CN115465799A

CN115465799A - Tower crane control parameter adjusting method and device, calculating equipment and storage medium

Info

Publication number: CN115465799A
Application number: CN202211294957.2A
Authority: CN
Inventors: 齐斌
Original assignee: Kyland Technology Co Ltd
Current assignee: Kyland Technology Co Ltd
Priority date: 2022-10-21
Filing date: 2022-10-21
Publication date: 2022-12-13

Abstract

The embodiment of the application relates to the field of intelligent manufacturing, and relates to a tower crane control parameter adjusting method and device, computing equipment and a storage medium. The method comprises the following specific scheme: predicting PID parameters by using a neural network model based on the position deviation between the track planning point position and the actual position of the tower crane hook to obtain a predicted value of the PID parameters; controlling the operation of the lifting hook based on the prediction value of the PID parameter, and calculating the fitting degree of the actual path and the planned path of the operation of the lifting hook; adjusting the learning rate and inertia parameters of the neural network model according to the fitting degree; and predicting the PID parameters by using the neural network model based on the adjusted learning rate and inertia parameters to obtain the PID parameters for controlling the operation of the lifting hook. According to the embodiment of the application, the PID parameters are accurately predicted by using the neural network model, the control parameters can be continuously learned and corrected in the tower crane operation process, the self-adaptive adjustment of the tower crane control parameters is realized, and the tower crane operation control precision is improved.

Description

Tower crane control parameter adjusting method and device, computing equipment and storage medium

Technical Field

The invention relates to the technical field of intelligent manufacturing, in particular to a tower crane control parameter adjusting method and device, computing equipment and a storage medium.

Background

In a current motion control system of the unmanned tower crane, a high-precision index requires that a lifting hook moves according to a planned path, and the deviation between the path of the lifting hook and the planned path needs to be controlled within a minimum range in the motion process. However, the mechanical structure and transmission mechanism of the tower crane determine the motion equation and power equation of the tower crane, and the accurate establishment is difficult. Therefore, a common method is to skip model building of the tower crane, directly analyze the output value and the current value of each motion axis by using a PID (proportional integral differential) in the motion process, and dynamically compensate the output value. However, the performance of the tower cranes of various models has deviation, and the transmission devices are different, so that the PID parameters are difficult to adjust in practical situations. The tuning process of the current PID parameters is usually tried and tried by the experience of the field engineer. The trial and error process is time-consuming, and the accuracy of PID parameters obtained by trial and error is not enough, so that the control precision of tower crane operation cannot be ensured.

Disclosure of Invention

In view of the above problems in the prior art, the embodiments of the present application provide a tower crane control parameter adjusting method and apparatus, a computing device, and a storage medium, which utilize a neural network model to accurately predict PID parameters, so that the control parameters can be continuously learned and corrected during the operation of the tower crane, thereby implementing adaptive adjustment of the tower crane control parameters and improving the control accuracy of the tower crane operation.

The above object is achieved, and in a first aspect of the present application, a method for adjusting tower crane control parameters is provided, including:

predicting the PID parameters by utilizing a neural network model based on the position deviation between the track planning point position and the actual position of the tower crane hook to obtain the predicted value of the PID parameters;

controlling the operation of the lifting hook based on the predicted value of the PID parameter; calculating the fitting degree of the actual path and the planned path of the operation of the lifting hook;

adjusting the learning rate and the inertia parameters of the neural network model according to the fitting degree;

and predicting the PID parameters by using the neural network model based on the adjusted learning rate and inertia parameters to obtain the PID parameters for controlling the operation of the lifting hook.

As a possible implementation manner of the first aspect, predicting PID parameters by using the neural network model includes:

acquiring a track planning point position and an actual position of a tower crane hook; calculating the position deviation between the position of the track planning point and the actual position;

and inputting the position of the track planning point, the actual position and the position deviation into the neural network model, and predicting PID parameters by using the neural network model.

As a possible implementation manner of the first aspect, the method further includes:

and collecting the actual position corresponding to the position of the track planning point of the lifting hook by using an encoder.

As a possible implementation manner of the first aspect, controlling the hook operation based on the predicted value of the PID parameter includes:

based on the predicted value of the PID parameter, a PID control module is utilized to obtain the correction quantity of the target speed of the current track point;

and taking the sum of the correction amount and the target speed as the command speed of the current track point, and controlling the operation of the lifting hook by using the command speed.

As a possible implementation manner of the first aspect, calculating a fitting degree of an actual path traveled by the hook and a planned path includes:

calculating the residual square sum and the total square sum of the actual path and the planned path;

and obtaining the fitting degree of the actual path of the lifting hook operation and the planned path according to the residual square sum and the total square sum.

and training the neural network model based on the set initial value of the learning rate and the initial value of the inertia parameter, and determining the weight coefficient of the neural network model.

As a possible implementation manner of the first aspect, adjusting the learning rate and the inertia parameter of the neural network model according to the fitting degree includes:

comparing the fitting degree with a preset fitting threshold;

under the condition that the fitting degree is smaller than the preset fitting threshold, adjusting the learning rate and the inertia parameters of the neural network model to enable the fitting degree obtained again on the basis of the adjusted neural network model to be larger than or equal to the preset fitting threshold;

wherein the process of reacquiring the fitness based on the adjusted neural network model comprises: based on the adjusted learning rate and inertia parameters, the neural network model is utilized to predict the PID parameters again; controlling the operation of the lifting hook based on the PID parameters obtained by the re-prediction; and recalculating the fitting degree of the actual path operated by the lifting hook and the planned path.

The application second aspect provides a tower crane control parameter adjusting device, includes:

a prediction unit to: predicting the PID parameters by utilizing a neural network model based on the position deviation between the track planning point position and the actual position of the tower crane hook to obtain the predicted value of the PID parameters;

a control unit for: controlling the operation of the lifting hook based on the predicted value of the PID parameter; calculating the fitting degree of the actual path and the planned path of the operation of the lifting hook;

an adjustment unit for: adjusting the learning rate and inertia parameters of the neural network model according to the fitting degree;

the prediction unit is further to: and predicting the PID parameters by using the neural network model based on the adjusted learning rate and inertia parameters to obtain the PID parameters for controlling the operation of the lifting hook.

As a possible implementation manner of the second aspect, the prediction unit includes:

the acquiring subunit is used for acquiring a track planning point position and an actual position of the tower crane hook and calculating a position deviation between the track planning point position and the actual position;

and the predicting subunit is used for inputting the position of the track planning point, the actual position and the position deviation into the neural network model and predicting the PID parameters by using the neural network model.

As a possible implementation manner of the second aspect, the obtaining subunit is configured to:

As a possible implementation manner of the second aspect, the control unit is configured to:

and taking the sum of the correction quantity and the target speed as the command speed of the current track point, and controlling the operation of the lifting hook by using the command speed.

As a possible implementation manner of the second aspect, the apparatus further includes a training unit, and the training unit is configured to:

As a possible implementation manner of the second aspect, the adjusting unit is configured to:

comparing the degree of fit to a preset fit threshold;

A third aspect of the present application provides a computing device comprising:

a communication interface;

at least one processor coupled with the communication interface; and

at least one memory coupled to the processor and storing program instructions that, when executed by the at least one processor, cause the at least one processor to perform the method of any of the first aspects.

A fourth aspect of the present application provides a computer readable storage medium having stored thereon program instructions that, when executed by a computer, cause the computer to perform the method of any of the first aspects described above.

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiment(s) described hereinafter.

Drawings

The individual features of the invention and the connections between the individual features are further explained below with reference to the drawings. The figures are exemplary, some features are not shown to scale, and some of the figures may omit features that are conventional in the art to which the application relates and are not essential to the application, or show additional features that are not essential to the application, and the combination of features shown in the figures is not intended to limit the application. In addition, the same reference numerals are used throughout the specification to designate the same components. The specific drawings are illustrated as follows:

fig. 1A is a schematic structural diagram of an application scenario according to various embodiments of the present application;

FIG. 1B is a schematic structural diagram of a tower crane applied in various embodiments of the present application;

fig. 2 is a schematic diagram of an embodiment of a tower crane control parameter adjusting method provided in the embodiment of the present application;

fig. 3 is a flowchart of an embodiment of a tower crane control parameter adjusting method provided in the embodiment of the present application;

fig. 4 is a schematic diagram of tower crane operation control of the tower crane control parameter adjustment method provided by the embodiment of the present application;

fig. 5 is a schematic diagram of a neural network structure of the tower crane control parameter adjustment method provided in the embodiment of the present application;

fig. 6 is a control operation flowchart of the tower crane control parameter adjustment method provided in the embodiment of the present application;

FIG. 7 is a control operation flow chart of a tower crane control parameter adjustment method provided in the embodiment of the present application;

FIG. 8 is a control operation flow chart of a tower crane control parameter adjustment method provided in the embodiment of the present application;

fig. 9 is a schematic diagram of an embodiment of a tower crane control parameter adjustment method provided in the embodiment of the present application;

fig. 10A is a diagram illustrating a comparison effect between a planned trajectory and an actual trajectory of a tower crane lifting direction after a third tower crane control parameter adjusting method provided by the embodiment of the application is applied;

fig. 10B is a diagram illustrating a comparison effect between a planned trajectory and an actual trajectory of a tower crane rotation direction after applying the third method for adjusting tower crane control parameters according to the embodiment of the present application;

fig. 10C is a diagram illustrating a comparison effect between a planned trajectory and an actual trajectory in the tower crane arm span direction after applying the third tower crane control parameter adjusting method provided in the embodiment of the present application;

fig. 10D is a diagram showing a comparison effect between a planned trajectory and an actual trajectory of a tower crane hook in a tower crane cylindrical coordinate system after applying the third method for adjusting tower crane control parameters provided by the embodiment of the present application.

Fig. 11 is a schematic diagram of an embodiment of a tower crane control parameter adjusting device provided in the embodiment of the present application;

fig. 12 is a schematic diagram of an embodiment of a tower crane control parameter adjusting device provided in the embodiment of the present application;

fig. 13 is a schematic diagram of an embodiment of a tower crane control parameter adjusting device provided in the embodiment of the present application;

fig. 14 is a schematic diagram of a computing device provided in an embodiment of the present application.

Detailed Description

The terms "first, second, third and the like" or "module a, module B, module C and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order, it being understood that specific orders or sequences may be interchanged where permissible to effect embodiments of the present application in other than those illustrated or described herein.

In the following description, reference numerals indicating steps such as S110, S120 \ 8230 \8230 \ 8230, etc. do not necessarily indicate that the steps are performed, and the order of the front and rear steps may be interchanged or performed simultaneously, where the case allows.

The term "comprising" as used in the specification and claims should not be construed as being limited to the items listed thereafter; it does not exclude other elements or steps. It should therefore be interpreted as specifying the presence of the stated features, integers, steps or components as referred to, but does not preclude the presence or addition of one or more other features, integers, steps or components, and groups thereof. Thus, the expression "a device comprising means a and B" should not be limited to a device consisting of only components a and B.

Reference in the specification to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the invention. Thus, appearances of the phrases "in one embodiment" or "in an embodiment" in various places throughout this specification are not necessarily all referring to the same embodiment, but may. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments, as would be apparent to one of ordinary skill in the art from this disclosure.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. In the case of inconsistency, the meaning described in the present specification or the meaning derived from the content described in the present specification shall control. In addition, the terminology used herein is for the purpose of describing embodiments of the present application only and is not intended to be limiting of the present application. To accurately describe the technical contents in the present application and to accurately understand the present invention, the terms used in the present specification are explained or defined as follows before the description of the specific embodiments:

1) PID (proportional integral Differentiation): in engineering practice, the most widely used control law of the regulator is proportional, integral and derivative control, which is called PID control for short, also called PID regulation. When the structure and parameters of the controlled object cannot be completely mastered or an accurate mathematical model is not obtained, and other technologies of the control theory are difficult to adopt, the structure and parameters of the system controller must be determined by experience and field debugging, and the application of the PID control technology is most convenient. I.e. when we do not have complete knowledge of a system and the controlled object, or cannot obtain system parameters by effective measurement means, the PID control technique is most suitable. The PID controller calculates the control quantity by using proportion, integral and differential according to the error of the system to control.

2) BP (back propagation) neural network: error Back Propagation tracing (Error Back Propagation tracing), abbreviated as BP. The system solves the problem of the learning of the hidden layer connection right of the multilayer neural network and gives a complete derivation in mathematics. A multi-layer feed-forward network that employs such an algorithm for error correction is commonly referred to as a BP network. The BP neural network has arbitrary complex pattern classification capability and excellent multidimensional function mapping capability, and solves the problems of XOR and other problems which cannot be solved by a simple perceptron. Structurally, a BP network has an input layer, a hidden layer, and an output layer; in essence, the BP algorithm calculates the minimum value of the objective function by using a gradient descent method with the square of the network error as the objective function.

3) TCP (Transmission Control Protocol): is a connection-oriented, reliable, byte stream-based transport layer communication protocol. TCP is intended to accommodate layered protocol hierarchies that support multiple network applications. Reliable communication services are provided by means of TCP between pairs of processes in host computers connected to different but interconnected computer communication networks. TCP assumes that it can obtain simple, possibly unreliable, datagram service from lower level protocols. In principle, TCP should be able to operate over a variety of communication systems connected from hard wire to packet switched or circuit switched networks.

The prior art method is described first, and then the technical solution of the present application is described in detail.

In a current motion control system of the unmanned tower crane, a high-precision index requires that a lifting hook moves according to a planned path, and the deviation between the path of the lifting hook and the planned path needs to be controlled within a minimum range in the motion process. However, the mechanical structure and transmission mechanism of the tower crane determine the motion equation and power equation of the tower crane, and the accurate establishment is difficult. Therefore, a common method is to skip the model building of the tower crane, directly analyze the output value and the current value of each motion axis by using a PID (proportional integral Differentiation) in the motion process, and dynamically compensate the output value. However, because the tower cranes of various models have different performances, the transmission devices are different, and the PID parameters are difficult to adjust under the actual condition. The same set of PID parameters is also difficult to generalize across models. The tuning process of the current PID parameters is usually tried and tried by the experience of the field engineer. The trial and error process is time-consuming, the accuracy of the PID parameters obtained by trial and error is not enough, and the control precision of the tower crane operation cannot be ensured.

The prior art has the following defects: the PID parameter adjustment and calibration process is time-consuming, and the accuracy of the PID parameter obtained by trial and error is not enough, so that the control precision of the tower crane operation cannot be ensured.

Based on the technical problems in the prior art, the application provides a tower crane control parameter adjusting method. According to the method and the device, the PID parameters are accurately predicted by using the neural network model, the self-adaptive adjustment of the tower crane control parameters is realized, and the control precision of the tower crane operation is improved, so that the technical problems that the accuracy of the PID parameters is not enough and the control precision of the tower crane operation cannot be guaranteed in the prior art are solved. In addition, the control parameters can be continuously learned and corrected in the tower crane operation process, and the PID parameter adjusting period can be shortened, so that the technical problem that the adjusting process is time-consuming in the prior art is solved.

The application provides various method embodiments, various device embodiments, computing equipment embodiments and storage medium embodiments for tower crane control parameter adjustment. Application scenarios of the embodiments of the present application are described below with reference to fig. 1A and 1B.

Fig. 1A shows a control system of a tower crane of the present application, which includes video equipment, a controller, a frequency converter, a motor, and an encoder.

The video equipment is used for acquiring a reference point in the operation process of the tower crane hook in advance, and the reference point comprises an initial position, a target position and an obstacle front accessible position which should be avoided of the tower crane hook. And the reference point coordinates acquired by the video equipment are coordinates of a user coordinate system. The video equipment and the controller are communicated through a Modubus TCP protocol.

Wherein, the motor is used for driving the removal of tower crane lifting hook. The tower crane includes a plurality of axle, a plurality of direction of adjusting promptly. Each shaft is driven by a motor.

The frequency converter is used for receiving the target speed output by the controller and controlling the rotating speed of the motor according to the target speed. One frequency converter is correspondingly used for controlling one axial direction.

The encoder is used for acquiring the actual position of the tower crane lifting hook in the moving process so as to help the tower crane lifting hook to avoid obstacles according to a planned track and reach a target position. The encoder and the controller are communicated through a Modubus 485 protocol.

The controller is used for generating a planning track of the tower crane hook according to a reference point, determining a target speed of each track point, correcting the target speed according to the position deviation of the actual position and the planning position of the tower crane hook acquired by the encoder in the operation process of the hook, obtaining an instruction speed of each track point, driving the frequency converter of the tower crane to control the hook to bypass an obstacle according to the planning track and accurately reach the target position, and the instruction speed is the rotating speed of the frequency converter of each shaft of the tower crane.

The method embodiments of the present application operate in a controller, and the device embodiments are deployed in the controller.

FIG. 1B shows a tower crane applied in various embodiments of the present application, which includes three axes of elevation, rotation, and reach.

The video equipment in fig. 1A obtains coordinates of the tower crane hook in a user coordinate system, and the coordinates can be converted into coordinates in a cartesian coordinate system in fig. 1B through a conversion matrix.

The structure of fig. 1B is an example of a tower crane, and in practical scenarios, the tower crane may include other numbers of axes, such as axes that move laterally and/or longitudinally along the guide rail.

Taking three shafts of lifting, rotating and arm spreading as an example, in the operation process of the tower crane lifting hook, the three shafts respectively drive the alternating current motor to execute movement through three frequency converters, and the current position of each shaft is acquired through an absolute value encoder. And the vision equipment transmits the position of the main path point to the controller through ModbusTCP according to the monitored position of the obstacle. The controller sends the motion instruction to three converters through the Modbus485 bus, comes with the position information acquisition of three encoders simultaneously. In one example, the communication cycle of the Modbus485 bus may be 50ms.

Fig. 2 is a schematic diagram of an embodiment of a tower crane control parameter adjusting method provided by the embodiment of the application. The method may be performed in the controller of fig. 1A. As shown in fig. 2, the method may include:

step S110, predicting PID parameters by utilizing a neural network model based on the position deviation between the track planning point position and the actual position of the tower crane hook to obtain the predicted value of the PID parameters;

step S120, controlling the operation of the lifting hook based on the predicted value of the PID parameter, and calculating the fitting degree of the actual path and the planned path of the operation of the lifting hook;

step S130, adjusting the learning rate and inertia parameters of the neural network model according to the fitting degree;

and S140, predicting the PID parameters by using the neural network model based on the adjusted learning rate and the inertia parameters to obtain the PID parameters for controlling the operation of the lifting hook.

Referring to fig. 1A and 2, a neural network model and a PID control module may be provided in the controller of fig. 1A. The PID control module is also referred to as a PID controller. In order to accurately control the operation of the tower crane, the PID parameters are accurately predicted by using the neural network model, and then the PID controller is used for controlling the operation of the tower crane hook. The PID parameters comprise a proportional parameter Kp, an integral parameter Ki and a differential parameter Kd. Generally, a control system improves the stability, accuracy and rapidity of real-time control by adjusting PID parameters, and the method specifically comprises the following steps:

1) Stability, namely after the system is disturbed in an equilibrium state, the controlled quantity of the system is required to quickly return to a steady state, and the stability of the system is improved by adjusting a differential parameter Kd;

2) The accuracy, namely when the system is in a steady state, the steady state error is as small as possible, and the proportional parameter Kp and the integral parameter Ki are adjusted to improve the steady state precision;

3) Rapidity, that is, the time of the system for dynamic response is short, and is generally measured by the length of the transition time; the response speed is generally increased by adjusting the proportional parameter Kp and the derivative parameter Kd, and the response speed is decreased by adjusting the integral parameter Ki.

In step S110, a trajectory planning point position of the tower crane hook and an actual position of the tower crane hook are first obtained. And inputting the position deviation between the position of the track planning point and the actual position into a neural network model, predicting the PID parameters by using the neural network model, and outputting the predicted value of the PID parameters. In one example, the PID parameters can be predicted using a BP neural network model.

In step S120, the predicted value of the PID parameter output by the neural network model in step S110 is used as an input parameter for the PID control module. And in the PID control module, controlling the operation of the lifting hook based on the predicted value of the PID parameter. And acquiring the actual path of the lifting hook running under the control, and calculating the fitting degree of the actual path and the planned path.

In step S130, it is first determined whether the degree of fitting calculated in step S120 reaches a standard. And if the fitting degree does not reach the standard, adjusting the learning rate and the inertia parameters of the neural network model until the fitting degree reaches the standard.

Wherein, the learning rate represents the amplitude of each parameter update of the neural network model. If the learning rate is too large, the parameters to be optimized fluctuate near the minimum value; if the learning rate is too small, the convergence speed of the parameters to be optimized is slow. The larger the inertia parameter of the neural network model is, the larger the inertia of the weight coefficient adjustment is, i.e. the adjustment amount of each weight coefficient is more closely related to the adjustment amount of the previous weight coefficient. Therefore, the learning rate and the inertia parameter are appropriately adjusted, and the PID parameter adjustment period can be shortened. And the learning rate and the inertia parameters of the neural network model are adjusted according to the fitting degree, so that the self-adaptive adjustment of the tower crane control parameters can be realized, and the control precision of the tower crane operation is improved.

In step S140, after the learning rate and the inertial parameter are adjusted, the position of the trajectory planning point of the tower crane hook and the actual position of the tower crane hook are obtained. And inputting the position deviation between the position of the track planning point and the actual position into a neural network model, predicting PID parameters by using the neural network model, and outputting the PID parameters for controlling the operation of the lifting hook.

According to the method and the device, the PID parameters are accurately predicted by using the neural network model, the control parameters can be continuously learned and corrected in the tower crane operation process, the self-adaptive adjustment of the tower crane control parameters is realized, and the tower crane operation control precision is improved.

Fig. 3 is a flowchart of an embodiment of a tower crane control parameter adjusting method provided in the embodiment of the present application. Referring to fig. 1A to 3, in the operation process of the tower crane, the obstacle is positioned by the vision device, and then the controller performs path planning according to the position of the obstacle and the starting point and ending point positions to determine the operation path of the tower crane hook. And the controller plans the track of the joint space according to the running path. Wherein for an operating arm with n degrees of freedom, all its link positions can be determined by a set of n joint variables. That is, the end pose of the mechanical arm with n degrees of freedom is determined by n joint variables, which are collectively referred to as n-dimensional joint vectors. The space formed by all the joint vectors is called joint space. In the example of a tower crane, the space represented by the positions on the three motion axes of lifting, slewing, and arm extension may be referred to as joint space. And performing PID control in the joint movement process, acquiring an actual movement path in the control process, and calculating the fitting degree of the actual movement path and the planned path. And if the fitting degree does not reach the standard, adjusting the PID parameters to ensure that the fitting degree of the actual motion path and the planned path reaches the standard.

In one embodiment, the predicting PID parameters using the neural network model comprises:

In one embodiment, the method further comprises:

Fig. 4 is a schematic diagram of tower crane operation control of the tower crane control parameter adjustment method provided by the embodiment of the application. As shown in fig. 1A-4, in one example, interpolation may be used to plan a planned trajectory of a hook, which is represented by discrete trajectory points. On the one hand, the position of a track planning point of a tower crane hook can be obtained. On the other hand, in the operation process of the tower crane, the coordinates of the actual positions of the shafts of the tower crane can be acquired through the encoder. And then calculating the position deviation between the position of the track planning point and the actual position, namely the position deviation between the actual position of the tower crane encoder shown in fig. 4 and the position of the track planning point. The letter "e" in fig. 4 indicates the above positional deviation. Referring to fig. 4, the actual position of the tower crane encoder, the position of the track planning point and the position deviation e are input into a BP neural network model, and the PID parameters are predicted by using the BP neural network model to obtain a proportional parameter Kp, an integral parameter Ki and a differential parameter Kd. And then inputting the proportional parameter Kp, the integral parameter Ki, the differential parameter Kd and the position deviation e into a PID control module, and outputting a control quantity U. The control amount U is used to control the controlled object. In the embodiment of the present application, the controlled object is the frequency converter in fig. 1A; the neural network model and the PID control module are provided in the controller of FIG. 1A.

Fig. 5 is a schematic diagram of a neural network structure of the tower crane control parameter adjustment method provided in the embodiment of the present application. As shown in fig. 5, the neural network model includes an input layer, an intermediate layer, and an output layer. And inputting the position of the track planning point, the actual position of the encoder, the deviation of the position of the planning point and the actual position of the encoder and a stability coefficient of 1 into the neural network model. And (3) predicting the PID parameters by using a neural network model, and outputting a proportional parameter Kp, an integral parameter Ki and a differential parameter Kd by an output layer.

In one example of a neural network model, the middle layer coefficients are a matrix of 4 rows and 5 columns:

the output layer coefficients are 3 rows and 5 columns of matrix:

the input layer outputs are:

R＝[ProfilePos EncoderPos PosError 1]

wherein, the Profile pos represents the position of the trajectory planning point; encoderPos represent the actual position of the encoder; posError represents the deviation of the planned point position from the actual encoder position; "1" represents a stability factor.

The inputs to the middle layer are:

wherein k represents the training step index of the neural network model, i.e., "k" represents the kth training step.

The interlayer activation function is:

the middle layer output is:

M _i (k)＝g(net2 _i (k))

wherein i =1,2,3,4,5.

The input variables of the output layer are:

the activation function of the output layer is:

the output of the output layer is:

O _m (k)＝f(net3 _m (k))

wherein m =1,2,3.

The performance indexes in the movement process are as follows:

wherein Y represents the output position, which is the command position to the frequency converter, i.e. the compensated command position; r represents a deviation of the actual position from the commanded position.

And (3) calculating a weight coefficient:

wherein β is a learning rate, and γ is an inertia coefficient.

The 'U' in the above formula represents the control quantity output by the PID control module, and is calculated by adopting a discrete incremental PID algorithm to obtain:

U(k)＝U(k-1)+O ₁ (k)（e(k)-e(k-1))+o ₂ (k)e(k)+O ₃ (k)(e(k)-2e(k-1)+e(k-2))

wherein O is three output parameters of the output layer, and O1, O2 and O3 are Kp, ki and Kd in sequence; e denotes the position deviation between the trajectory planning point position and the actual position.

Finally, the following can be obtained:

wherein, the first and the second end of the pipe are connected with each other,

wherein "sgn" in the formula is a sign function.

Similarly, the weight coefficient increment of the middle layer can be obtained:

in conclusion, the weight coefficients of W1 and W2 are increased in each period and added to the weight coefficient matrix of the previous period, so that Kp, ki and Kd are automatically corrected.

In one embodiment, controlling the hook operation based on the predicted value of the PID parameter comprises:

based on the predicted value of the PID parameter, obtaining the correction quantity of the target speed of the current track point by utilizing a PID control module;

Fig. 6 is a control operation flowchart of a tower crane control parameter adjustment method provided in the embodiment of the present application. As shown in fig. 6, in the first embodiment of controlling the operation of the hook, the method may specifically include the following steps:

and S210, obtaining a planned track of the lifting hook in a tower crane Cartesian coordinate system by using a speed planner according to the coordinates of a plurality of reference points of the tower crane lifting hook in the tower crane Cartesian coordinate system.

Wherein, in some embodiments, the reference point of the hook is obtained by a vision device, and the reference point of the hook is a passing point in front of an obstacle encountered during the movement of the hook, the starting position of the hook and the target position. In other embodiments, the reference point of the hook is obtained from a map of the site of the hoist.

Wherein, the tower crane cartesian coordinate system uses the central point of tower crane as the original point, including the three rectangular axis in cartesian space.

Wherein, in some embodiments, the coordinates of each reference point acquired by the vision device in the user coordinate system are obtained. In other embodiments, the coordinates of each reference point in the tower crane cartesian coordinate system are obtained by measurement.

The planning track of the lifting hook in the tower crane Cartesian coordinate system is planned by utilizing an interpolation method, the planning track is expressed by discrete track points, each track point has a coordinate and arrival time in the tower crane Cartesian coordinate system, so that the target speed of each axis of the tower crane can be planned, and the track points comprise the initial position and the target position of the lifting hook. In some embodiments, the positions of the trace points are distributed according to a time period, the arrival time of each trace point is distributed equidistantly, and one trace point is distributed every 50ms.

The interpolation method is characterized in that a plurality of points are interpolated in a Cartesian space between every two adjacent reference points of the lifting hook, each reference point and the interpolation point form a Cartesian space track together, and meanwhile, the continuity of speed and acceleration is realized on each track point so as to plan the track.

And S220, obtaining the coordinates of each track point in a tower crane cylindrical coordinate system through a kinematics inverse solution method according to the coordinates of each track point in the tower crane Cartesian coordinate system on the planned track.

Wherein, the tower crane cylindrical coordinate system uses the central point of tower crane as the initial point, and the coordinate axis of tower crane cylindrical coordinate system is each axle of tower crane, so can say that the space of tower crane cylindrical coordinate system is the joint space of tower crane.

The kinematic inverse solution method comprises an analytic method, a numerical method and a geometric method.

And step S230, obtaining the target speed of each track point according to the coordinate of each track point in the tower crane cylindrical coordinate system.

The target speed comprises the speed of each coordinate axis direction of the tower crane cylindrical coordinate system, namely the speed of each axis of the tower crane.

Wherein, in some embodiments, the tower crane includes three axes of lift, rotation and reach, and the target speed includes speeds on the three axes of lift, rotation and reach. In some embodiments, when the track points are periodically distributed in time, the target speed of one track point on one axis of the tower crane is the quotient of the coordinate difference on the axis from the track point to the next track point and the arrival time difference of the two track points.

And S240, correcting the target speed of the current track point by using the first PID controller according to the position deviation of the current track point, and obtaining the correction quantity of the target speed of the current track point.

And obtaining the correction quantity of the target speed of the current track point on each axis of the tower crane by utilizing a first PID controller according to the position deviation of the current track point on each axis of the tower crane, and forming the correction quantity of the target speed of the current track point.

The command speed realizes feedforward control on frequency converters of all shafts of the tower crane, and the lifting hook is driven to operate according to a planned track. Illustratively, when the tower crane comprises three motion axes of lifting, rotating and arm stretching, the command speed comprises the lifting speed of the hook, the rotating speed of the hook around the tower crane and the arm stretching speed of the hook, and the motors of the lifting, rotating and arm stretching are respectively subjected to feedforward control.

The position deviation is an operation position error of the lifting hook, and is a coordinate difference of an actual position, corresponding to the corresponding track point, of the lifting hook in each track point on the planning track in the tower crane cylindrical coordinate system, wherein the coordinate difference comprises coordinate differences on each axis of the tower crane cylindrical coordinate system.

In some embodiments, coordinates of the actual positions corresponding to the track points in the tower crane cylindrical coordinate system are acquired through the encoder. In other embodiments, the coordinates of the track points in the user coordinate system are obtained through the vision equipment, and are converted into the coordinates in the tower crane cartesian coordinate system, and then are converted into the coordinates in the tower crane cylindrical coordinate system.

The position deviation can influence the lifting hook to move according to a planned track and needs to be corrected. In some embodiments, the correction speed of the current track point is obtained by a PID method according to the position deviation of the current track point. In other embodiments, the correction speed of each track point is obtained according to the position deviation of the current track point and the running time between the current track point and the corresponding next track point.

It is to be emphasized that: if the current track point is the first track point, the position deviation is 0 because the lifting hook does not move, and the command speed is equal to the target speed.

S243: and taking the speed difference between the target speed of the current track point and the actual speed from the last track point to the current track point as the primary compensation speed of the current track point.

And taking the difference between the target speed of the current target and the actual speed of the last track point as the primary compensation speed of the current track point so as to correct the system-level error of the controller.

The steps S240 and S243 may be executed in parallel, or step S240 may be executed first and then step S243 is executed, or step S243 may be executed first and then step S240 is executed. The embodiment only illustrates that step S240 is executed first and then step S243 is executed.

S246: and taking the target speed of the current track point, the correction quantity of the target speed of the current track point, the sum of the primary compensation speed of the current track point and the speed feedforward of the next track point as the command speed of the current track point.

Wherein the velocity feed forward of a track point is the target velocity of the track point.

The target speed of the next track point is used as speed feedforward to correct the instruction speed of the current track point, and the stability of each motor of the tower crane in the moving process is improved.

And S250, sending the instruction speed of the current track point to a corresponding frequency converter so as to control motors of all shafts of the tower crane to drive the lifting hook to operate.

Wherein, each axle of tower crane lifting hook removes or rotatory through corresponding motor control, and the rotational speed of motor drives through the instruction speed of converter input. The controller in each trace point diagram 1A sends a command speed to the frequency converter to control the motor to drive the lifting hook to each trace point.

After the instruction speed of the current track point is sent, the next track point is taken as the current track point, and a new round of control is carried out until the instruction speed of the penultimate track point is sent. And after the command speed of the penultimate track point is sent, the lifting hook moves to the last track point, namely the target point.

In some embodiments, after the command speed of the penultimate trace point is sent, it is determined that the target point is reached, and if a position deviation still exists at the target point, the trace control is continued.

S260: and judging whether the current track point is the second to last track point or not. If the last trace point is the last trace point, the control is finished; otherwise, after step S270 is executed, the process proceeds to step S240.

S270: and taking the next track point as the current track point, and performing a new round of control.

In the above example of controlling the operation of the hook, the target speed of the current track point is corrected by using a PID method according to the position deviation during the motion process of the hook, and the primary compensation speed of the current track point is increased to correct the command speed of the current track point together with the speed feedforward. The technical scheme of this embodiment realizes the more accurate motion control to the tower crane, improves motor pivoted stability simultaneously, has further improved the security of unmanned hoist and mount.

Fig. 7 is a control operation flowchart of a tower crane control parameter adjustment method provided in the embodiment of the present application. As shown in fig. 7, in an embodiment two of controlling the operation of the hook, on the basis of the above embodiment one, a secondary PID control is added, and according to a position deviation of a last trajectory point, i.e., a target point, a second PID controller is used to generate a supplementary motion speed of the tower crane, i.e., a secondary compensation speed, so that the tower crane accurately reaches the target point according to the secondary compensation speed, and the safety of unmanned hoisting is further improved. The second embodiment is implemented in the controller of fig. 1A, and may specifically include the following steps:

s310: and according to the coordinate of the tower crane hook reference point in the tower crane Cartesian coordinate system, obtaining the planned track of the hook in the tower crane Cartesian coordinate system by using a speed planner.

Please refer to step S210 in the first embodiment of the tower crane control method for details and advantages of this step.

S320: and obtaining the coordinates of each track point in a tower crane cylindrical coordinate system by a kinematics inverse solution method according to the coordinates of each track point in the tower crane Cartesian coordinate system on the planned track.

Please refer to step S220 in the first embodiment of the tower crane control method for details and advantages of this step.

S330: and obtaining the target speed of each track point according to the coordinate of each track point in the tower crane cylindrical coordinate system and the time interval between two adjacent track points, and setting the current track point as the first track point.

Please refer to step S230 in the first embodiment of the tower crane control method for details and advantages of this step.

S340: and according to the position deviation of the current track point, correcting the target speed of the current track point by using the first PID controller, and obtaining the correction quantity of the target speed of the current track point.

Please refer to step S240 in the first embodiment of the tower crane control method for details and advantages of this step.

S343: and taking the difference between the target speed of the current track point and the actual speed from the last track point to the current track point as the primary compensation speed of the current track point.

Please refer to step S243 in the first embodiment of the tower crane control method for details and advantages of this step.

Step S340 and step S343 may be executed in parallel, or step S340 may be executed first and then step S343 may be executed, or step S343 may be executed first and then step S340 may be executed. The embodiment only illustrates that step S340 is executed first and then step S343 is executed.

S346: and taking the target speed of the current track point, the correction quantity of the target speed of the current track point, the sum of the primary compensation speed of the current track point and the speed feedforward of the next track point as the command speed of the current track point.

Please refer to step S246 in the first embodiment of the tower crane control method for describing the detailed method and advantages of this step.

S350: and sending the instruction speed of the current track point to a corresponding frequency converter so as to control motors of all shafts of the tower crane to drive the lifting hook to operate.

Please refer to step S250 in the first embodiment of the tower crane control method for details and advantages of this step.

S360: and judging whether the current track point is the penultimate track point. If the trace point is the penultimate trace point, the step S380 is operated; otherwise, after step S370 is executed, the process proceeds to step S340.

S370: and taking the next track point as the current track point, and performing a new round of control.

Please refer to step S270 in the first embodiment of the tower crane control method for details and advantages of this step.

S380: and judging whether the tower crane hook reaches the target position, namely the planning position of the last track point.

And if the actual position and the target position acquired by the encoder are the same in the tower crane cylindrical coordinate system, the lifting hook of the tower crane reaches the target position, the control on the tower crane is finished, and otherwise, the step S390 is executed.

S390: and according to the position deviation of the target position in the tower crane cylindrical coordinate system, obtaining the secondary compensation speed of the target position by using a second PID controller.

And obtaining secondary compensation speed components on each axis of the tower crane by using a second PID controller according to the position deviation of the target position on each axis of the tower crane, thereby forming the secondary compensation speed of the target position.

And when the tower crane finishes all the motions in the planning, the lifting hook is accurately controlled to the target position by utilizing the supplementary motion.

S393: and sending the secondary compensation speed to a corresponding frequency converter to control a motor of the tower crane to drive the lifting hook to operate.

Wherein, the lifting hook may not be accurately controlled to the target position through one supplementing movement, and the step S380 is executed after the step is executed, so as to judge whether more supplementing movements are needed.

In summary, the second embodiment of controlling the operation of the lifting hook is added with the second PID control on the basis of the first embodiment, and the second PID controller is used for generating the second compensation speed of the tower crane according to the position deviation of the target point, so that the tower crane accurately reaches the target point according to the second compensation speed, and the safety of unmanned lifting is further improved.

Fig. 8 is a control operation flowchart of the tower crane control parameter adjustment method provided in the embodiment of the present application. As shown in fig. 8, the third embodiment for controlling the operation of the hook is a detailed implementation of the second embodiment, which is operated in the controller of fig. 1A.

In this embodiment, each trace point is periodically distributed in time, the motion from the current trace point to the next trace point is the motion of the current period, and the speed of the current trace point is the speed of the current period.

The symbols involved in the subsequent flow are illustrated below:

{ U }: representing a user coordinate system, here a camera coordinate system, as a cartesian coordinate system;

{ B }: a base coordinate system is represented, wherein a tower crane coordinate system is represented, and the base coordinate system is a Cartesian coordinate system;

{ BP }' representing a base coordinate system, wherein the base coordinate system represents a tower crane coordinate system and is a cylindrical coordinate system;

and the pose transformation relation from the camera coordinate system to the tower crane coordinate system is shown, and the symbols of other coordinate transformation relations are the same as the pose transformation relation.

Also known as the T matrix.

Referring to fig. 8, the third embodiment may specifically include the following steps:

s410: and obtaining the coordinates of the reference points of the lifting hook in the user coordinate system according to the visual equipment.

Wherein, the reference point of lifting hook is the point in front of the barrier that the lifting hook met in the removal process, obtains through visual equipment. Because the vision device is human-centered, its coordinate system is the user coordinate system. The coordinates of each reference point in the user coordinate system, i.e., the { U } coordinate system, is a matrix of M x3, M representing the number of reference points, and 3 representing three dimensions of the { U } coordinate system. The vision device communicates the coordinates of the reference points in the user coordinate system to the controller via Modbus TCP.

S412: using transformation matrices

And obtaining the coordinates of each reference point in a tower crane Cartesian coordinate system.

Therein, benefit ofUsing conversion matrices from { U } to { B }

And obtaining the coordinates of each reference point in the { B } coordinate system. The content of the T matrix is shown in formula (1).

Wherein the content of the first and second substances,

the device comprises an attitude transformation part and a translation transformation part, wherein the attitude transformation part is represented by a 3X 3R rotation matrix, and the translation transformation part is represented by a 3X 1P translation vector.

S420: and obtaining periodic track points on the planning track of the lifting hook according to the coordinates of the reference points of the tower crane lifting hook in the { B } coordinate system.

And planning the planning track of the lifting hook in the B coordinate system by using an interpolation method in the speed planner, and obtaining the coordinates and the arrival time of each track point on the planning track in the B coordinate system. The planning track is provided with N +1 track points, the numbers of the track points are respectively 0 to N, and the track points comprise the initial position and the target position of the lifting hook.

S430: and obtaining the coordinates of periodic track points of each axis of the tower crane through inverse kinematics solution.

Specifically, according to the coordinates of each track point in the { B } coordinate system on the planned track, the coordinates of each track point in the tower crane cylindrical coordinate system are obtained through a kinematics inverse solution method.

And obtaining the coordinates of each track point in a tower crane cylindrical coordinate system, namely a { BP } coordinate system, by using an analytic method according to the formula (2).

representing the coordinates of the nth trace point in a { BP } coordinate system, particularly three-axis coordinates

r _i Coordinates of the spread axis of the arm, theta, representing the ith locus point _i Coordinates of the axis of rotation, z, representing the ith locus point _i The coordinates of the lifting axis of the ith trace point are shown,

coordinate in the B coordinate system representing the ith trace point, and height coordinate z thereof _i The same as the coordinates of the elevation axis in the { BP } coordinate system.

S440: and obtaining the target speed of each periodic track point of each shaft, and sending the target speed to a first PID controller.

Specifically, the first PID controller obtains the target speed of each track cycle according to the coordinate of each track point in the { BP } coordinate system and the time interval between two adjacent track points, and sets the current track point as the first track point.

In this embodiment, the trace points are periodically distributed in time, and the target speed of each trace period is obtained by using equation (3) according to the coordinate of each trace point in the { BP } coordinate system and the time period between two adjacent trace points:

a vector representing the target speed of the nth cycle.

Wherein r is _n Showing the spread of the nth locus, r _n-1 And (3) representing the spread position of the (n-1) th track point, wherein the spread position is the coordinate of the spread axis direction in the { BP } coordinate system.

Wherein, theta _n Indicating the angle of revolution, theta, of the nth track point _n-1 Denotes the firstThe rotation angle of n-1 points is the coordinate of the rotation axis direction in the { BP } coordinate system.

Wherein z is _n Showing the elevation of the nth track point, z _n-1 And the lifting height of the (n-1) th track point is shown, and the lifting height is the coordinate in the lifting shaft direction in the { BP } coordinate system.

Wherein, T _sm The time interval between two adjacent trace points, i.e. the time period of the trace points, is represented.

S450: the first PID controller obtains the correction amount of the target speed for the current cycle.

Specifically, the first PID controller obtains the correction amount of the target speed in the current period according to the position deviation of the coordinate of the current track point in the { BP } coordinate system.

The coordinates of the actual positions of the lifting hooks in the arrival time of the track points in the { BP } coordinate system are collected through the encoder, the communication cycle of the Modbus485 bus between the encoder and the controller is 50ms, and the actual positions of the lifting hooks can be conveniently and quickly returned.

And obtaining the position deviation of the coordinate of the current track point (nth track point) in the { BP } coordinate system by using the formula (4), wherein the position deviation is the coordinate difference between the actual position of the current track point and the planned position in the { BP } coordinate system.

Wherein the content of the first and second substances,

vector, er, representing the position deviation of the nth trace point _n 、Eθ _n And Ez _n Components of the vector representing the positional deviation on three coordinate axes in the { BP } coordinate system, r _{n_current} 、θ _{n_current} And z _{n_current} The coordinates of the actual position of the hook in the { BP } coordinate system at the time of arrival corresponding to the nth locus point are shown.

And inputting the position deviation of the current track point into a first PID controller, and obtaining the correction speed of the next track point by using an equation (5). The tuning method of the first PID controller is also referred to as the first PID method.

vector representing correction speed of nth track point, i.e. nth period, ur _n 、Uθ _n And Uz _n Respectively, components of the correction speed on three coordinate axes in the { BP } coordinate system.

Wherein, er _j 、Eθ _j And Ez _j Respectively representing the components of the vector of the position deviation of the jth track point on three coordinate axes in a { BP } coordinate system; er _n 、Eθ _n And Ez _n Components of vectors respectively representing the position deviation of the nth track point on three coordinate axes in a { BP } coordinate system; er _n-1 、Eθ _n-1 And Ez _n-1 And components of the vectors respectively representing the positional deviations of the (n-1) th trace point on three coordinate axes in the { BP } coordinate system.

Wherein, kr _p 、Kθ _p 、Kz _p Indicating the scale-loop coefficient of the first PID controller, kr _i 、Kθ _i 、Kz _i Integral loop coefficient, kr, representing the first PID controller _d 、Kθ _d 、Kz _d Representing the differential loop coefficients of the first PID controller.

Therefore, the correction speed of each track point is accurately corrected through a PID method, so that the lifting hook can safely operate according to the planned track accurately.

S453: and obtaining the primary compensation speed of the current period.

Specifically, the coordinates of the actual positions of the current track point and the previous track point in the { BP } coordinate system are obtained, and the primary compensation speed of the current period is obtained.

Wherein, the compensation speed of the current period can be obtained by using the equation (6).

Wherein the content of the first and second substances,

for the primary compensation speed of the nth period, r _{current_last} 、θ _{current_last} And z _{current_last} Is the coordinate in the { BP } coordinate system of the actual position of the (n-1) th trace point.

It should be further emphasized that the execution sequence of steps S450 and S453 can be executed in parallel or first, and the sequence of this embodiment is just an example.

S456: and taking the sum of the target speed of the current period, the correction quantity of the target speed, the current initial compensation speed and the feed-forward speed of the next period as the command speed of the current period.

Wherein, the command speed of each track point is obtained by equation (7).

Wherein the content of the first and second substances,

indicating the commanded speed for the nth trace point.

It should be noted that, for the 0 th trace point, there is no position deviation for the 0 th trace point, and there is no need to calculate

It is emphasized that in this step also the

The maximum speed is controlled to ensure the operation safety of the tower crane. By the above, by increasing the compensation speed in the command speed of each track pointThe results of past corrections come in, and the electron operation is made more stable by increasing the target speed of the next trace point in the command speed of each trace point.

S460: and sending the instruction speed of the current period to a corresponding frequency converter to control a motor of the tower crane to drive the lifting hook to operate.

And the controller sends the target speed to the frequency converter of each shaft of the tower crane through a Modbus485 bus.

S462: and obtaining the position of each shaft of the tower crane in the current period.

Specifically, the positions, namely the actual positions, of all axes of the tower crane in the current period are obtained through a digital code and are sent to a first PID controller.

S470: and judging whether the command speed of each cycle is sent to be completed or not.

If the transmission is completed, the step S490 is executed, otherwise, the step S480 is executed by regarding the next trace point as the current trace point.

S480: and taking the next track point as the current track point to perform a new round of control.

S490: and judging whether the tower crane hook reaches the target position, namely the planning position of the last track point.

And the target position is the planning position of the last track point. And comparing the actual position corresponding to the target position acquired by the encoder with the coordinate of the target position in the { BP } coordinate system, if the actual position and the coordinate of the target position are the same, enabling the tower crane hook to reach the target position, and ending the control on the tower crane, otherwise, operating the step S493.

S493: and obtaining the secondary compensation speed of the target position.

Specifically, the secondary compensation speed of the target position is obtained from the positional deviation of the coordinates of the target position in the { BP } coordinate system.

The position deviation of the target position, i.e. the nth track point, is the difference between the actual position of the target position acquired by the encoder and the coordinates of the target position in the { BP } coordinate system. Since the correction of the target position may be performed a plurality of times without loss of generality, the positional deviation of the target position is calculated using equation (8) here as an example of the lth correction.

vector Er 'representing position deviation before L-th correction of target position' _L 、Eθ′ _L And Ez' _L Components of the vector representing the positional deviation on three coordinate axes in the { BP } coordinate system, r _{L_current} 、θ _{L_current} 、z _{L_current} Coordinates in the { BP } coordinate system, r, representing the actual position of the target position before the L-th correction acquired by the encoder _target 、θ _target And z _target A planned position representing the target position.

Wherein, the positional deviation of the target position before the L-th correction is inputted to the second PID controller, the second compensation speed of the target position after the L-th correction is generated by the equation (9), and the (10) is used as the command speed after the L-th correction.

Wherein the content of the first and second substances,

the second compensation speed of the L-th correction representing the target position,

instruction speed, ur 'indicating the L-th correction of the target position' _L 、Uθ′ _L And Uz' _L The second compensation speed of the L-th correction representing the target position is divided into three coordinate axes in the { BP } coordinate system.

Wherein, er' _L 、Eθ′ _L And Ez' _L Bit input at Lth correction of target positionSetting deviation as component Er 'on three coordinate axes in { BP } coordinate system' _L-1 、Eθ′ _L-1 And Ez' _L-1 The component Er 'of the positional deviation inputted in the L-1 correction representing the target position on the three coordinate axes in the { BP } coordinate system' _j 、Eθ′ _j And Ez' _j The positional deviation input at the j-th correction representing the target position is divided into three coordinate axes in the { BP } coordinate system.

Wherein, kr' _p 、Kθ′ _p And Kz' _p Denotes the proportionality Ring coefficient, kr 'of the second PID controller' _i 、Kθ′ _i And Kz' _i Integral Loop coefficient, kr 'of the second PID' _d 、Kθ′ _d And Kz' _d Representing the differential loop coefficients of the second PID controller.

It is emphasized that this step is performed in a number of iterations, so that the hook will eventually reach the target position exactly.

Therefore, the lifting hook can finally and accurately reach the target position at the actual position corresponding to the target position by a PID method.

S496: and sending the secondary compensation speed to a corresponding frequency converter to control a motor of the tower crane to drive the lifting hook to operate.

And the controller sends the target speed to frequency converters of three shafts of the tower crane through a Modbus485 bus.

S499: and obtaining the position of each shaft of the tower crane in the current period.

Specifically, the positions, namely the actual positions, of all axes of the tower crane in the current period are obtained through a digital code and are sent to the second PID controller. Then, step S490 is executed.

To sum up, in the third embodiment of controlling the operation of the hook, the coordinates of the reference points of the hook, which are acquired by the video device, in the user coordinate system are converted into the coordinates in the cartesian coordinate system of the tower crane by using the conversion matrix, the correction speed of the current track point is obtained by using the PID method in the operation process of the hook, the instruction speed of the current track point is corrected by increasing the compensation speed of the current track point and the target speed of the next track point, the actual position of the hook is corrected by using the PID method in the target position, and the hook can accurately reach the target position. The technical scheme of this embodiment realizes the more accurate motion control to the tower crane, has further improved the security of unmanned hoist and mount.

In one embodiment, calculating a fit of the actual path traveled by the hook to the planned path comprises:

After the path operation of the tower crane hook is finished, the path fitting degree can be automatically calculated according to the following formula:

for m samples

The estimated value of a model is

Calculate the Total square Sum TSS (Total Sum of Squares) of the sample:

i.e., m times the pseudo-variance of the sample Var (Y) = TSS/m;

calculate Residual Sum of Squares RSS (Residual Sum of Squares):

RSS, sum of squared Error SSE (Sum of Squares for Error);

definition of R ² ＝1-RSS/TSS

In the above formula, R ² The larger the fitting effect, the better; r ² The optimum value of (1); if the model predicts a random value, R ² May be negative(ii) a If the predicted value is constant to the sample expectation, R ² Is 0.

In addition, it is also possible to define:

ESS(Explained Sum of Squares)：

TSS = ESS + RSS, the above equation is only true when unbiased estimation is performed, otherwise, TSS is larger than or equal to ESS + RSS;

among them, ESS is also called Regression Sum of Squares for Regression (SSR).

In one embodiment, the method further comprises:

In one embodiment, adjusting the learning rate and inertial parameters of the neural network model according to the fitness comprises:

comparing the degree of fit to a preset fit threshold;

under the condition that the fitting degree is smaller than the preset fitting threshold, adjusting the learning rate and inertia parameters of the neural network model so that the fitting degree obtained again based on the adjusted neural network model is larger than or equal to the preset fitting threshold;

the process of obtaining the fitting degree again based on the adjusted neural network model comprises the following steps: based on the adjusted learning rate and inertia parameters, the neural network model is utilized to predict the PID parameters again; controlling the operation of the lifting hook based on the PID parameters obtained by the re-prediction; and recalculating the fitting degree of the actual path operated by the lifting hook and the planned path.

Fig. 9 is a schematic diagram of an embodiment of a tower crane control parameter adjusting method provided by the embodiment of the application. Referring to fig. 9, a set of parameters of the learning rate and the inertia coefficient of the neural network may be first given, and after a proper amount of set of tracks are automatically run, the weight coefficients W1 and W2 of the neural network model will be substantially close to the final modulation interval. And then calculating the fitting degree of the actual path operated by the lifting hook and the planned path. If the fitting degree is smaller than the preset fitting threshold value, namely under the condition that the fitting degree does not reach the standard, the learning rate can be slightly adjusted, the inertia coefficient is reduced, and the planning track is executed again until the path reaches the standard. W1, W2, and learning rate and inertial coefficient parameters are then recorded and saved.

In one example, the preset fitting threshold may be set to 0.97.

By adopting the tower crane control parameter adjusting method provided by the embodiment of the application, the PID parameters are accurately predicted by utilizing the neural network model, and the field work can be greatly reduced. And because the same PID parameter is difficult to be used universally in each model, the neural network model can be adapted and trained aiming at different tower crane models, and the control parameter can be matched quickly.

The technical effect of the third embodiment of the hook operation will be described with reference to fig. 10A to 10D.

FIG. 10A is a diagram showing a comparison between a planned trajectory and an actual trajectory of a tower crane in a lifting direction; FIG. 10B is a diagram comparing a planned track and an actual track of a tower crane in a rotation direction; FIG. 10C is a diagram illustrating a comparison between a planned trajectory and an actual trajectory in the tower crane arm span direction; FIG. 10D is a comparison graph of a planned track and an actual track of a tower crane in a tower crane cylindrical coordinate system space. The planned trajectories of fig. 10A to 10D are almost the same as the corresponding actual trajectories. In the embodiment of the application, whether the track precision meets the requirement is judged through the calculation of the fitting degree, and the fitting degree is in accordance with the requirement from 0.97 to 1. The degree of fit between the planned trajectory and the corresponding actual trajectory in fig. 10A to 10D is greater than 0.97. In fig. 10D, the fitting degree of the planned trajectory and the actual trajectory in the space of the tower crane cylindrical coordinate system is 0.99, and the trajectory precision control requirement is met.

As shown in FIG. 11, the application further provides a corresponding tower crane control parameter adjusting device. The apparatus may be provided in the controller of fig. 1A. For the beneficial effects or technical problems to be solved of the apparatus, reference may be made to the description in the method corresponding to each apparatus, or to the description in the summary of the invention, and details are not repeated here.

In this tower crane control parameter adjusting device's embodiment, this device includes:

a prediction unit 100 for: predicting the PID parameters by utilizing a neural network model based on the position deviation between the track planning point position and the actual position of the tower crane hook to obtain the predicted value of the PID parameters;

a control unit 200 for: controlling the operation of the lifting hook based on the predicted value of the PID parameter; calculating the fitting degree of the actual path and the planned path of the operation of the lifting hook;

an adjusting unit 300 configured to: adjusting the learning rate and inertia parameters of the neural network model according to the fitting degree;

the prediction unit 100 is further configured to: and predicting the PID parameters by using the neural network model based on the adjusted learning rate and inertia parameters to obtain the PID parameters for controlling the operation of the lifting hook.

As shown in fig. 12, in one embodiment, the prediction unit 100 includes:

the obtaining subunit 110 is configured to obtain a planned trajectory point position and an actual position of the tower crane hook, and calculate a position deviation between the planned trajectory point position and the actual position;

and the predicting subunit 120 is configured to input the position of the trajectory planning point, the actual position, and the position deviation into the neural network model, and predict a PID parameter by using the neural network model.

In one embodiment, the obtaining subunit 110 is configured to:

In one embodiment, the control unit 200 is configured to:

As shown in fig. 13, in one embodiment, the apparatus further comprises a training unit 400, the training unit 400 being configured to:

In one embodiment, the adjusting unit 300 is configured to:

comparing the fitting degree with a preset fitting threshold;

Fig. 14 is a schematic structural diagram of a computing device 900 according to an embodiment of the present disclosure. The computing device 900 includes: a processor 910, a memory 920, and a communication interface 930.

It is to be appreciated that the communication interface 930 in the computing device 900 shown in fig. 14 may be used to communicate with other devices.

The processor 910 may be coupled to the memory 920. The memory 920 may be used to store the program codes and data. Therefore, the memory 920 may be a storage unit inside the processor 910, an external storage unit independent of the processor 910, or a component including a storage unit inside the processor 910 and an external storage unit independent of the processor 910.

Optionally, computing device 900 may also include a bus. The memory 920 and the communication interface 930 may be connected to the processor 910 through a bus. The bus may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc.

It should be understood that, in the embodiment of the present application, the processor 910 may adopt a Central Processing Unit (CPU). The processor may also be other general purpose processors, digital Signal Processors (DSPs), application Specific Integrated Circuits (ASICs), field Programmable Gate Arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. Or the processor 910 may employ one or more integrated circuits for executing related programs to implement the technical solutions provided in the embodiments of the present application.

The memory 920 may include a read-only memory and a random access memory, and provides instructions and data to the processor 910. A portion of the processor 910 may also include non-volatile random access memory. For example, the processor 910 may also store device type information.

When the computing device 900 is running, the processor 910 executes the computer-executable instructions in the memory 920 to perform the operational steps of the above-described method.

It should be understood that the computing device 900 according to the embodiment of the present application may correspond to a corresponding main body for executing the method according to the embodiments of the present application, and the above and other operations and/or functions of each module in the computing device 900 are respectively for implementing corresponding flows of each method of the embodiment, and are not described herein again for brevity.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

It can be clearly understood by those skilled in the art that, for convenience and simplicity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the units is only one type of logical functional division, and other divisions may be realized in practice, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit.

The functions may be stored in a computer-readable storage medium if they are implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solution of the present application or portions thereof that substantially contribute to the prior art may be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

Embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, where the computer program is used to execute a diversification problem generation method, where the method includes at least one of the solutions described in the above embodiments.

The computer storage media of embodiments of the present application may take any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the context of this document, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device.

A computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device.

Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Computer program code for carrying out operations for aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

It is noted that the foregoing is only illustrative of the preferred embodiments of the present application and the technical principles employed. Those skilled in the art will appreciate that the present invention is not limited to the particular embodiments described herein, and that various obvious changes, rearrangements and substitutions will now be apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present application has been described in more detail with reference to the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention.

Claims

1. A tower crane control parameter adjusting method is characterized by comprising the following steps:

controlling the operation of the lifting hook based on the predicted value of the PID parameter, and calculating the fitting degree of the actual path and the planned path of the operation of the lifting hook;

2. The method of claim 1, wherein predicting PID parameters using the neural network model comprises:

acquiring a track planning point position and an actual position of a tower crane hook, and calculating a position deviation between the track planning point position and the actual position;

3. The method of claim 2, further comprising:

4. A method according to any one of claims 1 to 3, wherein controlling the hook operation based on the predicted value of the PID parameter comprises:

5. The method of any one of claims 1 to 3, wherein calculating a fit of the actual path traveled by the hook to a planned path comprises:

6. The method according to any one of claims 1 to 3, further comprising:

7. The method of claim 6, wherein adjusting learning rates and inertial parameters of the neural network model according to the fitness comprises:

comparing the fitting degree with a preset fitting threshold;

8. The utility model provides a tower crane control parameter adjusting device which characterized in that includes:

a control unit for: controlling the operation of the lifting hook based on the predicted value of the PID parameter, and calculating the fitting degree of the actual path and the planned path of the operation of the lifting hook;

9. A computing device, comprising:

a communication interface;

at least one processor coupled with the communication interface; and

at least one memory coupled to the processor and storing program instructions that, when executed by the at least one processor, cause the at least one processor to perform the method of any of claims 1-7.

10. A computer-readable storage medium having stored thereon program instructions, which, when executed by a computer, cause the computer to perform the method of any of claims 1-7.