An intelligent control method for the boom of a concrete placing boom

Abstract

An intelligent control method for the boom of a concrete placing boom, which relates to the field of control technology for construction engineering machinery. Aiming at the problem that the existing concrete placing boom uses the PID control algorithm to control the hydraulic cylinder to drive the boom to move, resulting in reduced control accuracy of the boom and increased vibration amplitude. The steps are as follows: According to the training samples of the boom hydraulic control parameters randomly generated and the corresponding boom attitude parameters, using one or more of amplitude, construction time, energy consumption, and position accuracy as the optimized motion index of the boom, establish an initial model of the boom control algorithm, train the initial model through a machine learning algorithm to obtain the final model of the boom control algorithm, calculate the boom hydraulic control parameters of the current concrete placing boom through the final model of the boom control algorithm for the cloth placing target position coordinates, and the boom control system of the placing boom drives the boom to move to the cloth placing target position according to the boom hydraulic control parameters of the current concrete placing boom.

Description

Technical Field

[0001] The present invention relates to the technical field of construction engineering machinery control, and particularly relates to an intelligent control method for the boom of a concrete placer. Background Art

[0002] At present, a concrete placer is an essential tool in construction. The boom of the concrete placer realizes reciprocating motion under the push of a hydraulic cylinder. Usually, the PID control algorithm is used to control the hydraulic cylinder. The PID control algorithm has the characteristic of wide applicability. However, since the PID control algorithm controls a single cylinder, after the motion results of the hydraulic cylinders of multiple booms are superimposed, the overall error and vibration effect of the boom of the concrete placer will be continuously amplified, resulting in a reduction in the control accuracy of the boom of the concrete placer and an increase in the vibration amplitude. Summary of the Invention

[0003] Aiming at the problem that the existing concrete placer uses the PID control algorithm to control the hydraulic cylinder to push the boom to move, resulting in a reduction in the control accuracy of the boom of the concrete placer and an increase in the vibration amplitude. The purpose of the present invention is to provide an intelligent control method for the boom of a concrete placer.

[0004] The technical solution adopted by the present invention to solve its technical problems is: an intelligent control method for the boom of a concrete placer, the steps are as follows:

[0005] S11: Through a machine learning algorithm, the training sample data of randomly generated boom hydraulic control parameters and their corresponding boom attitude parameters are used to train to obtain a boom control algorithm, and an initial model of the boom control algorithm of the placer is established. Using one or more of amplitude, construction time, energy consumption, and position accuracy as the optimized motion index, the initial model of the boom control algorithm of the placer is trained to obtain the final model of the boom control algorithm;

[0006] S12: Establish the final model of the boom control algorithm in the boom control system of the placer, convert the cloth target position coordinates into boom attitude parameters and input them into the boom control system of the placer, and calculate the current boom hydraulic control parameter Y of the concrete placer through the final model of the boom control algorithm ie (q jie , v jie , p jie , Δl jie );

[0007] θ ji represents the azimuth angle after the movement of the j-th boom in the i-th group of boom hydraulic control parameters;

[0008] β ji represents the included angle between the j-th boom and the adjacent (j - 1)-th boom after the movement in the i-th group of boom hydraulic control parameters;

[0009] q jie ,v jie ,p jie ,Δl jie respectively represent the hydraulic oil flow rate, flow velocity, oil pressure and cylinder stroke of the hydraulic cylinder of the j-th section in the i-th group of boom hydraulic control parameters;

[0010] i represents the number of groups of boom hydraulic control parameters, i = 1, 2, 3... N, N is a natural number;

[0011] j represents the j-th section boom hydraulic cylinder;

[0012] S13: The boom control system of the concrete placer drives each boom to move according to the boom hydraulic control parameter Y i (q ji ,v ji ,p ji ,Δl ji ) respectively, so that it reaches the concrete placing target position under the condition of satisfying one or more optimized motion indexes of amplitude, construction time, energy consumption, position accuracy.

[0013] For the intelligent control method of the boom of the concrete placer of the present invention, first, according to the training samples of the randomly generated boom hydraulic control parameters and the corresponding boom attitude parameters, using one or more of amplitude, construction time, energy consumption, and position accuracy as the optimized motion indexes of the boom, an initial model of the boom control algorithm is established. The initial model of the boom control algorithm is trained through a machine learning algorithm to obtain the final model of the boom control algorithm and input it into the boom control system of the concrete placer. Finally, the coordinates of the concrete placing target position are converted into boom attitude parameters and input into the boom control system of the concrete placer. The boom hydraulic control parameters of the current concrete placer are calculated through the boom control algorithm. The boom hydraulic control parameters at least include the hydraulic oil flow rate, flow velocity, oil pressure and cylinder stroke of the hydraulic cylinder. The boom control system of the concrete placer drives the boom to move to the concrete placing target position according to the current boom hydraulic control parameters of the concrete placer; for the intelligent control method of the boom of the concrete placer of the present invention, the training samples of the boom hydraulic control parameters, boom attitude parameters and optimized motion indexes are used to train the boom control algorithm through a machine learning algorithm. The boom hydraulic control parameters of the current concrete placer can be calculated through this boom control algorithm for the concrete placing target position, so that the boom control system of the concrete placer can accurately control and automatically drive the boom to move to the concrete placing target position, thereby improving the control accuracy of the boom of the concrete placer and achieving the vibration suppression effect by reducing the vibration amplitude of the boom.

[0014] Further, in the step S11, the steps of the training samples of the boom attitude parameters and the boom hydraulic control parameters are as follows:

[0015] S01: Randomly generate N groups of boom hydraulic control parameters:

[0016] The hydraulic control parameters Y of the i-th group of boom i (q ji , v ji , p ji , Δl ji ; θ 0i , β j0i );

[0017] q ji , v ji , p ji , Δl ji respectively represent the hydraulic oil flow rate, flow velocity, oil pressure and cylinder stroke of the j-th hydraulic cylinder in the hydraulic control parameters of the i-th group of boom;

[0018] θ 0i represents the azimuth angle of the j-th boom before movement in the hydraulic control parameters of the i-th group of boom;

[0019] β j0i represents the included angle between the j-th boom and the adjacent upper boom before movement in the hydraulic control parameters of the i-th group of boom;

[0020] S2: Input the hydraulic control parameters of N groups of booms into the boom control system of the placer respectively to drive the boom to move, and monitor the boom attitude parameters after the corresponding deformation of the hydraulic control parameters of each group of booms:

[0021] Input the hydraulic control parameters of the i-th group of booms into the boom control system of the placer in sequence. The boom control system of the placer issues a control signal to drive the hydraulic cylinder, so that the boom realizes the corresponding movement. After the movement of the boom driven by the hydraulic control parameters of the i-th group of booms is completed, monitor the attitude parameters X of the deformed i-th group of booms i (θ 1i , β j1i ; A ji , B ji , C ji , D ji );

[0022] Among them, θ 1i represents the azimuth angle of the j-th boom after movement in the hydraulic control parameters of the i-th group of booms;

[0023] β j1i represents the included angle between the j-th boom and the adjacent upper boom after movement in the hydraulic control parameters of the i-th group of booms;

[0024] A ji , B ji , C ji , D ji respectively represent the movement indexes of the j-th boom in the hydraulic control parameters of the i-th group of booms, A jiis the amplitude, B ji is the construction time, C ji is the energy consumption, D ji is the position accuracy. The boom attitude parameters are selected from one or more of the above motion indexes as required;

[0025] S3: Generate training samples: Combine the i-th group of boom attitude parameters X i with the i-th group of boom hydraulic control parameters Y i to form a new data element (X i , Y i ). Generate training samples {(X i , Y i ), i = 1, 2, 3... N} from N groups of data elements (X i , Y i ).

[0026] Furthermore, in the step S12, the method for converting the cloth target position coordinates into boom attitude parameters is as follows: Obtain the boom attitude function represented by a spatial point sequence, and determine the boom attitude parameter X e (θ jie , β jie )

[0027]

[0028] Where:

[0029]

[0030] Where, R j is the length of the j-th boom of the concrete placer;

[0031] ɑ j is the angle between the j-th boom and the positive direction of the ρ axis or the horizontal plane;

[0032] β i is the rotation angle of the j-th boom relative to the j-1-th boom;

[0033] θ 1ie represents the azimuth angle of the j-th boom after movement;

[0034] β j1ie represents the angle between the j-th boom and the j-1-th boom after movement.

[0035] K is the number of booms of the concrete placer. Description of the Drawings

[0036] Figure 1 is a flowchart of an embodiment of the intelligent control method for the boom of the concrete placer of the present invention. Detailed Embodiment

[0037] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that the accompanying drawings are in very simplified forms and use non-precise scales, only for conveniently and clearly assisting in explaining the objectives of the embodiments of the present invention.

[0038] In this embodiment, a concrete placing boom with three boom sections is taken as an example. The intelligent control method for the boom of the concrete placing boom of the present invention will be described below with reference to Figure 1 The specific steps are as follows:

[0039] S11: Select a machine learning algorithm, and train the training sample data of the randomly generated boom hydraulic control parameters and the corresponding boom attitude parameters to obtain the boom control algorithm Y = F(X):

[0040] According to the training samples of the randomly generated boom hydraulic control parameters and the corresponding boom attitude parameters {(X i , Y i ), i = 1, 2, 3... N}, taking one or more of the amplitude A ji , construction time B ji , energy consumption C ji , and position accuracy D ji as the optimization motion indexes of the boom, establish the initial model of the boom control algorithm, and train the initial model of the boom control algorithm through the machine learning algorithm to obtain the final model Y = F(X) of the boom control algorithm;

[0041] Wherein, i represents the number of groups of boom hydraulic control parameters, i = 1, 2, 3... N, and N is a natural number; j represents the jth boom hydraulic cylinder; X i is the ith group of boom attitude parameters; Y i is the ith group of boom hydraulic control parameters; A ji , B ji , C ji , D ji are the amplitude, construction time, energy consumption, and position accuracy of the jth boom hydraulic cylinder in the ith group of boom hydraulic control parameters;

[0042] S12: Input the final model Y = F(X) of the boom control algorithm into the boom control system of the placing boom, and calculate the boom hydraulic control parameters of the current concrete placing boom:

[0043] Establish the final model Y = F(X) of the boom control algorithm in the boom control system of the placing boom, synchronously update the various parameters of the final model Y = F(X) of the boom control algorithm, and convert the cloth target position coordinates into the boom attitude parameters X e (θ lie , β jlie) and input it into the boom control system of the concrete placing boom. Calculate the boom hydraulic control parameters Y of the current concrete placing boom through the final model Y = F(X) of the boom control algorithm ie (q jie ,v jie ,p jie ,Δl jie );

[0044] θ ji represents the azimuth angle after the movement of the j-th boom in the i-th set of boom hydraulic control parameters;

[0045] β ji represents the angle between the j-th boom and the adjacent (j - 1)-th boom after the movement of the j-th boom in the i-th set of boom hydraulic control parameters;

[0046] q jie ,v jie ,p jie ,Δl jie respectively represent the hydraulic oil flow rate, flow velocity, oil pressure and cylinder stroke of the j-th hydraulic cylinder in the i-th set of boom hydraulic control parameters;

[0047] S13: The boom control system of the concrete placing boom drives the boom to move to the concrete placing target position according to the boom hydraulic control parameters of the current concrete placing boom:

[0048] The boom control system of the concrete placing boom drives each boom to move according to the boom hydraulic control parameters Y io (q jio ,v jio ,p jio ,Δl jio ) respectively, so that it reaches the concrete placing target position while meeting one or more optimized motion indexes among the amplitude A ji , construction time B ji , energy consumption C ji , position accuracy D ji .

[0049] The intelligent control method for the boom of a concrete placing boom of the present invention is as follows. First, according to the training samples of the boom hydraulic control parameters randomly generated and the corresponding boom attitude parameters, using one or more of amplitude, construction time, energy consumption, and position accuracy as the optimized motion indicators of the boom, an initial model of the boom control algorithm is established. The initial model of the boom control algorithm is trained through a machine learning algorithm to obtain the final model of the boom control algorithm and input it into the boom control system of the placing boom. Finally, the cloth placement target position coordinates are converted into boom attitude parameters and input into the boom control system of the placing boom. The boom hydraulic control parameters of the current concrete placing boom are calculated through the boom control algorithm. The boom hydraulic control parameters at least include the hydraulic oil flow rate, flow velocity, oil pressure, and cylinder stroke of the hydraulic cylinder. The boom control system of the placing boom drives the boom to move to the cloth placement target position according to the boom hydraulic control parameters of the current concrete placing boom. In the intelligent control method for the boom of a concrete placing boom of the present invention, the training samples of the boom hydraulic control parameters, boom attitude parameters, and optimized motion indicators are used to train the boom control algorithm through a machine learning algorithm. The boom hydraulic control parameters of the current concrete placing boom can be calculated through the boom control algorithm for the cloth placement target position, enabling the boom control system of the placing boom to accurately control and automatically drive the boom to move to the cloth placement target position, thereby improving the control accuracy of the boom of the concrete placing boom and achieving a vibration suppression effect by reducing the boom vibration amplitude.

[0050] In the step S11, the steps of the training samples of the boom hydraulic control parameters and the boom attitude parameters are as follows:

[0051] S1: Randomly generate N groups of boom hydraulic control parameters:

[0052] For example: Randomly generate the i-th group of boom hydraulic control parameters Y i (q ji , v ji , p ji , Δl ji ; θ 0i , β j0i );

[0053] Among them, q ji , v ji , p ji , Δl ji respectively represent the hydraulic oil flow rate, flow velocity, oil pressure, and cylinder stroke of the j-th hydraulic cylinder in the i-th group of boom hydraulic control parameters; θ 0i represents the azimuth angle of the j-th boom before movement in the i-th group of boom hydraulic control parameters, and β j0i represents the angle between the j-th boom and the (j - 1)-th boom before movement in the i-th group of boom hydraulic control parameters;

[0054] S2: Input the N groups of boom hydraulic control parameters into the boom control system of the concrete placing boom respectively to drive the boom to move, and monitor the boom attitude parameters after the corresponding movement of the boom under each group of boom hydraulic control parameters respectively;

[0055] Starting from the initial state, input the i-th group of boom hydraulic control parameters into the boom control system of the concrete placing boom in sequence. The boom control system of the concrete placing boom issues a control signal to drive the hydraulic cylinder, so that the boom realizes the corresponding movement. After the relative movement of the boom driven by the i-th group of boom hydraulic control parameters is completed, the i-th group of boom attitude parameters X i (θ 1i , β j1i ; A ji , B ji , C ji , D ji ) are monitored through monitoring devices such as displacement sensors and angle sensors.

[0056] Among them, θ 1i represents the azimuth angle after the movement of the j-th boom under the i-th group of boom hydraulic control parameters;

[0057] β j1i represents the included angle between the j-th boom and the (j - 1)-th boom after the movement under the i-th group of boom hydraulic control parameters;

[0058] A ji , B ji , C ji , D ji are respectively the amplitude, construction time, energy consumption and position accuracy of the hydraulic cylinder of the j-th boom under the i-th group of boom hydraulic control parameters. The boom attitude parameters are selected according to needs from one or more of the above optimized motion indexes;

[0059] S3: Generate training samples: Combine the i-th group of boom attitude parameters X i with the i-th group of boom hydraulic control parameters Y i to form a new data element (X i , Y i ). Generate training samples {(X i , Y i ), i = 1, 2, 3... N} from the N groups of data elements (X i , Y i ).

[0060] The method of the above training samples of boom hydraulic control parameters and boom attitude parameters is only an example and is not limited to this.

[0061] Generally, the coordinates of the concrete placing target position are three-dimensional rectangular coordinates. To facilitate the operation of the boom control system of the concrete placing boom, it is necessary to convert the three-dimensional rectangular coordinates of the concrete placing target position into boom attitude parameters.

[0062] Assume that the three-dimensional rectangular coordinates of a spatial point are M(x, y, z), and convert it into three-dimensional cylindrical coordinates with the center of rotation of the concrete distributor as the origin O(x 0 , y 0 , z 0 ). The conversion principle is as follows:

[0063]

[0064] Among them, θ is the angle rotated counterclockwise from the x-axis to OM;

[0065] ρ is the distance from the origin O to the projection of the point M(x, y, z) on the plane;

[0066] h is the height difference between the concrete distribution point M and the origin of the rectangular coordinate system;

[0067] The method of converting the coordinates of the concrete distribution target position into the boom attitude parameters in step S12 is as follows: According to the operation of vectors, obtain the boom attitude function represented by a series of spatial points, and determine the boom attitude parameters X e (θ jie , β jie ) through the conversion of the boom attitude function:

[0068]

[0069] Moreover:

[0070]

[0071] Among them, R j is the length of the j-th boom of the concrete distributor;

[0072] ɑ j is the angle of the j-th boom relative to the positive direction of the ρ-axis or the horizontal plane;

[0073] β i is the rotation angle of the j-th boom relative to the (j - 1)-th boom;

[0074] θ 1ie represents the azimuth angle of the j-th boom after movement;

[0075] β j1ie represents the angle between the j-th boom and the (j - 1)-th boom after movement.

[0076] K is the number of booms of the concrete distributor. In this embodiment, K = 3.

[0077] The machine learning algorithm adopted in this embodiment can adopt genetic algorithm, neural network algorithm, etc.

[0078] The above description is only a description of the preferred embodiments of the present invention and does not limit the scope of the present invention in any way. Any changes or modifications made by those of ordinary skill in the art of the present invention based on the above disclosure fall within the scope of the claims.

Claims

1. An intelligent control method for the boom of a concrete placing boom, characterized in that, the steps are as follows: S11: Through a machine learning algorithm, train the training sample data of randomly generated boom hydraulic control parameters and their corresponding boom attitude parameters to obtain a boom control algorithm, establish an initial model of the boom control algorithm for the concrete placing boom, and use one or more of amplitude, construction time, energy consumption, and position accuracy as optimized motion indicators to train the initial model of the boom control algorithm for the concrete placing boom to obtain the final model of the boom control algorithm; S12: Establish the final model of the boom control algorithm in the boom control system of the placing boom, convert the placing target position coordinates into boom attitude parameters and input them into the boom control system of the placing boom, and calculate the boom hydraulic control parameter Y of the current concrete placing boom through the final model of the boom control algorithm ie (q jie , v jie , p jie , Δl jie ); Y ie represents the boom hydraulic control parameters of the current concrete placing boom; θ ji represents the azimuth angle after the movement of the j-th boom section among the hydraulic control parameters of the i-th group of booms; β ji denotes the angle between the j-th boom after movement and the adjacent (j - 1)-th boom among the hydraulic control parameters of the i-th group of booms. q jie ,v jie ,p jie， Δl jie respectively represent the hydraulic oil flow rate, flow velocity, oil pressure, and cylinder stroke of the j-th hydraulic cylinder in the i-th boom hydraulic control parameters; i represents the number of groups of boom hydraulic control parameters, i = 1, 2, 3... N, and N is a natural number; j represents the jth boom hydraulic cylinder; S13: The boom control system of the placing boom drives each boom to move according to the boom hydraulic control parameters Y i (q ji ,v ji ,p ji ,Δl ji ) respectively, so that it reaches the placing target position while meeting one or more optimized motion indexes of amplitude, construction time, energy consumption, and position accuracy; Y i represents the hydraulic control parameters of the i-th boom.

2. The intelligent control method for the boom of a concrete placing boom according to claim 1, characterized in that: In the step S11, the steps of the training samples of the boom attitude parameters and the boom hydraulic control parameters are as follows: S1: Randomly generate N groups of boom hydraulic control parameters: The hydraulic control parameters Y of the i-th boom i (q ji , v ji , p ji , Δl ji ; θ 0i , β j0i ); q ji ,v ji ,p ji， Δl ji respectively represent the hydraulic oil flow rate, flow velocity, oil pressure and cylinder stroke of the j-th hydraulic cylinder in the i-th boom hydraulic control parameters; θ 0i represents the azimuth angle of the j-th boom before movement in the i-th group of boom hydraulic control parameters; β j0i denotes the angle between the j-th boom before movement and the adjacent upper boom among the hydraulic control parameters of the i-th boom group; S2: Input the N groups of boom hydraulic control parameters into the boom control system of the concrete placing boom to drive the boom to move, and respectively monitor the boom attitude parameters after the corresponding deformation of each group of boom hydraulic control parameters: Input the i-th set of boom hydraulic control parameters into the boom control system of the concrete placer in sequence. The boom control system of the concrete placer issues control signals to drive the hydraulic cylinders, enabling the boom to achieve corresponding movements. After the i-th set of boom hydraulic control parameters drive the boom to complete the movement, monitor the attitude parameters X of the deformed i-th set of booms i (θ 1i， β j1i ; A ji， B ji， C ji， D ji ); Among them, θ 1i represents the azimuth angle after the movement of the j-th boom section in the i-th group of boom hydraulic control parameters; β j1i denotes the included angle between the j-th boom after movement and the adjacent upper boom among the hydraulic control parameters of the i-th boom group; A ji， B ji， C ji， D ji respectively represent the motion indexes of the j-th boom section in the i-th group of boom hydraulic control parameters. A ji is the amplitude, B ji is the construction time, C ji is the energy consumption, D ji is the position accuracy. The boom attitude parameters are selected from one or more of the above motion indexes as required; S3: Generate training samples: Combine the i-th set of boom attitude parameters X i with the i-th set of boom hydraulic control parameters Y i to form a new data element (X i , Y i ). Generate training samples {(X i , Y i ), i = 1, 2, 3... N} from N sets of data elements (X i , Y i ).

3. The intelligent control method for the boom of a concrete placing boom according to claim 1, characterized in that , in the step S12, the method for converting the coordinate of the target position of the fabric into the boom attitude parameter is as follows: obtain the boom attitude function represented by a series of spatial points, and determine the boom attitude parameter X through the conversion of the boom attitude function e (θ jie， β jie ) wherein: Among them, R j is the length of the j-th boom of the concrete placing boom; ɑ j is the angle of the j-th boom relative to the positive direction of the ρ-axis or the horizontal plane; β i is the rotation angle of the j-th boom section relative to the (j - 1)-th boom section; θ 1ie represents the azimuth angle after the movement of the j-th boom section; β j1ie represents the included angle between the j-th boom section after movement and the (j - 1)-th boom section; K is the number of boom sections of the concrete placing boom.

Images