CN105446373A - Intelligent control method for optimizing oil unloading of marine cargo oil pump based on fuzzy immune PID - Google Patents

Abstract

The invention relates to an intelligent control method for optimizing the oil unloading of a marine cargo oil pump based on a fuzzy immune PID. According to the method, the deviation amount of a required oil unloading rate and an actual oil unloading rate and the change rate of the deviation amount are selected as input variables; the parameters of the controller are adjusted online and adaptively in real time according to fuzzy rules; and therefore, the rotation speed of a turbine and the opening degree of a flow regulating valve can be controlled, and control and adjustment of the actual oil unloading rate can be completed. Compared with a conventional PID controller, the fuzzy immune PID controller adopted by the method of the invention has excellent dynamic and steady state characteristics, makes the oil conveying rate of the cargo oil pump stabilized at a required value, rapidly and effectively responds to the change of working conditions and load, control and adjust the oil unloading rate of the cargo oil pump in real time, and can greatly improve the work stability and oil unloading efficiency of the marine cargo oil pump.

Description

The intelligent control method of Cargo Oil Pump emptying peculiar to vessel is optimized based on fuzzy immune PID

Technical field

The present invention relates to Cargo Oil Pump field of intelligent control, particularly relate to a kind of intelligent control method optimizing Cargo Oil Pump emptying peculiar to vessel based on fuzzy immune PID.

Background technology

In the use procedure of Cargo Oil Pump reality, that often wishes crude oil maintains certain setting value to shore current amount and pressure, and this can not only accelerate rate of discharging, improves the operation efficiency of boats and ships; The safety of oil pipeline can also be ensured, prevent the generation of oil accident; More can change crewman into overseer from operator, alleviate its working strength.How to adopt advanced control theory, the performance flow control valve of maximum efficiency and the adjusting function of turbine, in order to reach the two ore control to Cargo Oil flow and pressure.Optimize Cargo Oil Pump emptying control method, improve Cargo Oil Pump oil discharging rate, just become a problem demanding prompt solution.

At present, turbine regulates and is most popularly still typical PID and controls, these controllers are mostly design on the basis with the turbine model determining parameter, when turbine changes away from operating point, Parameter uncertainties or operational factor, these models and real system can produce mismatch, control performance desired when making system be difficult to reach design.

Summary of the invention

The object of the invention is to the intelligent control method proposing a kind of optimization based on fuzzy immune control Cargo Oil Pump emptying peculiar to vessel, stablizing by controller control realization Cargo Oil Pump emptying speed, improve Cargo Oil Pump oil discharging rate.

For reaching above-mentioned purpose, the technical solution adopted in the present invention is:

A kind of intelligent control method optimizing Cargo Oil Pump emptying peculiar to vessel based on fuzzy immune PID is provided, comprises the following steps:

Obtaining detection signal by organizing sensor more, calculating the actual emptying speed of Cargo Oil Pump according to detection signal, and with required emptying speed ratio comparatively, obtain the rate of change Ec of emptying rate variance E and departure;

Using the input of the rate of change Ec of emptying rate variance E and departure as controller, the fuzzy rule one according to presetting obtains controller regulating parameter Δ K _i, Δ K _d; Fuzzy rule two according to presetting obtains nonlinear function f (u (k), Δ u (k)), control law in fuzzy rule two is u (k)=K (1-η f (u (k), Δ u (k))) e (k), wherein u (k) is a kth sampling instant, the output of controller; E (k) is the deviation of a kth sampling instant set-point; K is for controlling reaction velocity; η is for controlling stablizing effect;

By controller online real-time adaptive adjustment controller parameter, obtain:

$\begin{array}{c}{K_P}^{\′}=K\left(1-\mathrm{\η}f\left(u\left(k\right),\mathrm{\Δ}u\left(k\right)\right)\right)\\ {K_I}^{\′}=K_I+f_I{\mathrm{\ΔK}}_I\\ {K_D}^{\′}=K_D+f_D{\mathrm{\ΔK}}_D\end{array},$ Wherein K _p, K _i, K _dfor Initial controller parameter, f _i, f _dfor correction factor;

According to the self-adaptative adjustment controller parameter adjustment turbine rotating speed calculated and flow regulation valve opening.

In method of the present invention, before controller online real-time adaptive adjustment controller parameter, by controller parameter K _i, K _dinitial value be set to 0, K _pput less initial value, make system drop into stable operation; Increase K gradually _p, until continuous oscillation appears in system, i.e. threshold oscillation, records now threshold oscillation gain K _pcritwith critical concussion cycle T _pcrit;

Calculate Initial controller parameter:

K _P＝0.6K _Pcrit

$K_I=\frac{1.2K_{Pcrit}}{T_{crit}}.$

K _D＝0.075K _PcritT _crit

In method of the present invention, step " calculates the actual emptying speed of Cargo Oil Pump " and is specially: the flow Q being detected Cargo Oil by flowmeter, obtains actual emptying speed v=4Q/ (the π D of Cargo Oil Pump according to known tubing internal diameter D ²).

In method of the present invention, the method also comprises step:

After Cargo Oil Pump startup work, detect Cargo Oil temperature signal and pressure signal by sensor, according to Cargo Oil temperature signal, pressure signal and Cargo Oil kind, search from preset database and obtain the kinetic viscosity μ of Cargo Oil and the density p of Cargo Oil;

Calculate reynolds number Re=Dv ρ/μ during Cargo Oil flowing in pipeline, calculate friction factor λ according to Reynolds number and experimental formula;

Calculate friction loss h _f=λ Lv ²/ 2Dg, wherein L is flow process, and g is acceleration of gravity;

Controller according to the watt level of the corresponding adjustment Cargo Oil Pump of the size of friction loss, with make up maintain Cargo Oil conveying energy loss.

In method of the present invention, it is specific as follows that step " calculates friction factor λ according to Reynolds number and experimental formula ":

When Re≤2320, λ=64/Re;

As 2320<Re<4000, λ=0.3164/Re ^0.25;

If pipeline is smooth, clean, then work as 4000<Re<10 ⁵time, λ=0.3164/Re ^0.25; When 10 ⁵<Re<10 ⁸time, λ=0.0056+0.5/Re ^0.32;

If pipeline is coarse, there is deposition, then work as 4000<Re<10 ⁶time, [1+ (20 ε/D+1 × 10, λ=0.0055 ⁶/ Re) ^1/3]; When 10 ⁶during <Re,

λ=0.0055+0.15 × (ε/D) ^1/3, wherein D is internal diameter of the pipeline, and unit m, ε are pipeline absolute roughness, unit m.

In method of the present invention, the method also comprises step:

Detect gas-liquid separator liquid level h, itself and gas-liquid separator liquid level setting height H are contrasted, if h is less than normal, turn flow regulation valve opening down, vacuum pump system is started working; If h is bigger than normal, tune up flow regulation valve opening, vacuum pump system quits work.

The beneficial effect that the present invention produces is: the control thought of binding immunoassay feedback mechanism of the present invention and artificial intelligence, fuzzy-immune PID controllers is adopted to control for turbine governing system, by chaos optimization technology, design is optimized to its parameter, avoid parameter testing work a large amount of in design process, system is made to obtain Optimal Control performance, realize the stable of Cargo Oil Pump emptying speed, improve Cargo Oil Pump oil discharging rate.

Accompanying drawing explanation

Below in conjunction with drawings and Examples, the invention will be further described, in accompanying drawing:

Fig. 1 is embodiment of the present invention Cargo Oil Pump emptying control block diagram.

Fig. 2 is the fuzzy immune PID realistic model of embodiment of the present invention Cargo Oil Pump unloading system.

Embodiment

In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.

Optimize the intelligent control method of Cargo Oil Pump emptying peculiar to vessel based on fuzzy immune PID, it is characterized in that Controlling model have employed two-output impulse generator.

The departure E of input emptying speed and the rate of change Ec of departure.The fuzzy subset of the input variable of Rules1 module is respectively E, Ec, and the fuzzy subset of output variable is Δ K _i, Δ K _d, fuzzy subset is with negative large, in negative, negative little, and zero, just little, center, honest seven descriptive words describe, and its fuzzy language correspondence is designated as: NB, NM, NS, Z0, PS, PM, PB.Membership function is the strong trigonometric function of sensitivity, and domain is the integer between [-6,6].Rules1 module, according to fuzzy experience, sets up fuzzy reasoning table through revising.

Rules1 fuzzy reasoning table

During Rules2 module controls, application Immune Feedback Mechanism designs following control law:

u(k)＝K(1-ηf(u(k),Δu(k)))e(k)

Wherein u (k) is a kth sampling instant, the output of controller; The deviation that e (k) is a kth sampling instant sampled value and setting value; K is for controlling reaction velocity, and namely the speed of response made by controller, and K increases, and response speed is accelerated; η is for controlling stablizing effect, and η increases, and the overshoot of system reduces.Reasonably select and adjust K and η, making control system have response speed and less overshoot faster.F () is about u (k), the nonlinear function of Δ u (k).

Considering the difficulty that nonlinear function f () chooses, the present invention adopts fuzzy logic inference to carry out Nonlinear Function Approximation f (), thus utilizes fuzzy immunization feedback mechanism to adjust the proportional gain K of controller in real time _p'.

Output u (k) being input as Immunocontroller of fuzzy controller and the fuzzy quantity of changes delta u (k) exported, be just expressed as, zero, negative, positive, negative, the fuzzy language of its correspondence is respectively P, Z, N and P, N; The output of fuzzy controller is the fuzzy quantity of nonlinear function f (), is just being expressed as, zero-sum bears, and corresponding fuzzy language is P, N.The fuzzy inference rule of this controller is derived by liapunov's theorem of stability and is drawn, is used for the stability of the system in nonlinear function f () approximate procedure that ensures.Set up the fuzzy reasoning table of Rules2 module.

Rules2 module fuzzy reasoning table

The initial parameter K of its middle controller _p, K _i, K _dsetting, rule of thumb and emulation experiment adjustment determine, enable to reach a good regulating effect.K _p, K _i, K _dconcrete defining method is employed herein Ziegler-Nichols method, and detailed process is as follows.

1) by K _i, K _dbe set to 0, K _pput less value, make system drop into stable operation;

2) K is increased gradually _p, until continuous oscillation appears in system, i.e. threshold oscillation, records now threshold oscillation gain K _pcritwith critical concussion cycle T _pcrit;

3) by formulae discovery controller parameter K _p, K _i, K _d;

K _P＝0.6K _Pcrit

$K_I=\frac{1.2K_{Pcrit}}{T_{crit}}$

K _D＝0.075K _PcritT _crit

So far controller achieves the online real-time adaptive adjustment of parameter, and specifically adjustment is as follows for it:

K _P'＝K(1-ηf(u(k),Δu(k)))

K _I'＝K _I+f _IΔK _I

K _D'＝K _D+f _DΔK _D

Wherein K _p, K _i, K _d, K, η be initial parameter; Δ K _i, Δ K _dobtained by Rules1; F (u (k), Δ u (k)) is obtained by Rules2; f _i, f _dfor correction factor, obtained by experience and experimental debugging.

According to the departure E of emptying speed and the rate of change Ec of departure, controller is by above fuzzy immunization rule, and real-time online calculates adjustment parameters, according to result of calculation obtain new needed for turbine rotating speed and flow regulation valve opening.

After the Controlling model employing of controller, because controller is more intelligent, control effects is optimized lifting, to the control of flow control valve and vacuum pump system on off control more intelligent rationally effectively, thus Cargo Oil degree of vaporizing effectively is regulated, reduce vacuum pump system working load and make again oil transportation Quality advance simultaneously.

The intelligent control method optimizing Cargo Oil Pump emptying peculiar to vessel based on fuzzy immune PID of the embodiment of the present invention, its core is, in fact controller carries out controlling based on the change of real-time working condition, comprises Cargo Oil temperature, pressure, gas-liquid separator liquid level.

After Cargo Oil Pump startup work, many reasons all may cause the change of emptying speed.Emptying rate variance is determined by calculating friction loss in one embodiment of the present of invention.By organizing sensor detection signal, detect temperature, the pressure of Cargo Oil, looking into preset database according to Cargo Oil kind obtains the kinetic viscosity μ of Cargo Oil and the density p of Cargo Oil simultaneously more.Detect flow Q and known tubing internal diameter D according to flowmeter and obtain Cargo Oil flow velocity v=4Q/ (π D ²), then can obtain reynolds number Re=Dv ρ/μ during Cargo Oil flowing in pipeline, thus judge Cargo Oil mobility status in pipeline, and then rule of thumb formula obtains friction factor λ, concrete condition is as follows:

When Re≤2320, λ=64/Re

As 2320<Re<4000, λ=0.3164/Re ^0.25

If pipeline is smooth, clean, then work as 4000<Re<10 ⁵time, λ=0.3164/Re ^0.25; When 10 ⁵<Re<10 ⁸time, λ=0.0056+0.5/Re ^0.32

If pipeline is coarse, deposition, then work as 4000<Re<10 ⁶time, [1+ (20 ε/D+1 × 10, λ=0.0055 ⁶/ Re) ^1/3]; When 10 ⁶during <Re,

λ=0.0055+0.15 × (ε/D) ^1/3; Through unit m, ε pipeline absolute roughness in D pipeline, unit m.

So friction loss h _f=λ Lv ²/ 2Dg, wherein L is flow process, and g is acceleration of gravity.Controller calculates friction loss, normally carries for maintaining Cargo Oil, according to the watt level of the corresponding adjustment Cargo Oil Pump of the size of friction loss, to make up the energy loss maintaining Cargo Oil conveying.When friction loss changes, Cargo Oil Pump power is constant, and Cargo Oil emptying speed can change, thus produce emptying rate variance, for maintaining Cargo Oil emptying rate stabilization, controller utilizes Fig. 2 realistic model according to deviation and deviation variation rate, regulate through fuzzy immunization, thus eliminate the emptying rate variance because friction loss produces, maintain the stable of emptying speed.

According to the gas-liquid separator liquid level h detected, controller receives signal, contrasts with gas-liquid separator liquid level setting height H.H is less than normal, and turn flow regulation valve opening down, vacuum pump system is started working; H is bigger than normal, and tune up flow regulation valve opening, vacuum pump system quits work.

Controller of the present invention makes dynamically and under steady state conditions, and Cargo Oil Pump system works performance obtains good improvement and optimization.

Should be understood that, for those of ordinary skills, can be improved according to the above description or convert, and all these improve and convert the protection domain that all should belong to claims of the present invention.

Claims (6)

1. optimize an intelligent control method for Cargo Oil Pump emptying peculiar to vessel based on fuzzy immune PID, it is characterized in that, comprise the following steps:

Obtaining detection signal by organizing sensor more, calculating the actual emptying speed of Cargo Oil Pump according to detection signal, and with required emptying speed ratio comparatively, obtain the rate of change Ec of emptying rate variance E and departure;

Using the input of the rate of change Ec of emptying rate variance E and departure as controller, the fuzzy rule one according to presetting obtains controller regulating parameter Δ K _i, Δ K _d; Fuzzy rule two according to presetting obtains nonlinear function f (u (k), Δ u (k)), control law in fuzzy rule two is u (k)=K (1-η f (u (k), Δ u (k))) e (k), wherein u (k) is a kth sampling instant, the output of controller; E (k) is the deviation of a kth sampling instant set-point; K is for controlling reaction velocity; η is for controlling stablizing effect;

By controller online real-time adaptive adjustment controller parameter, obtain:

K _P'＝K(1-ηf(u(k),Δu(k)))

K _i'=K _i+ f _iΔ K _i, wherein K _p, K _i, K _dfor Initial controller parameter, f _i, f _dfor

K _D'＝K _D+f _DΔK _D

Correction factor;

According to the self-adaptative adjustment controller parameter adjustment turbine rotating speed calculated and flow regulation valve opening.

2. method according to claim 1, is characterized in that, before controller online real-time adaptive adjustment controller parameter, by controller parameter K _i, K _dinitial value be set to 0, K _pput less initial value, make system drop into stable operation; Increase K gradually _p, until continuous oscillation appears in system, i.e. threshold oscillation, records now threshold oscillation gain K _pcritwith critical concussion cycle T _pcrit;

Calculate Initial controller parameter:

K _P＝0.6K _Pcrit

$K_I=\frac{1.2K_{Pcrit}}{T_{crit}}.$

K _D＝0.075K _PcritT _crit

3. method according to claim 1, it is characterized in that, step " calculates the actual emptying speed of Cargo Oil Pump " and is specially: the flow Q being detected Cargo Oil by flowmeter, obtains actual emptying speed v=4Q/ (the π D of Cargo Oil Pump according to known tubing internal diameter D ²).

4. method according to claim 3, is characterized in that, the method also comprises step:

After Cargo Oil Pump startup work, detect Cargo Oil temperature signal and pressure signal by sensor, according to Cargo Oil temperature signal, pressure signal and Cargo Oil kind, search from preset database and obtain the kinetic viscosity μ of Cargo Oil and the density p of Cargo Oil;

Calculate reynolds number Re=Dv ρ/μ during Cargo Oil flowing in pipeline, calculate friction factor λ according to Reynolds number and experimental formula;

Calculate friction loss h _f=λ Lv ²/ 2Dg, wherein L is flow process, and g is acceleration of gravity;

Controller according to the watt level of the corresponding adjustment Cargo Oil Pump of the size of friction loss, with make up maintain Cargo Oil conveying energy loss.

5. method according to claim 4, is characterized in that, it is specific as follows that step " calculates friction factor λ according to Reynolds number and experimental formula ":

When Re≤2320, λ=64/Re;

As 2320<Re<4000, λ=0.3164/Re ^0.25;

If pipeline is smooth, clean, then work as 4000<Re<10 ⁵time, λ=0.3164/Re ^0.25; When 10 ⁵<Re<10 ⁸time, λ=0.0056+0.5/Re ^0.32;

If pipeline is coarse, there is deposition, then work as 4000<Re<10 ⁶time, [1+ (20 ε/D+1 × 10, λ=0.0055 ⁶/ Re) ^1/3]; When 10 ⁶during <Re,

λ=0.0055+0.15 × (ε/D) ^1/3, wherein D is internal diameter of the pipeline, and unit m, ε are pipeline absolute roughness, unit m.

6. method according to claim 1, is characterized in that, the method also comprises step:

Detect gas-liquid separator liquid level h, itself and gas-liquid separator liquid level setting height H are contrasted, if h is less than normal, turn flow regulation valve opening down, vacuum pump system is started working; If h is bigger than normal, tune up flow regulation valve opening, vacuum pump system quits work.