CN102540886A - Self-adaptive control method and system for operation power of pipe-laying trencher - Google Patents
Self-adaptive control method and system for operation power of pipe-laying trencher Download PDFInfo
- Publication number
- CN102540886A CN102540886A CN2011104420982A CN201110442098A CN102540886A CN 102540886 A CN102540886 A CN 102540886A CN 2011104420982 A CN2011104420982 A CN 2011104420982A CN 201110442098 A CN201110442098 A CN 201110442098A CN 102540886 A CN102540886 A CN 102540886A
- Authority
- CN
- China
- Prior art keywords
- pipe
- pace
- laying trencher
- depth
- neural network
- 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.)
- Granted
Links
Images
Landscapes
- Excavating Of Shafts Or Tunnels (AREA)
Abstract
The invention relates to a self-adaptive control method for operation power of a pipe-laying trencher, which includes: firstly, determining trenching slope and depth of the pipe-laying trencher according to the construction requirements; secondly, building a nerve network model, testing revolution speed of a cutter chain and advancing speed and excavation power of the pipe-laying trencher on the condition with set trenching depth, then changing the trenching depth, the revolution speed of the cutter chain and the advancing speed of the pipe-laying trencher continuously and testing the corresponding excavation power, synthesizing test data in a certain quality, and building a relation model of the trenching depth, the revolution speed of the cutter chain, the advancing speed of the pipe-laying trencher and the excavation power by the nerve network algorithm; thirdly, optimizing by the genetic algorithm, determining the matching value of the revolution speed of the cutter chain and the advancing speed of the pipe-laying trencher, which is close to the set designed excavation power to the utmost extent, on the condition of linear variation of the trenching depth, and extracting a group of optimum values according to the precedence principle of the advancing speed of the pipe-laying trencher; and fourthly, controlling the revolution speed of the cutter chain and the advancing speed of the pipe-laying trencher according to the optimum matching values.
Description
Technical field
The present invention is directed to agricultural machinery equipment technical field of automatic control, relate to the system and method for pipe-laying trencher operation power adaptive control.
Background technology
Pipe-laying trencher is one type of large-scale Work machine that is used for farmland water drainage row's alkali, reasonably dispatches and uses the operation power of pipe-laying trencher to help giving full play to the usefulness of pipe-laying trencher, improves the reliability and the economy of facility.
Generally speaking, the power that the pipe-laying trencher ditching operation consumed is closely related with the soil property structural environment of builder's yard, the trench digging degree of depth (or referring to penetrating distance), cutter chain rotating speed (or referring to the speed of fetching earth), the depth of cut (that is pace) etc.Hard like soil property, the trench digging degree of depth is big, cutter chain rotating speed is fast and depth of cut when big, power demand is big, on the contrary power demand is little.Generally, the soil property structural environment is complicated, and the trench digging degree of depth changes along with gradient constantly, and the cutter chain rotating speed and the depth of cut are controlled by manual work, make that ditching operation power is complicated and changeable.
Traditional ditching operation is the trench digging degree of depth and the gradient by construction requirement; According to soil property structural environment, cutter chain part intensity and pipe-laying trencher power performance etc.; The artificial observation control cutter chain rotating speed and the depth of cut can't guarantee that pipe-laying trencher actual job power is consistent with DESIGN THEORY operation power.So scenarios often occurs: the cutter chain rotating speed and the depth of cut are less than normal, cause the pipe-laying trencher power dissipation; Or cutter chain rotating speed and the depth of cut are excessive, cause pipe-laying trencher power under-supply, and tool setting chain part and associated mechanical system form injury.
In order to guarantee that actual job power is as far as possible near theoretical operation power; Make engine output remain on optimum level; Seek a kind of intelligentized automatic control mode; Making the trench digging degree of depth, cutter chain rotating speed and the depth of cut reasonably mate, is that pipe-laying trencher has a thorny difficult problem to be solved for a long time.
Summary of the invention
For addressing the above problem, the present invention discloses a kind of pipe-laying trencher operation power adaptive control method, it is characterized in that, comprising:
Step 200; Set up neural network model, under given trench digging depth conditions, detect cutter chain speed of gyration and pipe-laying trencher pace and excavate power; And continuous the trench digging degree of depth, cutter chain speed of gyration and the pipe-laying trencher pace of changing; Detect the corresponding power that excavates, the detection data of comprehensive some are utilized the neural network algorithm foundation trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and are excavated the power relation model;
Step 300; The optimizing of utilization genetic algorithm; Under the linear change condition of the trench digging degree of depth, confirm to excavate the cutter chain speed of gyration of power and the matching value of pipe-laying trencher pace at utmost to approach given expectation, get wherein one group of optimal value by pipe-laying trencher pace priority principle;
Step 400 is by Optimum Matching value control cutter chain speed of gyration and pipe-laying trencher pace.
Described operation power adaptive control method is characterized in that said step 100 also comprises:
Step 110 uses generating laser to be used for to laser pickoff the reference laser plane of setting gradient being provided as light source; Said laser pickoff is used to measure the high and low position at self relative laser plane place, and the elevation signal is sent to said data processing and control center.
Described operation power adaptive control method is characterized in that said step 110 also comprises:
Step 111 is obtained the attitude information of current trench digging arm;
Step 112 sends control information to power transmission shaft and walking CD-ROM drive motor;
Step 113, power transmission shaft motion association cutter chain speed of gyration and operation power obtain the rotating speed and the torque information (input speed of power transmission shaft is adjusted by hydraulic speed regulating device) of power transmission shaft;
Step 114, walking CD-ROM drive motor motion association pipe-laying trencher pace is obtained pace information;
Step 115 is sent to data processing and control center with the rotating speed and the torque information of power transmission shaft through radio transmitting device, simultaneously pace information is sent to data processing and control center.
Described operation power adaptive control method is characterized in that said step 200 also comprises:
Step 210 is selected and is designed neural network, forms neural network structure;
Step 220 forms the training sample set of said neural network;
Step 230 is trained neural network.
Described operation power adaptive control method is characterized in that step 300 also comprises:
Step 310 is carried out the computing of chromosomal coding, coding/decoding method;
Step 320 is set up objective function;
Step 330 is carried out the design of fitness function;
Step 340 is selected, the design of intersection and mutation operator.
The present invention also discloses a kind of pipe-laying trencher operation power adaptive control system, it is characterized in that, comprising:
Laser system is used for to pipe-laying trencher trench digging gradient reference surface and trench digging depth monitoring being provided according to construction requirement;
Data processing and control center are used for signals collecting, data computation, and control algolithm etc., and CD-ROM drive motor and depth adjustment hydraulic cylinder sent control signal;
Neural network module; Be used to set up neural network model, under given trench digging depth conditions, detect cutter chain speed of gyration and pipe-laying trencher pace and excavate power; And continuous the trench digging degree of depth, cutter chain speed of gyration and the pipe-laying trencher pace of changing; Detect the corresponding power that excavates, the detection data of comprehensive some are utilized the neural network algorithm foundation trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and are excavated the power relation model;
Genetic algorithm module; Be used to use the genetic algorithm optimizing; Under the linear change condition of the trench digging degree of depth; Confirm to excavate the cutter chain speed of gyration of power and the matching value of pipe-laying trencher pace, get wherein one group of optimal value by pipe-laying trencher pace priority principle at utmost to approach given expectation; By Optimum Matching value control cutter chain speed of gyration and pipe-laying trencher pace.
Described operation power adaptive control system is characterized in that, said laser system also is used to use generating laser to be used for to laser pickoff the reference laser plane of setting gradient being provided as light source; Said laser pickoff is used to measure the high and low position at self relative laser plane place, and the elevation signal is sent to said data processing and control center.
Described operation power adaptive control system is characterized in that, also comprises:
Obliquity sensor is used to measure the attitude information of current trench digging arm;
Control and driving module is used for sending control information to power transmission shaft and walking CD-ROM drive motor;
Rotating speed and torque sensor are used to obtain the rotating speed and the torque information of power transmission shaft;
Tachogenerator is used to obtain pipe-laying trencher pace information;
Wireless receiving module is used for rotating speed and torque information to data processing and control center's real-time Transmission power transmission shaft;
Wherein, the input speed of power transmission shaft is adjusted by hydraulic speed regulating device.
Described operation power adaptive control system is characterized in that neural network module also comprises:
Network type confirms and the structure choice unit, is used for neural network is selected and designed, and forms neural network structure;
The training sample unit is used to form the training sample set of said neural network;
The neural metwork training unit is used for neural network is trained.
Described operation power adaptive control system is characterized in that genetic algorithm module also comprises:
The chromosome unit is used to carry out the computing of chromosomal coding, coding/decoding method;
The objective function unit is used to set up objective function;
The fitness function unit is used to carry out the design of fitness function;
The operator design cell is used to select, the design of intersection and mutation operator.
The present invention also discloses a kind of pipe-laying trencher, it is characterized in that, comprises above-mentioned described system.
The pipe-laying trencher operation power adaptive control system of embodiment of the present invention; Has following effect: according to the determined theoretical operation power of pipe-laying trencher overall design; Under the prerequisite of trench digging degree of depth linear transformation; The pace that speed of gyration that preset cutter chain part is allowed and pipe-laying trencher make every effort to reach; Set up model and be optimized through neural network and genetic algorithm and find the solution, draw the Optimum Matching of cutter chain speed of gyration and pipe-laying trencher pace, the input speed of FEEDBACK CONTROL adjustment power transmission shaft and the running speed of walking CD-ROM drive motor; Make it to bring into play the maximum trench digging power that engine distributes, thereby reach the maximum efficiency of pipe-laying trencher ditching operation.
The current operation power of the demonstration pipe-laying trencher of the running speed of monitoring walking CD-ROM drive motor and power transmission shaft, and measurement in real time.
Adopt neural network and genetic algorithm that operation power is optimized decision-making.In the construction operation process, through modeling with find the solution and obtain the two optimum matching point of cutter chain speed of gyration and pipe-laying trencher pace fast, and implement FEEDBACK CONTROL, reach the optimum performance of pipe-laying trencher operation power.
Description of drawings
Fig. 1 forms synoptic diagram for pipe-laying trencher operation power adaptive control system of the present invention;
Fig. 2 is a pipe-laying trencher operation power adaptive control system theory diagram of the present invention;
Fig. 3 is a pipe-laying trencher operation power adaptive control method process flow diagram of the present invention;
Fig. 4 is for the present invention is based on neural network and genetic algorithm process flow diagram;
Fig. 5 is a neural network model of the present invention.
Fig. 6 is the computing synoptic diagram of crossover operator of the present invention;
Fig. 7 is the computing synoptic diagram of the present invention's " variation " operator.
Embodiment
Provide embodiment of the present invention below, the present invention is made further description in conjunction with accompanying drawing.
In one embodiment of this invention, a kind of pipe-laying trencher operation power adaptive control system as shown in Figure 1 comprises by generating laser 1 laser pickoff 2; The walking CD-ROM drive motor 3 of band tachogenerator, hydraulic speed regulating device 4, data processing and control center 5; The rotating speed of band wireless transmission and the power transmission shaft 6 of torque sensor, depth adjustment hydraulic cylinder 7, sounding rod 8; Power transmission box 9, obliquity sensor 10, trench digging arm 11, cutter chain combination 12 etc.Wherein, Walking CD-ROM drive motor 3, the obliquity sensor 10 of said laser pickoff 2, band tachogenerator are connected to said data processing and control center 5 respectively, and rotating speed on the power transmission shaft 6 and torque sensor are sent to signal through self radio transmitting device the wireless receiving module of data processing and control center 5.
The main actuating unit of adaptive control system is depth adjustment hydraulic cylinder 7, hydraulic speed regulating device 4 and walking CD-ROM drive motor 3, through adjusting its three elements, can change the trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace.
Pipe-laying trencher during ditching operation, at first according to the engineering construction requirement, is confirmed the irrigation canals and ditches gradient and the initial trench digging degree of depth through laser system in the farmland; Then to set the linear embedded depth of adjusting trench digging arm 11 of slope relation of gradient; Again according to the pipe-laying trencher kinematic behavior; The pace that speed of gyration that the preliminary given one group of cutter chain part of control system is allowed and pipe-laying trencher make every effort to reach; Monitor and pass judgment on its trench digging (excavation) power and whether reach Design Theory power; As not as good as or mistakes, data processing and control center 5 adjust the size of cutter chain speed of gyration and pipe-laying trencher pace based on neural network and genetic algorithm, make the power of ditching can approach Design Theory power as far as possible.
Said data processing and control center 5 are used for signals collecting, data computation, and control algolithm etc., and walking CD-ROM drive motor 2 sent control signal with depth adjustment hydraulic cylinder 7.Mainly comprise computing machine CPU, A/D conversion and data transmission unit (conversion between simulating signal and the digital signal, and with the data transmission that collects to the processing unit), wireless receiving module, control and driving module, display module and power supply etc.
The walking CD-ROM drive motor 3 of said band tachogenerator mainly is used for driving the pipe-laying trencher walking.The size of the speed decision complete machine speed of travel of walking CD-ROM drive motor operation, and through being installed in the tachogenerator in the CD-ROM drive motor rotating shaft, to data processing and control center's 5 output speed of travel signals.Walking CD-ROM drive motor 3 adopts hydraulic system to drive, and is controlled by data processing and control center 5.
The rotating speed of said band wireless transmission and the power transmission shaft of torque sensor 6 are mainly used in to the cutter chain and make up 12 transmitting powers, and through carrying radio transmitting device to data processing and control center's 5 real-time output speed and torque signals.The input speed of power transmission shaft 6 is controlled by data processing and control center 5 by hydraulic speed regulating device 4 adjustment.
Described laser system comprises generating laser 1 and laser pickoff 2, is mainly used in the monitoring that the trench digging gradient reference surface and the trench digging degree of depth are provided to pipe-laying trencher by construction requirement.Said generating laser 1 is used for to laser pickoff 2 the reference laser plane of setting gradient being provided as light source.Said laser pickoff 2 is used to measure the high and low position at self relative laser plane (being provided by generating laser 1) place, and the elevation signal is sent to said data processing and control center 5.
Said laser pickoff 2 is installed in sounding rod 8 tops.
Said sounding rod 8 adopts has the fixed bar that better rigidity and height can be complementary with the trench digging degree of depth.
Said obliquity sensor 10 is used to measure the attitude information of current trench digging arm 11, and signal is sent to said data processing and control center 5.
Said display module is used to show elevation information, and too high, moderate through diode displaying, cross low three kinds of situations, be convenient to the trench digging arm job state of understanding directly perceived.In addition, also be used for showing car load pace, cutter chain speed of gyration and current operation power.
Pipe-laying trencher operation power adaptive control system as shown in Figure 2 comprises:
Data processing and control center 5 are used for signals collecting, data computation, control algolithm and control decision etc.；
Neural network module; Be used to set up neural network model, under given trench digging depth conditions, detect cutter chain speed of gyration and pipe-laying trencher pace and excavate power; And continuous the trench digging degree of depth, cutter chain speed of gyration and the pipe-laying trencher pace of changing; Detect the corresponding power that excavates, the detection data of comprehensive some are utilized the neural network algorithm foundation trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and are excavated the power relation model;
Genetic algorithm module; Be used to use the genetic algorithm optimizing; Under the linear change condition of the trench digging degree of depth; Confirm to excavate the cutter chain speed of gyration of power and the matching value of pipe-laying trencher pace, get wherein one group of optimal value by pipe-laying trencher pace priority principle at utmost to approach given expectation; By Optimum Matching value control cutter chain speed of gyration and pipe-laying trencher pace.
Described operation power adaptive control system is characterized in that, said laser system also is used to use generating laser to be used for to laser pickoff the reference laser plane of setting gradient being provided as light source; Said laser pickoff is used to measure the high and low position at self relative laser plane place, and the elevation signal is sent to said data processing and control center.
Described operation power adaptive control system is characterized in that, also comprises:
Obliquity sensor is used to measure the attitude information of current trench digging arm;
Control and driving module is used for sending control information to power transmission shaft and walking CD-ROM drive motor;
Rotating speed and torque sensor are used to obtain the rotating speed and the torque information of power transmission shaft;
Tachogenerator is used to obtain pipe-laying trencher pace information;
Wireless receiving module is used for rotating speed and torque information to data processing and control center's real-time Transmission power transmission shaft.
Described operation power adaptive control system is characterized in that neural network module also comprises:
Network type confirms and the structure choice unit, is used for neural network is selected and designed, and forms neural network structure;
The training sample unit is used to form the training sample set of said neural network;
The neural metwork training unit is used for neural network is trained.
Described operation power adaptive control system is characterized in that genetic algorithm module also comprises:
The chromosome unit is used to carry out the computing of chromosomal coding, coding/decoding method;
The objective function unit is used to set up objective function;
The fitness function unit is used to carry out the design of fitness function;
The operator design cell is used to select, the design of intersection and mutation operator.
Pipe-laying trencher as shown in Figure 3 is in ditching operation adaptive control process, and control procedure roughly can be divided into following step:
Step 100 is confirmed the trench digging gradient of pipe-laying trencher and the degree of depth of ditching according to construction requirement;
Step 200; Set up neural network model, under given trench digging depth conditions, detect cutter chain speed of gyration and pipe-laying trencher pace and excavate power; And continuous the trench digging degree of depth, cutter chain speed of gyration and the pipe-laying trencher pace of changing; Detect the corresponding power that excavates, the detection data of comprehensive some are utilized the neural network algorithm foundation trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and are excavated the power relation model;
Step 300; The optimizing of utilization genetic algorithm; Under the linear change condition of the trench digging degree of depth; Confirm to excavate the cutter chain speed of gyration of power and the matching value (can obtain one or more groups) of pipe-laying trencher pace, get wherein one group of optimal value by pipe-laying trencher pace priority principle at utmost to approach given expectation;
Step 400 is by Optimum Matching value control cutter chain speed of gyration and pipe-laying trencher pace.
Based on neural network and genetic algorithm process flow diagram shown in accompanying drawing 4.
Realize that based on neural network and genetic algorithm the control procedure of this system is divided into following step:
1. utilize neural network to set up the trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and excavation power relation model
Owing to do not have definite calculation methods or experimental formula between the trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and the excavation power, adopt neural network to set up the relational model between them.Mainly constitute by following steps:
1.1 the selection of neural network and design
Compare with other conventional models, BP (feed-forward type back-propagation algorithm) neural network be use the most extensively, satisfactory method, it has better persistence and real-time prediction property.With regard to native system, can construct three layers of BP neural network shown in accompanying drawing 5.
The three-layer network structure is set up the BP neural network, and the transport function of latent layer adopts tangent Sigmoid function, and last one deck neuron adopts the purelin type function, and whole network output can be got arbitrary value.For fear of the neural metwork training over-fitting, following empirical rule is arranged generally:
N/TW≥1 (1-1)
In the formula: N---the training sample size;
The sum of TW---connection weight in the network (comprising biasing).
The input node of this model has three: cutter chain speed of gyration, pipe-laying trencher pace, the trench digging degree of depth, (calculating of hidden layer node number has the experience formula to 4 nodes of hidden layer, also need pass through test when specifically selecting and obtain.Four nodes here are to calculate feedback through test to obtain), output node is optimum excavation power.
1.2 generation set of data samples
Under given trench digging depth conditions; Detect cutter chain speed of gyration and pipe-laying trencher pace and excavate power, constitute the training sample of neural network through the continuous change trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace simultaneously according to the excavation power that detects.
1.3BP the training of neural network
By the Neural Network Toolbox of Matlab, adopt the conjugate gradient learning algorithm (function of corresponding Neural Network Toolbox: trainscg); Usually the training process target error is made as 0.01.The preset training step number of BP neural network is 2000.Training and forecasting process to network are found the solution, and the BP neural network can constitute the relational model of correct reflection input parameter and output parameter after the training of a large amount of learning samples.
P
w＝f(v
1，v
2，v
3) (1-2)
Wherein: P
wFor excavating power, v
1Be the trench digging degree of depth, v
2Be cutter chain speed of gyration, v
3Be the pipe-laying trencher pace.
Adopt the Matlab interface software to generate the dll dynamic link library automatically, embed in the VB software, constitute real-time computation model.
2. genetic algorithm is confirmed the globally optimal solution region
After setting up the trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and excavating the power relation model; This step is by genetic algorithm; Excavate performance number through importing given expectation; Obtain the Optimum Matching value of cutter chain speed of gyration and pipe-laying trencher pace under the trench digging depth conditions of setting, at utmost to approach given expectation excavation power.Mainly constitute by following steps:
2.1 chromosomal coding, coding/decoding method
Confirm at first the trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace respectively are divided into how many sub-interval (five equilibrium); If the ditching degree of depth is 0.7～2m for example; Operating rate 0.1～2km/h; Cutter chain speed of gyration 0.7～7.0m/s respectively divides 256 intervals (five equilibrium), then divides 768 intervals altogether, and 256 intervals of the degree of depth of wherein ditching are that the determined value of setting only participates in calculating; Do not participate in optimizing processs such as selection, intersection, variation, then hypothesis evenly is divided into N sub-interval (like routine N=768).Chromosome (Chromosome) X=x
1x
2X
NLength setting be that N is counted in the sub-range, chromosomal each x
i(being gene) represented a sub-interval, each gene x
iAll values (being allele) be " 1 ", " 0 "; If chromosomal a certain position (being gene) is taken as " 1 "; Represent that this sub-range participates in excavating power calculation,, represent that this sub-range does not participate in calculating to excavate power if chromosomal a certain position (being gene) is taken as " 0 ".All gene x
iBe arranged in binit string X together, just represent the chromosome of body one by one, as follows:
Sub-range sequence number 12.....................N
All values have an implicit constraint condition here for the gene position number of " 1 " is r in the chromosome x: will remain with a sub-interval at least, therefore, when producing each individuals of initial population, all must guarantee:
r≥1 (2-1)
All individual needs that does not satisfy this constraint regenerate.
2.2 set up objective function
For a certain chromosome x; Value condition according to its locus; The value of selecting corresponding locus is that the numerical value (this is a decode operation) of " 1 " obtains the concrete trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace value; According to neural network model (1-2), calculate target excavation performance number and be shown below:
P
W(X)＝f(x
1，x
2，…，x
N) (2-2)
When power p is excavated in given expectation
WgThe time,
F(X)＝1/|P
Wg-P
W(X)| (2-3)
This functional value is exactly a target function value to be optimized.
2.3 the design of fitness function
According to the definition of target function value, the codomain of this objective function is non-negative, and optimization aim is to ask the maximal value of function.The Optimum Matching value of preferred cutter chain speed of gyration and pipe-laying trencher pace under the trench digging depth conditions of setting is satisfied in the foundation of this objective function, at utmost to approach given expectation excavation power principle.So can get k chromosome x in the population
kFitness equal its target function value:
F(X
k)＝1/|P
Wg-P
W(X
k)| (2-4)
2.4 the design of selection, intersection and mutation operator
(1) select operator design:
Usage ratio is selected (or duplicating) operator.Promptly individual selected and hereditary (or duplicating) is directly proportional with this individual fitness size to the probability in the population of future generation.Concrete implementation is:
● the first step, earlier to all the individual fitness summations in the population:
Here M is a population size, i.e. the individual number that comprises of population.
● second step, calculate each individual relative adaptation degree in the population, with this as the probability of this individuality selected and hereditary (or duplicating) in the population of future generation.
● the 3rd step, adopt the operation of simulation roulette, produce the random number between [0,1], confirm each individual selected number of times.Obviously the individual big individuality of fitness, it selects probability also big, can repeatedly be chosen, and its gene will enlarge in population.
(2) design of crossover operator:
Adopt the single-point crossover operator.Concrete implementation is:
● the first step, earlier the individuality in the population is carried out random pair in twos, establish the population size and be M, then always have [M/2] group of individuals to mutual pairing.Here [x] expression is not more than the maximum integer of x.
● in second step, to each individuality to mutual pairing, the position after a certain locus of picked at random is the point of crossing, and total (N-1) individual possible position can be selected as the point of crossing.Here N is the length (being the sub-range number) of chromosome x.
● the 3rd goes on foot, and to each individuality to mutual pairing, according to the crossover probability Pcrossover that selects in advance, at place, determined point of crossing of second step, exchanges the chromosome dyad of two individuals each other, produces two new individualities.The computing synoptic diagram of crossover operator is as shown in Figure 6.
● the 4th step, constraint condition check.According to implicit constraint condition: will remain with a sub-interval at least, therefore, produce newly when individual through interlace operation, all must guarantee: r >=1.If this condition does not satisfy, changeed for second step, choose the point of crossing again.
(3) design of mutation operator:
Adopt basic position mutation operator, concrete implementation is:
● the first step, to each individual in population locus, mutaion confirms whether this locus is change point with the variation probability P.
● in second step, to the change point of each appointment, its genic value is done " negate " computing (i.e. " 1 " change " 0 ", or " 0 " change " 1 "), thereby produce a new individuality.
● the 3rd step, constraint condition check.According to implicit constraint condition: will remain with a sub-interval at least, therefore, produce newly when individual through mutation operation, all must guarantee: r >=1.If this condition does not satisfy, change the first step, choose change point again.
The computing of " variation " operator is as shown in Figure 7.
Mate after duplicating according to the string of cutter chain speed of gyration, pipe-laying trencher pace, these three input variables of the trench digging degree of depth, select the point of crossing, produce new population, carry out optimizing through selection, intersection and variation then.
Usually the scope of the operational factor of setting genetic algorithm is generally following:
The population size: M=20～100 stop algebraically: G=50～500
Crossover probability: P
Crossover=0.4～0.99 variation probability: P
Mutation=0.0001～0.1
3. under the condition of the given excavation power and the setting trench digging degree of depth, realize the Optimum Matching of cutter chain speed of gyration and pipe-laying trencher pace
(population is made up of some strings when population; The corresponding argument value of each string; In this problem; Population is meant cutter chain speed of gyration, the value of complete machine pace in interval range) adaptation value during more and more near target function value, can obtain one or more groups through genetic algorithm and excavate the cutter chain speed of gyration of power and the matching value of pipe-laying trencher pace at utmost to approach given expectation, get wherein one group of optimal value by pipe-laying trencher pace priority principle; Realize that pipe-laying trencher is under the given trench digging degree of depth; By Optimum Matching value control cutter chain speed of gyration and pipe-laying trencher pace, reach optimum excavation power, to satisfy the designing requirement of system.
Those skilled in the art can also carry out various modifications to above content under the condition that does not break away from the definite the spirit and scope of the present invention of claims.Therefore scope of the present invention is not limited in above explanation, but confirm by the scope of claims.
Claims (11)
1. a pipe-laying trencher operation power adaptive control method is characterized in that, comprising:
Step 100 is confirmed the trench digging gradient of pipe-laying trencher and the degree of depth of ditching according to construction requirement;
Step 200; Set up neural network model, under given trench digging depth conditions, detect cutter chain speed of gyration and pipe-laying trencher pace and excavate power; And continuous the trench digging degree of depth, cutter chain speed of gyration and the pipe-laying trencher pace of changing; Detect the corresponding power that excavates, the detection data of comprehensive some are utilized the neural network algorithm foundation trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and are excavated the power relation model;
Step 300; The optimizing of utilization genetic algorithm; Under the linear change condition of the trench digging degree of depth, confirm to excavate the cutter chain speed of gyration of power and the matching value of pipe-laying trencher pace at utmost to approach given expectation, get wherein one group of optimal value by pipe-laying trencher pace priority principle;
Step 400 is by Optimum Matching value control cutter chain speed of gyration and pipe-laying trencher pace.
2. operation power adaptive control method as claimed in claim 1 is characterized in that said step 100 also comprises:
Step 110 uses generating laser to be used for to laser pickoff the reference laser plane of setting gradient being provided as light source; Said laser pickoff is used to measure the high and low position at self relative laser plane place, and the elevation signal is sent to said data processing and control center.
3. operation power adaptive control method as claimed in claim 2 is characterized in that said step 110 also comprises:
Step 111 is obtained the attitude information of current trench digging arm;
Step 112 sends control information to power transmission shaft and walking CD-ROM drive motor;
Step 113, power transmission shaft motion association cutter chain speed of gyration and operation power obtain the rotating speed and the torque information of power transmission shaft, and the input speed of power transmission shaft is adjusted by hydraulic speed regulating device;
Step 114, walking CD-ROM drive motor motion association pipe-laying trencher pace is obtained pace information;
Step 115 is sent to data processing and control center with the rotating speed and the torque information of power transmission shaft through radio transmitting device, simultaneously pace information is sent to data processing and control center.
4. operation power adaptive control method as claimed in claim 1 is characterized in that said step 200 also comprises:
Step 210 is selected and is designed neural network, forms neural network structure;
Step 220 forms the training sample set of said neural network;
Step 230 is trained neural network.
5. operation power adaptive control method as claimed in claim 1 is characterized in that step 300 also comprises:
Step 310 is carried out the computing of chromosomal coding, coding/decoding method;
Step 320 is set up objective function;
Step 330 is carried out the design of fitness function;
Step 340 is selected, the design of intersection and mutation operator.
6. a pipe-laying trencher operation power adaptive control system is characterized in that, comprising:
Laser system is used for to pipe-laying trencher trench digging gradient reference surface and trench digging depth monitoring being provided according to construction requirement;
Data processing and control center are used for signals collecting, data computation, and control algolithm etc., and CD-ROM drive motor and depth adjustment hydraulic cylinder sent control signal;
Neural network module; Be used to set up neural network model, under given trench digging depth conditions, detect cutter chain speed of gyration and pipe-laying trencher pace and excavate power; And continuous the trench digging degree of depth, cutter chain speed of gyration and the pipe-laying trencher pace of changing; Detect the corresponding power that excavates, the detection data of comprehensive some are utilized the neural network algorithm foundation trench digging degree of depth, cutter chain speed of gyration and pipe-laying trencher pace and are excavated the power relation model;
Genetic algorithm module; Be used to use the genetic algorithm optimizing; Under the linear change condition of the trench digging degree of depth; Confirm to excavate the cutter chain speed of gyration of power and the matching value of pipe-laying trencher pace, get wherein one group of optimal value by pipe-laying trencher pace priority principle at utmost to approach given expectation; By Optimum Matching value control cutter chain speed of gyration and pipe-laying trencher pace.
7. operation power adaptive control system as claimed in claim 6 is characterized in that, said laser system also is used to use generating laser to be used for to laser pickoff the reference laser plane of setting gradient being provided as light source; Said laser pickoff is used to measure the high and low position at self relative laser plane place, and the elevation signal is sent to said data processing and control center.
8. operation power adaptive control system as claimed in claim 6 is characterized in that, also comprises:
Obliquity sensor is used to measure the attitude information of current trench digging arm;
Control and driving module is used for sending control information to power transmission shaft and walking CD-ROM drive motor;
Rotating speed and torque sensor are used to obtain the rotating speed and the torque information of power transmission shaft;
Tachogenerator is used to obtain pipe-laying trencher pace information;
Wireless receiving module is used for rotating speed and torque information to data processing and control center's real-time Transmission power transmission shaft.
9. operation power adaptive control system as claimed in claim 6 is characterized in that neural network module also comprises:
Network type confirms and the structure choice unit, is used for neural network is selected and designed, and forms neural network structure;
The training sample unit is used to form the training sample set of said neural network;
The neural metwork training unit is used for neural network is trained.
10. operation power adaptive control system as claimed in claim 6 is characterized in that genetic algorithm module also comprises:
The chromosome unit is used to carry out the computing of chromosomal coding, coding/decoding method;
The objective function unit is used to set up objective function;
The fitness function unit is used to carry out the design of fitness function;
The operator design cell is used to select, the design of intersection and mutation operator.
11. a pipe-laying trencher is characterized in that, comprises the described system of claim 6-10.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110442098.2A CN102540886B (en) | 2011-12-26 | 2011-12-26 | Self-adaptive control method and system for operation power of pipe-laying trencher |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110442098.2A CN102540886B (en) | 2011-12-26 | 2011-12-26 | Self-adaptive control method and system for operation power of pipe-laying trencher |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102540886A true CN102540886A (en) | 2012-07-04 |
| CN102540886B CN102540886B (en) | 2014-12-10 |
Family
ID=46347984
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110442098.2A Active CN102540886B (en) | 2011-12-26 | 2011-12-26 | Self-adaptive control method and system for operation power of pipe-laying trencher |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102540886B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06327990A (en) * | 1993-05-21 | 1994-11-29 | Kobe Steel Ltd | Control method of roller mill |
| US6496761B1 (en) * | 1999-01-18 | 2002-12-17 | Yamaha Hatsudoki Kabushiki Kaisha | Optimization control method for shock absorber |
| CN1790197A (en) * | 2005-12-22 | 2006-06-21 | 安徽工业大学 | Simple method for neural network decoupling of multi-variable system based on model reference adaptive control |
| JP2009048524A (en) * | 2007-08-22 | 2009-03-05 | Toshiba Corp | Plant optimum operation control system |
| CN102004446A (en) * | 2010-11-25 | 2011-04-06 | 福建师范大学 | Self-adaptation method for back-propagation (BP) nerve cell with multilayer structure |
-
2011
- 2011-12-26 CN CN201110442098.2A patent/CN102540886B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06327990A (en) * | 1993-05-21 | 1994-11-29 | Kobe Steel Ltd | Control method of roller mill |
| US6496761B1 (en) * | 1999-01-18 | 2002-12-17 | Yamaha Hatsudoki Kabushiki Kaisha | Optimization control method for shock absorber |
| CN1790197A (en) * | 2005-12-22 | 2006-06-21 | 安徽工业大学 | Simple method for neural network decoupling of multi-variable system based on model reference adaptive control |
| JP2009048524A (en) * | 2007-08-22 | 2009-03-05 | Toshiba Corp | Plant optimum operation control system |
| CN102004446A (en) * | 2010-11-25 | 2011-04-06 | 福建师范大学 | Self-adaptation method for back-propagation (BP) nerve cell with multilayer structure |
Non-Patent Citations (1)
| Title |
|---|
| 张红彦,赵丁选,陈宁,尚涛: "基于遗传算法的工程车辆自动变速神经网络控制", 《中国公路学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102540886B (en) | 2014-12-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101533000B (en) | Method for constructing water eutrophication risk analysis model | |
| Mohanty et al. | Artificial neural network modeling for groundwater level forecasting in a river island of eastern India | |
| CN110887790B (en) | Urban lake eutrophication simulation method and system based on FVCOM and remote sensing inversion | |
| CN109543237B (en) | Foundation pit displacement prediction method based on GA-BP neural network | |
| CN113283095A (en) | Evolutionary digital twin watershed construction method | |
| CN1760912A (en) | Modeling method of uncertain hydraulics model for urban seweage and drainage system | |
| CN101344389A (en) | Method for estimating tunnel surrounding rock displacement by neural network | |
| CN105353427A (en) | Tunnel surrounding rock dynamic grading method and device thereof | |
| CN101847171A (en) | Back analysis method of slope displacement based on safety monitoring | |
| CN105824993A (en) | Method for artificial sand beach, sand filling and maintenance engineering construction | |
| KR20180116922A (en) | Apparatus for predicting net penetration rate of shield tunnel boring machine and method thereof | |
| CN103968841A (en) | Improved fireflyalgorithm based AUV (autonomous underwater vehicle) three-dimensional track planning method | |
| Schafhirt et al. | Reanalysis of jacket support structure for computer-aided optimization of offshore wind turbines with a genetic algorithm | |
| CN115773127A (en) | Intelligent decision-making method, system, equipment and medium for slurry balance shield | |
| CN115455727A (en) | Digital twin-driven fully-mechanized mining equipment and coal bed coupling system optimization decision method | |
| Afradi et al. | Prediction of TBM penetration rate using the imperialist competitive algorithm (ICA) and quantum fuzzy logic | |
| CN117150880B (en) | Deformation prediction and reverse design method and system for foundation pit supporting structure | |
| CN117688844A (en) | Urban waterlogging real-time simulation method and system based on deep neural network | |
| CN102540886B (en) | Self-adaptive control method and system for operation power of pipe-laying trencher | |
| CN116562448B (en) | Quick prediction method for flood inundation of flood storage area based on coupling of grid analysis and artificial intelligence technology | |
| CN117575094A (en) | Crop yield prediction and optimization method and device based on digital twin | |
| CN110930024B (en) | Intelligent analysis system and method for deep sea emergency situation based on Bayesian network | |
| Zhu et al. | The Integrator Dual-Delay model for advanced controller design of the open canal irrigation systems with multiple offtakes | |
| Alzoubi et al. | Integrating artificial neural network and imperialist competitive algorithm (ICA), to predict the energy consumption for land leveling | |
| CN111783323B (en) | Land hydrologic coupling simulation system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |