the deep learning model specifically includes:

[0101] The operating status of the rotating work equipment includes the equipment operating speed, vehicle tilt angle, boom vibration amplitude and frequency;

[0102] The equipment attitude includes the position, attitude, and load of the boom;

[0103] Based on the pre-collected operating status and attitude parameters of the rotating equipment, a training parameter set is constructed according to the operation time series. Combined with the upper limit of the allowable rotation speed of the rotating equipment itself, an initial deep learning model is established using a model predictive control algorithm. The output of the initial deep learning model is the allowable rotation speed corresponding to each operation time series training parameter set of the rotating equipment.

[0104] The actual rotation speed corresponding to the pre-collected time series training parameter sets of each operation is compared with the allowable rotation speed output by the initial deep learning model. The initial deep learning model is optimized according to the comparison result until the error between the two comparison results is within the expected range. The training process is then stopped, and the trained deep learning model that meets the requirements is deployed in the controller.

[0105] Specifically, the operating status, posture, and rotation speed of the rotating equipment, which are pre-collected by sensors, are stored in a database.

[0106] Specifically, based on the same principle as the method shown in the embodiments of the present invention, the embodiments of the present invention also provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the rotation speed control method for the rotary work equipment.

[0107] Specifically, embodiments of the present invention also include a computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of the rotation speed control method for the rotary work equipment.

[0108] The processor can be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), a FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. The processor can also be a combination that implements computational functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0109] The memory can be ROM (Read-Only Memory) or other types of static storage devices that can store static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or EEPROM (Electrically Erasable Programmable Read-Only Memory), CD-ROM (Compact Disc Read-Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited thereto.

[0110] The memory stores application code (computer program) that executes the present invention, and its execution is controlled by a processor. The processor executes the application code stored in the memory to implement the content shown in the foregoing method embodiments.

[0111] The above description is merely a preferred embodiment of the present invention and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of disclosure involved in this invention is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept.

Claims

1. A method for controlling the rotation speed of a rotary work equipment, characterized in that: The method specifically includes the following steps: S1. Construct and train a deep learning model based on pre-collected data on the operating status, posture, and rotation speed of the rotating equipment; specifically including: The operating status of the rotating work equipment includes the equipment operating speed, vehicle tilt angle, boom vibration amplitude and frequency; The equipment attitude includes the position, attitude, and load of the boom; Based on the pre-collected operating status and attitude parameters of the rotating equipment, a training parameter set is constructed according to the operation time series. Combined with the upper limit of the allowable rotation speed of the rotating equipment itself, an initial deep learning model is established using a model predictive control algorithm. The output of the initial deep learning model is the allowable rotation speed corresponding to each operation time series training parameter set of the rotating equipment. The actual rotation speed corresponding to the pre-collected operation time series training parameter sets is compared with the allowable rotation speed output by the initial deep learning model. The initial deep learning model is optimized according to the comparison result until the error between the two comparison results is within the expected range. The training process is stopped, and the trained deep learning model that meets the requirements is deployed in the controller. The specific form of the training parameter set is as follows: The permissible rotation speed is ; The The training parameter set is constructed for the parameters of the rotating equipment collected at time i; For the equipment operating speed, the The weighted value for equipment operating speed; For the vehicle body tilt angle, the stated The vehicle body tilt angle is the weighted value; A is the boom vibration amplitude. Here, f is the weighted value for the boom vibration amplitude, and f is the boom vibration frequency. The weighted value for the boom vibration frequency; S is the boom position; J is the boom attitude. The weight value for the boom attitude; L is the load capacity of the boom, and the... This is the weight value for the boom load. The Set an upper limit on the permissible rotational speed of the rotating work equipment; the Output values ​​for deep learning models; S2. Based on the set sensor modules, collect the current operating status and current posture of the rotating work equipment, and use the deep learning model to output the current permissible rotation speed of the rotating work equipment; S3. The controller parses the target position from the task and calculates precise hydraulic valve or motor control commands based on the current permissible rotation speed output using a PID algorithm, thereby controlling the rotating equipment to complete the rotating task.

2. The rotation speed control method for rotary work equipment as described in claim 1, characterized in that: Step S2 specifically includes: The rotating work equipment is equipped with a sensor module to collect data in real time on the current operating speed, tilt angle, boom vibration amplitude and frequency, as well as the boom position, attitude and load. The collected real-time parameters of the rotating equipment are used to construct a current parameter set according to the training parameter set format. Then, the current parameter set is input into the trained deep learning model, and the deep learning model outputs the current allowable rotation speed of the rotating equipment, that is, the maximum allowable rotation speed in the current state.

3. The rotation speed control method for rotary work equipment as described in claim 2, characterized in that: Step S3 specifically includes: The controller parses the target position from the current task, calculates the difference between the current position of the boom and the target position, and calculates the control quantity of the hydraulic valve or motor based on the current allowable rotation speed output by the PID algorithm. The control quantity is then converted into a corresponding control command and sent to the hydraulic valve or motor controller to complete the current task.

4. The rotation speed control method for rotary work equipment as described in claim 3, characterized in that: Step S3 specifically also includes: High-definition cameras are also installed around the boom and body of the rotating work equipment to capture high-definition images of the boom's rotation range in real time. The controller processes the captured high-definition images to detect whether there are obstacles within the boom's rotation range. If there are, the speed and direction of the obstacle are extracted. Based on the boom's rotation direction and speed, the permissible rotation speed of the rotating work equipment when there are obstacles is recalculated. Based on the recalculated permissible rotation speed, precise hydraulic valve or motor control commands are calculated to control the rotating work equipment to complete the rotating operation task.

5. The rotation speed control method for rotary work equipment as described in claim 4, characterized in that: The recalculation of the permissible rotation speed of the rotating work equipment when there are obstacles specifically includes: If there are obstacles, extract the speed and direction of the obstacles: If the obstacle is a static obstacle, the position difference between the current position of the boom and the static obstacle is calculated. If the position difference is greater than the first preset threshold, the rotation operation is still performed at the current allowable rotation speed output by the deep learning model until the position difference is equal to the first preset threshold. If the position difference is less than the first preset threshold and greater than the second preset threshold, then the rotation operation is performed at 50% of the current allowable rotation speed until the position difference is equal to the second preset threshold. If the position difference is less than the second preset threshold, deceleration control is performed based on the position error.

6. The rotation speed control method for a rotary work equipment as described in claim 5, characterized in that: If there are obstacles, the method of extracting the speed and direction of the obstacles also includes: If the obstacle is dynamic, extract its speed and direction. If the obstacle moves in a direction opposite to the boom's direction of movement, the rotation operation will continue at the current permissible rotation speed output by the deep learning model, while continuously monitoring whether the dynamic obstacle is moving away from the rotating equipment. If it is not moving away, the position difference between the current position of the boom and the dynamic obstacle will be continuously monitored. If the position difference is greater than the third preset threshold, the rotation operation will continue at the current permissible rotation speed output by the deep learning model until the position difference equals the third preset threshold. If the position difference is less than the third preset threshold but greater than the fourth preset threshold, the rotation operation will continue at 30% of the current permissible rotation speed until the position difference equals the fourth preset threshold. If the position difference is less than the fourth preset threshold, the boom will be stopped rotating, and the rotating equipment will be controlled to move to avoid the dynamic obstacle. If the obstacle's movement direction is opposite to the boom's movement direction, the position difference between the current boom position and the dynamic obstacle is calculated. If the position difference is greater than the third preset threshold, the rotation operation is still performed at the current allowable rotation speed output by the deep learning model until the position difference equals the third preset threshold. If the position difference is less than the third preset threshold, the boom is controlled to stop rotating, and the rotating operation equipment is controlled to move to avoid the dynamic obstacle.

7. A rotation speed control system for a rotary work equipment, characterized in that: The system includes a construction module, a sensor module, a controller, and an execution module; The building module constructs and trains a deep learning model based on the pre-collected operating status, posture, and rotation speed of the rotating equipment, and deploys it within the controller. The construction and training of the deep learning model specifically includes: The operating status of the rotating work equipment includes the equipment operating speed, vehicle tilt angle, boom vibration amplitude and frequency; The equipment attitude includes the position, attitude, and load of the boom; Based on the pre-collected operating status and attitude parameters of the rotating equipment, a training parameter set is constructed according to the operation time series. Combined with the upper limit of the allowable rotation speed of the rotating equipment itself, an initial deep learning model is established using a model predictive control algorithm. The output of the initial deep learning model is the allowable rotation speed corresponding to each operation time series training parameter set of the rotating equipment. The actual rotation speed corresponding to the pre-collected operation time series training parameter sets is compared with the allowable rotation speed output by the initial deep learning model. The initial deep learning model is optimized according to the comparison result until the error between the two comparison results is within the expected range. The training process is stopped, and the trained deep learning model that meets the requirements is deployed in the controller. The specific form of the training parameter set is as follows: The permissible rotation speed is ; The The training parameter set is constructed for the parameters of the rotating equipment collected at time i; For the equipment operating speed, the The weighted value for equipment operating speed; For the vehicle body tilt angle, the stated The vehicle body tilt angle is the weighted value; A is the boom vibration amplitude. Here, f is the weighted value for the boom vibration amplitude, and f is the boom vibration frequency. The weighted value for the boom vibration frequency; S is the boom position; J is the boom attitude. The weight value for the boom attitude; L is the load capacity of the boom, and the... This is the weight value for the boom load. The Set an upper limit on the permissible rotational speed of the rotating work equipment; the Output values ​​for deep learning models; The sensor module is used to collect the current operating status and current posture of the rotating equipment, and transmit the collected data to the deep learning model in the controller; The controller uses the deep learning model to output the current permissible rotation speed of the rotating work equipment; it parses the target position from the work task, and calculates precise hydraulic valve or motor control commands based on the output current permissible rotation speed using a PID algorithm. The execution module includes a hydraulic valve or a motor, which controls the rotating work equipment to complete the rotating work task according to the received control command.

Images