CN102426418B - Single-channel electro-hydraulic position servo control method of omnibearing propeller of submarine - Google Patents

Abstract

The invention provides a single-channel electro-hydraulic position servo control method of an omnibearing propeller of a submarine. The method mainly comprises: a model identification link, a control parameter optimization link and a position deviation detection link. When the submarine is in a certain sea state, through identification, a mathematical model of omnibearing propeller motion of the submarine can be acquired. Based on the model, a genetic algorithm is used to optimize an operation of a PID parameter. The parameter obtained through optimization is given to an actual control system of the omnibearing propeller of the submarine. The submarine performs posture adjustment or motion under the parameter. During an adjusting process, data statistics values of pitch angle deviation are continuously detected. If the sea state changes, the parameter of a submarine motion equation changes. When the parameter value of the submarine motion equation exceeds a setting scope, the model is identified again and the PID parameter is optimized again. The method of the invention possesses a good adaptive function to the sea state change. By using the method of the invention, system energy consumption can be saved. The method is simple and easy to be realized. The method possesses the engineering application value.

Description

Single-channel electro-hydraulic position servo control method of omnibearing propeller of submarine

Technical field

What the present invention relates to is the angle of rake control method of a kind of underwater hiding-machine, specifically a kind of omnibearing propeller of submersible hydraulic control system method.

Background technology

Omnibearing propeller of submersible is by adjustment airscrew pitch angle, produces the power (square) of different directions, thereby makes omnibearing propeller of submersible by the direction motion of expection.The adjustment of the angle of pitch is to drive realization by being attached to a mechanical driving device on the swash plate that rotatablely moves.So the control to omnibearing propeller of submersible is to realize by the control to swash plate.The rotation of motion swash plate is completed with the displacement movement of three hydraulic jacks, the Combination conversion of three hydraulic cylinder displacements motion is become motion (two rotation angle of motion swash plate space three degree of freedom, and the displacement movement of each hydraulic jack is controlled and driven by a closed loop electrohydraulic servo system displacement).

For satisfying the requirement of motion swash plate locus and athletic posture, will three independently electrohydraulic servo system organically coordinate, make it according to the requirement of design, realize the various combination of three hydraulic cylinder extension degree and stretching speed, determine swash plate at diverse location and the athletic posture of 3 dimension spaces, and the control accuracy of single channel electro-hydraulic position servo system is the assurance of omnibearing propeller of submersible normal operation.For problems such as electrohydraulic servo system PID controller parameter are difficult to determine, can adopt genetic algorithm to regulate parameter to PID and carry out optimizing, but traditional basic genetic algorithmic can be brought the fraud problem of genetic algorithm usually, be i.e. premature convergence problem and the slow problem of evolving.For the problem that basic genetic algorithmic exists, adopt crossover probability and variation probability to carry out self-adjusting self-adapted genetic algorithm according to fitness value the parameter in the PID controller is chosen automatically.Self-adapted genetic algorithm has very strong search capability, thereby provides guarantee for seeking globally optimal solution, for the parameter On The Choice that solves PID provides an effective approach, thereby has improved control accuracy and the robustness of electrohydraulic servo-controlling system.

Through being retrieved, existing technical literature finds, (the genetic algorithm optimization pid parameter is arranged, but the not application in the omnibearing propeller of submersible electrohydraulic system) do not find and the same or similar bibliographical information of theme of the present invention.

Summary of the invention

The object of the present invention is to provide a kind of variation for sea condition to have adaptation function preferably, can save the single-channel electro-hydraulic position servo control method of omnibearing propeller of submarine of system energy consumption.

Technical scheme of the present invention is as follows:

Single-channel electro-hydraulic position servo control method of omnibearing propeller of submarine mainly comprises Model Distinguish, Optimization about control parameter, position deviation detection; When latent device is under a certain sea situation, draw the mathematical model of omnibearing propeller of submersible motion by identification, and carry out the operation of genetic algorithm optimization pid parameter on the basis of described model, the parameter of optimizing gained is assigned to actual omnibearing propeller of submersible control system, and the device of diving carries out attitude adjustment or action under the parameter of described optimization gained; Be in course of adjustment, constantly detect the data statistics value of angle of pitch deviation, if sea condition changes, the parameter of the device equation of motion of diving changes thereupon, when the parameter value of the device equation of motion of diving exceeds setting range, and identification model and Optimize Multivariable PID Controller again again.

The method of described genetic algorithm optimization pid parameter is:

At first three to the PID controller control parameter scale-up factor k _p, integral coefficient k _i, differential coefficient k _dCarry out binary coding by certain span and obtain three word strings, then three word strings are strung consist of individuality in hereditary space; The random individuality that generates some, by genetic algorithm, each individuality is carried out the fitness assessment, with this foundation as the selection operation, according to crossover probability and variation probability, each individuality is operated the new individuality of generation again, the crossover probability of self-adapted genetic algorithm and variation probability are not changeless, but according to the adaptive change of ideal adaptation degree situation, new individuality carries out fitness assessment and genetic manipulation again, and iterative cycles is until satisfy Optimality Criteria.

The method of the data statistics value of described continuous detection angle of pitch deviation is:

Angle of pitch deviation detection is explained by the standard deviation of angle of pitch deviation in every period time interval, if have: the standard deviation of angle of pitch deviation is less than detecting angle of pitch deviation statistics value K, latent device attitude or the action control precision in the time interval meets the demands at this section to think control system, and system continues operation; Otherwise think that system's control accuracy does not meet the demands, illustrate that pid parameter has been not suitable for the device control system of diving, this moment is identification model again, carries out the pid parameter Optimum Operation of a new round.

The object of the invention is to omnibearing propeller of submersible single channel electro-hydraulic position servo control system and change according to sea condition, utilize improved genetic algorithms method self-adaptation to regulate the PID controller parameter, reach the Control system resolution requirement.if sea condition changes, obtain the mathematical model of omnibearing propeller of submersible attitude motion by the System Discrimination link, and utilize self-adapted genetic algorithm to be optimized design to the PID controller of traditional electrical fluid servo system on the basis of this model, PID controller after optimizing, in the situation that do not need to provide the regulator initial parameter, still can search out suitable parameter, make and control the goal satisfaction requirement, and effectively improve control efficiency, and also can carry out as requested very soon parameter and regulate when systematic parameter changes, satisfy the demand of system in real work, solve and manyly often be operated in the automatic adjusting pattern and discrete from the pattern of adjusting from the PID of setting parameter controller, overcome traditional basic genetic algorithmic and usually can bring the fraud problem of genetic algorithm, at last optimum results is given actual omnibearing propeller of submersible attitude control system.

The invention has the advantages that:

Omnibearing propeller of submersible single channel electro-hydraulic position servo system is a kind of mechanical, electrical, liquid coupled system of complexity, exist non-linear, uncertain, external interference, it is higher due to electro-hydraulic position servo system is controlled quality requirements, traditional PID based on Linear Time Invariant controls and can not satisfy the needs of controlling quality, has proposed a kind of omnibearing propeller of submersible single channel electro-hydraulic position servo intelligence control method that conventional PID controls, genetic algorithm optimizing strategy combines with intelligent coordinated controller.The method has adaptation function preferably with the variation of comparing for sea condition based on simple PID control system, simultaneously, because parameter tuning is timely, has effectively suppressed the increase of omnibearing propeller of submersible electro-hydraulic position amplitude, has saved system energy consumption.This system is simple, be easy to realize to have the engineering using value.

Description of drawings

Fig. 1 omnibearing propeller of submersible single channel electro-hydraulic position servo control algolithm process flow diagram;

Fig. 2 a-Fig. 2 b self-adapted genetic algorithm is optimized electro-hydraulic position servo PID controller parameter process schematic diagram;

Fig. 3 electro-hydraulic position servo control system square-wave response figure.

Embodiment

For example the present invention is done more detailed description below in conjunction with accompanying drawing:

In conjunction with Fig. 1, by introducing course error detection, System Discrimination link and genetic algorithm optimization link on the basis of controlling at conventional PID, both utilized the optimizational function of genetic algorithm to pid parameter, make again the course-keeping system can adaptive work.

For obtaining satisfied transient process dynamic perfromance, the minimum target function that adopts Error Absolute Value time integral performance index to select as parameter.Excessive in order to prevent controlling energy, access control input quadratic term in objective function.The optimum index of selecting following formula to choose as parameter:

$J=\underset{0}{\overset{\∞}{\∫}}\left({\mathrm{\ω}}_1\right|e\left(t\right)|+{\mathrm{\ω}}_2{u}^2\left(t\right))\mathrm{dt}+{\mathrm{\ω}}_3\·t_u---\left(1\right)$

In formula, e (t) is systematic error, and u (t) is controller output, t _uBe the rise time, ω ₁, ω ₂, ω ₃Be weights.

For fear of overshoot, adopted punitive function, namely in case produce overshoot, with overshoot as optimum index one, this moment, optimum index J was:

If ey (t)＜0, $J=\underset{0}{\overset{\∞}{\∫}}\left({\mathrm{\ω}}_1\right|e\left(t\right)|+{\mathrm{\ω}}_2{u}^2\left(t\right)+{\mathrm{\ω}}_4|\mathrm{ey}\left(t\right)\left|\right)\mathrm{dt}+{\mathrm{\ω}}_3\·t_u---\left(2\right)$

In formula, ω ₄Be weights, and ω ₄ω ₁, overshoot ey (t)=y (t)-y (t-1), y (t) is object output.

To three parameter k of PID controller _p, k _i, k _dChoose respectively binary string and represent, and the relation between foundation and parameter, then three binary strings are linked up just form a long scale-of-two word string, this word string is the manipulable object of self-adapted genetic algorithm; In the PID motion controller, N gets 30, M and gets 10, and namely the individual number in population is 30, scale-up factor k _p, integral coefficient k _i, differential coefficient k _dPrecision get respectively 10, thereby each individual total length is 3 * 10=30 position; In order to make the control better effects if of PID controller, with controlled quentity controlled variable, error and the rise time objective function as constraint condition, because fitness function is relevant with objective function, after so objective function is determined, directly its inverse is carried out parameter optimization as fitness function, optimum control parameter namely makes optimum index J hour satisfying under constraint condition, corresponding controller parameter; k _p, k _i, k _dThe operation that the initial population that forms is carried out, in the major part that is the genetic algorithm master routine, comprise selection, three steps of crossover and mutation, in basic genetic algorithmic, the value of selection, crossover and mutation is fixed, in the improved genetic algorithms method, their value is adaptive change, by the operation of three kinds of genetic operators, an Evolution of Population generation, thus the population of a new generation can be brought into the pid algorithm subroutine to obtain new fitness value.

The size that wherein detects angle of pitch deviation statistics value K has reflected the requirement of omnibearing propeller of submersible being controlled effect, when latent device running environment sea condition changes complicated, for fear of system identification and Optimum Operation frequently, can suitably increase the K value, the precision that the device of diving is controlled reduces, and the problem that detection should be noted that is the size of detection time and choosing of K value.Detection time is long, obviously is unfavorable in time controlling the device attitude of diving; Detection time is too short, easily is subjected to the impact of uncertain noises, causes unnecessary identification operation and Optimum Operation.Choosing of K value need to be by a large amount of experiments and in conjunction with the requirement of omnibearing propeller of submersible control accuracy is determined.

In conjunction with Fig. 2, Fig. 3, using number of samples in self-adapted genetic algorithm is 30, excessive for avoiding when using genetic algorithm the parameter selection range, at first by the experience Selecting All Parameters, and sets respectively PID controller scale-up factor k _pSpan be [0,60], integral coefficient k _iSpan is [0,2], differential coefficient k _dSpan is [0,2], fitness function weighting value ω ₁=0.999, ω ₂=0.001, ω ₃=100, ω ₄=2.0; Evolutionary generation is set as 100, crossover probability P _cWith the variation probability P _mCan automatically change with fitness; Setting under sea condition, algorithm through 100 generation iteration after, the PID controller parameter result of adjusting is scale-up factor k _p=58.5696, integral coefficient k _i=0.4938, differential coefficient k _d=0.1304, optimum index J=19.7706; When system's input amplitude was 1 square-wave signal, omnibearing propeller of submersible single channel electro-hydraulic position servo system tracking results can keep well controlling effect, has satisfied the demand of system in real work; This shows, in the situation that do not need to provide initial parameter, still can search out suitable parameter through the PID of self-adapted genetic algorithm optimization controller, make and control the goal satisfaction requirement, can effectively improve control efficiency.

Claims (1)

1. single-channel electro-hydraulic position servo control method of omnibearing propeller of submarine comprises that the mould type distinguishes that knowing ﹑ controls excellentization of ginseng number ﹑ position deviation detection; It is characterized in that: when latent device is under a certain sea situation, draw the mathematical model of omnibearing propeller of submersible motion by identification, and carry out the operation of genetic algorithm optimization pid parameter on the basis of described model, the parameter of optimizing gained is assigned to actual omnibearing propeller of submersible control system, and the device of diving carries out attitude adjustment or action under the parameter of described optimization gained; Be in course of adjustment, constantly detect the data statistics value of angle of pitch deviation, if sea condition changes, the parameter of the device equation of motion of diving changes thereupon, when the parameter value of the device equation of motion of diving exceeds setting range, and identification model and Optimize Multivariable PID Controller again again;

The method of described genetic algorithm optimization pid parameter is: at first three to the PID controller control parameter scale-up factor k _p, integral coefficient k _i, differential coefficient k _dCarry out binary coding by certain span and obtain three word strings, then three word strings are strung consist of individuality in hereditary space; The random individuality that generates some, by genetic algorithm, each individuality is carried out the fitness assessment, with this foundation as the selection operation, according to crossover probability and variation probability, each individuality is operated the new individuality of generation again, the crossover probability of self-adapted genetic algorithm and variation probability are not changeless, but according to the adaptive change of ideal adaptation degree situation, new individuality carries out fitness assessment and genetic manipulation again, and iterative cycles is until satisfy Optimality Criteria;

The method of the data statistics value of described continuous detection angle of pitch deviation is: angle of pitch deviation detection is explained by the standard deviation of angle of pitch deviation in every period time interval, if have: the standard deviation of angle of pitch deviation is less than detecting angle of pitch deviation statistics value K, latent device attitude or the action control precision in the time interval meets the demands at this section to think control system, and system continues operation; Otherwise think that system's control accuracy does not meet the demands, illustrate that pid parameter has been not suitable for the device control system of diving, this moment is identification model again, carries out the pid parameter Optimum Operation of a new round.

Images