CN101840201A - Genetic optimization controlling technology of piler - Google Patents
Genetic optimization controlling technology of piler Download PDFInfo
- Publication number
- CN101840201A CN101840201A CN201010136402A CN201010136402A CN101840201A CN 101840201 A CN101840201 A CN 101840201A CN 201010136402 A CN201010136402 A CN 201010136402A CN 201010136402 A CN201010136402 A CN 201010136402A CN 101840201 A CN101840201 A CN 101840201A
- Authority
- CN
- China
- Prior art keywords
- piler
- speed
- neural network
- optimization
- control
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention relates to a genetic optimal controlling technology of a piler, which is used for performing speed optimization control on the operation process of a piler by means of a neutral network and a genetic algorithm so as to reduce vibration produced in the high-speed operation process of the piler. A neutral network mapping model between the vibration state of the piler and main operating parameters is created on the basis of the main operating parameters in the operation process of the piler, and vibration data produced in the operation process of the piler are used; the operating speed optimization for minimum vibration in the operating process of the piler is carried out on the basis of the created neutral network mapping model by using the genetic algorithm; and an operating speed control curve of the piler is obtained on the basis of the piler operating speed obtained through optimization, and the speed control on the piler is realized by using the numeric conversion of a programmable logic controller and a variable-frequency controller.
Description
Technical field:
The invention belongs to piler control association area, particularly a kind of genetic optimization controlling technology at the piler operational process.
Technical background:
Because automatic stereowarehouse has space availability ratio height, automaticity advantages of higher, its range of application is more and more wider.As the piler of key equipment in the three-dimensional warehouse, its control technology is for very crucial to the stack machine performance.Along with the widespread use of technology such as programmable logic controller (PLC) (PLC), frequency control, laser ranging in piler control field, the piler travelling speed is significantly improved.But the raising of piler travelling speed has brought problems such as vibration, thereby need be optimized the vibration that reduces piler to the speed of piler operational process.
Carrying out some research work aspect the piler control technology at present.For example people such as Chen Juan and Zhong Yongyan has inquired into the hardware and software implementation of piler control system at the characteristics of piler control in the automatic stereowarehouse.People such as Wang Yongjun and all rare talent with laser ranging technique as lane stacker recognize the location means, use frequency converter stepless speed regulation close-loop control mode, a kind of control method of novel lane stacker high-speed cruising is studied.People such as Lv Quanhai and Shen Minde has inquired into the realization technology of the hardware and software of piler closed-loop speed control in conjunction with laser ranging technique.People such as literary composition group and Miu Xingfeng have not designed and have been suitable for the PLC control system that piler walking, lifting and retractable fork mechanism use.Ge Gaofeng analyzes composition, structure and the characteristics of piler control system, and has inquired into the piler control system scheme that adopts S7-300PLC.People such as the cleer and peaceful Yan Shu of Luo Zhi field are applied to the piler of mailbag tiered warehouse facility with fuzzy control and PREDICTIVE CONTROL, have studied the intelligent predicting fuzzy control method, and have carried out emulation according to the actual condition of piler motion.Utilize neural network and genetic Optimization Algorithm to carry out the piler travelling speed at present and control and reduce piler vibration, the research work of this respect yet there are no the open source literature report.
The present invention is the genetic optimization controlling technology of relevant piler.The piler control new technology that the present invention proposes is to be means, to be target to reduce the vibratory output that produces in the piler high-speed cruising process with genetic algorithm and neural network.Utilize neural network to set up mapping model between piler vibrational state and the piler operational factor, utilize genetic algorithm to carry out the optimization of piler minimum vibration state travelling speed; Carry out the control of piler operational process on this basis, thereby realize reducing the target of piler operational process Oscillation Amplitude.The piler genetic optimization control new technology that the present invention proposes adopt neural network and genetic algorithm to realize, thereby the control new technology that the present invention proposes belongs to the piler intelligence control method.The piler genetic optimization controlling technology that the present invention proposes can reduce the vibratory output that produces in the piler high-speed cruising process, for efficient, the safe operation of guaranteeing piler provides effective control device.
Summary of the invention:
The objective of the invention is to be means, to be target, carry out the speed-optimization control of piler operational process to reduce the vibratory output that produces in the piler high-speed cruising process with neural network and genetic algorithm.In order to reach above-mentioned target, the technical solution used in the present invention is: based on the vibration data that produces in key operation parameter in the piler operational process and the piler operational process, set up the neural network mapping model between piler vibrational state and the main operational factor; Neural network model to set up utilizes genetic algorithm to carry out the travelling speed optimization of vibratory output minimum in the piler operation process; The piler travelling speed that obtains with optimization is the speed control curve that the basis obtains the piler operation, utilizes the speed control of the numerical value conversion realization piler of programmable logic controller (PLC) and frequency-variable controller.
The present invention includes piler operational process sample data and set up, make up contents such as neural network model, the optimization of piler travelling speed, piler speed control.The concrete steps that the present invention includes are as follows:
1) sets up piler operational process sample data
At first analyze the immediate cause that vibration produces in the piler operational process, determine on this basis and relevant major influence factors piler type, operator's console height, load-carrying, the travelling speed of piler vibration.According to orthogonal test method, carry out the experiment of influence relation between piler type, operator's console height, load-carrying, travelling speed and the vibration of piler operational process, obtain the mass data of influence relation between parameters such as reflecting relevant piler type and the piler vibration.Piler service data with acquisition is a base configuration piler operational process sample data.The present invention is the output of piler vibratory output as sample data, and parameters such as piler type, operator's console height, load-carrying, travelling speed are formed sample data as the input of sample data.This sample data has reflected the relation between piler vibrational state and relevant piler travelling speed, piler type, operator's console height, the load-carrying.
2) make up neural network model
Piler operational process sample data with foundation is the structure that neural network model is carried out on the basis.Piler operational process sample data has reflected the relation between piler vibrational state and the parameters such as piler travelling speed, piler type, therefore utilize multilayer neural network or radial basis function neural network to carry out sample learning, just piler vibratory output and piler parametric relationship that sample data contains can be described by neural network model.
Making up neural network model needs to determine earlier the structure and the attribute of neural network, comprises the number of plies of neural network, the implicit number of plies, every layer neuron number, every layer activation function setting and the input of this neural network model etc.Wherein neural network input layer number comprises piler type, operator's console height, pay load, travelling speed corresponding to the number of parameters of sample importation.Neural network output layer neuron number comprises the vibratory output of piler corresponding to the number of parameters of sample output.After determining neural network structure, sample data is processed into the requirement of satisfying the neural network learning needs, select suitable learning algorithm to carry out the study of neural network.After finishing the sample learning process of neural network, just can set up the neural network model of parametric relationships such as reactor stack machine vibratory output and piler travelling speed.
3) the piler travelling speed is optimized
Based on the neural network model between the parameters such as the piler vibratory output that makes up and piler travelling speed, piler type, utilizing genetic algorithm to carry out with the minimum vibration amount is the optimization of the piler travelling speed of optimization aim.In the genetic optimization process, the vibratory output that optimization aim relates to will utilize the neural network that makes up to calculate; Needs are carried out the input parameter of the velocity amplitude of piler type that travelling speed optimizes, operator's console height, pay load, current design variable as neural network, these data processing are become to satisfy the requirement of neural network prediction needs, the output of the neural network that can obtain; Output data to neural network is handled, and can calculate the piler vibratory output.Be optimized piler travelling speed preferred of design variable by genetic optimization, obtain the travelling speed of piler minimum vibration amount correspondence.
4) piler speed control
The travelling speed of the piler minimum vibration that obtains based on neural network prediction, the velocity amplitude curve fitting disperses, rate curve after the match is stored in the data block of programmable logic controller (PLC) of piler, for the speed control of piler ready.In the control of piler actual speed, programmable logic controller (PLC) is found out corresponding velocity amplitude according to the piler current location on rate curve after the match, by the frequency converter of analog quantity or fieldbus mode drive current vector, and drive the speed control that piler walking AC asynchronous motor is finished piler.
Advantage of the present invention: the genetic optimization controlling technology of piler, can carry out travelling speed optimization and control under the minimum vibration state in the piler operation process, reduce the vibratory output in the piler operational process.
Description of drawings:
Accompanying drawing 1 is a piler genetic optimization control flow chart;
Accompanying drawing 2 is piler genetic optimization model synoptic diagram;
Accompanying drawing 3 is piler genetic optimization control principle synoptic diagram
Embodiment:
Below in conjunction with accompanying drawing, the genetic optimization controlling technology at automatic warehouse stacker that the present invention proposes is described, its specific implementation method is as follows:
Fig. 1 is a piler evolutionary optimization control flow chart, the piler control new technology that the present invention proposes is to be means, to be target to reduce the vibratory output that produces in the piler high-speed cruising process with neural network and genetic algorithm, carries out the speed-optimization control of piler operational process.In order to reach above-mentioned target, the technical solution used in the present invention is: based on the vibration data that produces in key operation parameter in the piler operational process and the piler operational process, set up the neural network mapping model between piler vibrational state and the main operational factor; Neural network model to set up utilizes genetic algorithm to carry out the travelling speed optimization of vibratory output minimum in the piler operation process; The piler travelling speed that obtains with optimization is the speed control curve that the basis obtains the piler operation, utilizes the speed control of the numerical value conversion realization piler of programmable logic controller (PLC) and frequency-variable controller.
Fig. 2 is a piler evolutionary optimization model synoptic diagram, and the piler genetic optimization model among the present invention comprises two parts: neural network technology and genetic algorithm optimization technology.Neural network model to set up among the present invention utilizes genetic algorithm to carry out the travelling speed optimization of vibratory output minimum in the piler operation process.
1. the present invention utilizes nerual network technique to carry out the prediction of piler travelling speed, and the neural network of structure is the BP neural network model with three-decker, and detailed is described below:
1) determines input/output variable
In the three-dimensional Automatic Warehouse, normally goods is placed on the operating platform of piler, transport by the mobile of piler, in the goods operational process, because the starting of piler, acceleration, at the uniform velocity, slowing down causes the nuisance vibration of piler body easily up to the compound motion that stops etc., influenced piler and moved safely and effectively.These factors mainly comprise the travelling speed of the current load-carrying of type, the height of piler shelf, the piler of piler, piler each section on working line.Therefore, can adopt the input variable of several variablees such as type, shelf height, dead weight capacity, each section velocity amplitude of piler as piler neural network Based Intelligent Control model, and with the travelling speed of piler each section on the working line output variable as piler neural network Based Intelligent Control model.
2) network structure is described
In this piler neural network control technique, if we are divided into N part as required with the one-stop operation running route of piler, the neuron number that input layer comprised of neural network model is N+3 in this invention so, respectively velocity amplitude (the V of each section on the height of the type of corresponding piler, operator's console, current load-carrying and the piler one-stop operation working line
1, V
2... V
N); This neural network model output node number is N, the velocity amplitude (V of each section on the corresponding piler one-stop operation route
1, V
2... V
N); Middle one deck is the hidden layer of neural network model, and the neuronic number that it comprised is
Individual, wherein t represents any one integer between 0 to 10.In neural network model of the present invention, each neuron is realized connecting between input layer, middle layer (hidden layer) and output layer, the levels, does not connect with having between the layer, and the activation function of hidden layer all adopts Sigmoid function (S type function), and function representation is as follows:
In the formula, a is the Slope Parameters of Sigmoid function, by changing parameter a, can obtain the Sigmoid function of Different Slope.
3) network training
The training sample of network comes by random acquisition in the piler actual job process, and neural network adopts 3 layers neural network model, N input neuron of structure employing,
Individual hidden layer node (t represents any one integer between 0 to 10) and N output node.The process of neural metwork training is the process of a study just, at first input information is successively handled and is calculated the real output value of each unit through hidden layer by input layer, according to the difference of desired output, begin successively to regulate weights then by output terminal with actual output.
2. to utilize genetic algorithm to carry out with the minimum vibration amount be the optimization of the piler travelling speed of optimization aim in the present invention:
1) design optimization variable
In the operation process of piler, the type T of the main and piler of the speed of piler, the height of shelf is H in the operation process, current pay load is that G is relevant, so optimization variable can be designated as:
V
i＝f(T，H
i，G
i)
In the following formula, V
iThe expression piler is at the travelling speed in the sub-highway section of i, H
iBe illustrated in the shelf height in the sub-highway section of i, G
iBe illustrated in the sub-highway section of the i pay load of piler constantly.I=1,2 ... N, N by in the piler one-stop operation as required the number in the sub-highway section of division.
2) design optimization objective function
At the operation process and the principle of work of piler in the automatic stereowarehouse, the type of establishing piler is T, and the height of shelf is H in the operation process, and current pay load is G, and the vibratory output in each sub-highway section in the piler one-stop operation process is respectively A
1, A
2... A
n(n=1,2......) influencing each factor of piler vibratory output during in conjunction with the piler operation, the design variable of design piler operating speed optimization mainly comprises: the type T of piler, the height H of shelf, pay load G, the synthesis oscillation amount A of piler one-stop operation.Wherein the synthesis oscillation amount A of piler one-stop operation is calculated as follows:
A＝(a
1*A
1+a
2*A
2+…+a
N*A
N)/N
A wherein
1, a
2A
NBe respectively the weighting coefficient of the vibratory output in each sub-highway section in the piler one-stop operation, N=1,2 ..., be the number in the sub-highway section divided as required.
3) design optimization constraint function
In the operation process of piler, the speed of piler can not surpass the maximum safe speed of regulation, and the body vibration of piler should be minimum in the operation process simultaneously, can determine the constraint function of piler intelligent optimization according to this.
V
min＜V
i＜V
max
minA
i＝g(V
i，T，H
i，G
i)
4) computation optimization
The intelligent optimization that the present invention carries out piler with genetic algorithm calculates, and in the optimizing process of piler, by the operation of genetic algorithm piler is carried out accurate optimization and calculates, and finally obtains the accurate optimization result.
Fig. 3 is the control principle synoptic diagram of piler neural network control, and the ANN (Artificial Neural Network) Control algorithm of piler is mainly finished by Programmable Logic Controller PLC among the present invention.Training prediction piler by neural network travelling speed value of every section on the single operation path, what this functional block was exported is the speed control value of day part, these velocity amplitudes are some discrete points, how to be directly used in control, because each point is discontinuous, can cause certain impact to piler, operates steadily in order to make piler, here by these discrete points being fitted to the operation control curve of piler, send frequency-variable controller as speed setting value with the each point of getting on this curve.
The present invention adopts closed-loop control to speed, and the pace of change of the piler horizontal level that the speed of piler can detect by laser range finder is calculated.The D/A module of speed control amount after the computing by PLC is converted into current signal and is used for the control of conversion device, walking among the present invention, hoist and the pallet fork motor is realized stepless time adjustments by two frequency converter timesharing control respectively, frequency converter is again by changing supply frequency control rotating speed of motor.
Claims (3)
1. piler genetic optimization controlling technology is characterized in that being means, being target to reduce the vibratory output that produces in the piler high-speed cruising process with genetic algorithm and neural network, carries out the speed control of piler operational process, may further comprise the steps:
Set up piler operational process sample data;
The genetic optimization of piler operational process speed;
Make up neural network model;
The prediction of piler travelling speed;
The piler speed control.
2. piler genetic optimization controlling technology according to claim 1 is characterized in that: the genetic optimization of described piler operational process velocity amplitude utilizes genetic algorithm that sample data is optimized and realizes.
3. piler genetic optimization controlling technology according to claim 1, it is characterized in that: the prediction of described piler travelling speed value is on the basis of piler operational process velocity amplitude genetic optimization, realizes by setting up neural network model and network being trained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010136402A CN101840201A (en) | 2010-03-30 | 2010-03-30 | Genetic optimization controlling technology of piler |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010136402A CN101840201A (en) | 2010-03-30 | 2010-03-30 | Genetic optimization controlling technology of piler |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101840201A true CN101840201A (en) | 2010-09-22 |
Family
ID=42743624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010136402A Pending CN101840201A (en) | 2010-03-30 | 2010-03-30 | Genetic optimization controlling technology of piler |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101840201A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011120200A1 (en) * | 2010-04-01 | 2011-10-06 | 江苏六维物流设备实业有限公司 | Genetic optimization control technology for stacking machines |
| CN106005952A (en) * | 2016-05-26 | 2016-10-12 | 淮南市鸿裕工业产品设计有限公司 | Overload preventing device of stacking machine |
| CN117075647A (en) * | 2023-10-17 | 2023-11-17 | 合肥焕智科技有限公司 | Control method and device for stacker |
-
2010
- 2010-03-30 CN CN201010136402A patent/CN101840201A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2011120200A1 (en) * | 2010-04-01 | 2011-10-06 | 江苏六维物流设备实业有限公司 | Genetic optimization control technology for stacking machines |
| CN106005952A (en) * | 2016-05-26 | 2016-10-12 | 淮南市鸿裕工业产品设计有限公司 | Overload preventing device of stacking machine |
| CN117075647A (en) * | 2023-10-17 | 2023-11-17 | 合肥焕智科技有限公司 | Control method and device for stacker |
| CN117075647B (en) * | 2023-10-17 | 2024-01-12 | 合肥焕智科技有限公司 | Control method and device for stacker |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108140157B (en) | Computer-implemented process and system for evaluating and adjusting industrial vehicle performance | |
| CN102107819B (en) | Anti-shaking control method for container shore bridge hanger | |
| Yi et al. | Anti-swing and positioning control of overhead traveling crane | |
| CN110316657A (en) | A kind of anti-swing control system and its control method of heavy object of crane | |
| Fernández et al. | Dynamic fuzzy logic elevator group control system with relative waiting time consideration | |
| CN101024471B (en) | Bridge-type crane control apparatus with rocking-prevention function | |
| CN101840201A (en) | Genetic optimization controlling technology of piler | |
| CN102120545A (en) | Crane anti-swing system | |
| CN113156980A (en) | Tower crane path planning method and system based on deep reinforcement learning | |
| Lee et al. | Autonomous construction hoist system based on deep reinforcement learning in high-rise building construction | |
| US20240053752A1 (en) | Motion Control of a Motion Device | |
| CN110228754A (en) | A kind of industrial overhead crane control method and system of adaptive speed planning | |
| Sun et al. | The Structure Optimization of Main Beam for Bridge Crane Based on An Improved PSO. | |
| CN110589684A (en) | Anti-swing driving control method for crane | |
| JPH0122198B2 (en) | ||
| CN101846974A (en) | Piler neural network control technique | |
| Yi et al. | Anti-swing fuzzy control of overhead traveling crane | |
| JP2512854B2 (en) | Control system for the cavern lane | |
| WO2023248529A1 (en) | Information processing system, operation planning method, and program | |
| CN114803866B (en) | Staged optimization control method and device for lifting motion state of intelligent tower crane | |
| CN110950241A (en) | Electronic anti-swing method of intelligent crane | |
| WO2011120200A1 (en) | Genetic optimization control technology for stacking machines | |
| CN101846973A (en) | Piler adaptive fuzzy optimal control method | |
| Burkhardt et al. | A graph-based path planning algorithm for the control of tower cranes | |
| WO2011120201A1 (en) | Piler neural network control technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100922 |