JP2540567B2 - Self-referenced adaptive control system for ship motion reduction - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a control system for reducing motion of a hull (including a floating body) in waves.

[Prior Art] Conventionally, as an example of a control system, there is a model reference adaptive control system (hereinafter referred to as MRAS).

This model reference adaptive control system, as shown in the block diagram of FIG. 3, gives a reference input to the reference model and also gives the same input to the controlled object via the controller, and the difference between the response outputs of the two becomes an error. By giving the error signal as a signal to the adaptive mechanism, the parameter of the controller is changed so that the error signal becomes zero, that is, the two coincide with each other, whereby the response of the controlled object (including the controller). It seeks to match the characteristics to the response of the reference model with the desired characteristics.

[Problems to be solved by the invention]

By the way, considering the case where the model reference adaptive control system described above is applied to the motion control of a ship in waves, the same wave disturbance as that received by the ship to be controlled is applied to the reference model, and the response output You will have to compare the differences.

However, in order to give a wave disturbance to the reference model, it is necessary to measure the wave and estimate the magnitude of the wave forcing force or moment based on the wave measurement. Moreover, it has to be performed in real time (real time) for irregular waves on the actual sea surface, but there is a problem in the accuracy of measurement of waves and the accuracy of estimation of wave forcing, and the calculation processing speed of them. However, there is a limit, and although it is possible as a concept, it is impossible to actually execute it.

Since the goal of motion control in waves is to keep the motion amplitude as small as possible, it is not necessary to use a normative model as long as there is a normative output to be targeted, and the target value is zero. May be

The present invention was created in view of the above points, and does not require a normative model, and operates an adaptive mechanism in a direction of decreasing the output of the controlled object itself as a reference, which is a hull motion. It is an object to provide a self-reference adaptive control system for reduction.

[Means for solving problems]

In order to achieve the above-mentioned object, a self-reference adaptive control system for reducing ship motion of the present invention is an adaptive control system for reducing motion of a ship in waves, which is a control target that operates based on a manipulated variable from feedback. Means for producing an error signal from the response output of ## EQU1 ## and a response value obtained by multiplying the response output by a coefficient between 0 and 1, and the error signal becomes zero by operating with the error signal from the means. As described above, the self-normative adaptive mechanism for outputting the adaptive signal for adjusting the manipulated variable is provided.

That is, in the self-reference adaptive control system for reducing ship motion of the present invention, the target output Xm is obtained by multiplying the output Xp (t) from the controlled object by the target output coefficient S.
(T), the error signal e (t) is created as follows.

Xm (t) = S · Xp (t), where 0 ≦ S <1 e (t) = Xm (t) −Xp (t) = S · Xp (t) −Xp (t) = (1-S)・ Xp (t) Then, this e (t) is input to the adaptive mechanism. The operation principle thereafter is the same as the existing model reference adaptive control.

[Action]

In the above-described self-reference adaptive control system for reducing ship motion of the present invention, the target output coefficient S is a value between 0 and 1, and S · Xp (t) is the target value. In other words, the present system operates the adaptive mechanism with the target value of S · Xp (t) which is always smaller than the current output value Xp (t), and always reduces the wave motion amplitude of the controlled object. It will guide you and control you. This is effective even when the state quantity of the controlled object, for example, the drainage rate, the ship speed, or the disturbance condition (for example, the wave period or the wave height) changes, and is adaptive control.

In this way, this system operates the adaptive mechanism in the direction of decreasing it based on the output of the controlled object itself. Therefore, this system is referred to as Self Reference Adaptive System (SRAS). (Abbreviated)).

〔Example〕

FIG. 1 shows a block diagram of the self-normative adaptive control system of the present invention. The controlled object is subject to seed feedback determined in the design reference state of the system. In this case, since the control is aimed at reducing the motion amplitude in the waves, the operation amount is given by feedback. Therefore, the reference input is zero.

The adjustable parameters Ga (t) and Gb (t) are the wave disturbance W.
The response output Xp (t) of the control means by (t) is the target output Xm.
It is determined by activating the adaptation mechanism to match (t). The target output coefficient S is a value that can be arbitrarily given between 0 and 1, but if S = 0, the target output is controlled as Xm (t) = 0, and if S = 1. For example, Xm (t) = Xp (t), so the output error is e
(T) = 0 and the adaptive controller does not operate. Therefore, if this function is used and S = 1 is set at an appropriate time while observing the convergence state of the response, that state is held, and the system has room for judgment intervention.

The calculation formula of the self-normative adaptive control system of the present invention is shown below.

It is assumed that the equation of motion of the controlled object is given by the SRAS calculation formula in the form of the state equation as follows.

p (t) = AoXp (t) + BoUp (t) + W (t) (1) where Ao and Bo are coefficient matrices, for example, AoεR
It is assumed that ^{5 × 5} and BoεR ^{5 × 2} .

W (t) is the wave disturbance.

In general MRAS, Up (t) = Um (t) is given as a reference input, whereas when targeting motion control in waves, Up (t) is a feedback that cancels the influence of disturbance. It is given as a manipulated variable. An adaptive controller of Up (t) = − Ga (t) Xp (t) + Gb (t) r (t) Ga ∈ R ^{2 × 5} , Gb ∈ R ^{2 × 2} ... (2) is added to this system. Then, p (t) = AoXp (t) + Bo {−Ga (t) Xp (t) + Gb (t) r (t)} + W (t) = {Ao−BoGa (t)} Xp (t) + BoGb ( t) r (t) + W (t) (3) However, r (t) is the main feedback gain F
Is given by r (t) = − F · Xp (t).

Here, if Ao-BoGa (t) = Ap (t) BoGb (t) = Bp (t) (4) is set, the control target including the adjustable parameters Ga (t) and Gb (t) The state equation can be expressed as p (t) = Ap (t) Xp (t) + Bp (t) r (t) + W (t) (5).

By the way, the ship motion in waves is a kind of forced vibration due to the disturbance W (t), and even if it is controlled, the motion amplitude cannot be made completely zero, and a certain amplitude Xp
o (t) will remain.

 The convergence state is po (t) = ApoXpo (t) + Bporo (t) + W (t) (6).

Now, assuming that the target output (or reference output) is Xm (t), s times Xp (t) Xm (t) = s · Xp (t) where 0 ≦ s <1 s: target output coefficient The difference from the plant output Xp (t) is e (t) = Xm (t) −Xp (t) = (s−1) Xp (t) (7).

As mentioned above, Xp (t) does not converge to zero, so
As for the state error e (t), eo (t) = (s-1) Xpo (t) (8) remains.

 Therefore, instead of directly targeting e (t), E (t) = e (t) −eo (t) = (s−1) {Xp (t) −Xpo (t)} ... ( Let us consider applying Lyapunov's stability theorem to 9).

When the equation (9) is differentiated, (t) = (t) -o (t) = (s-1) Xp (t)-(s-1) Xpo (t) = (s-1) {Ap (t ) Xp (t) + Bp (t) r (t) + W (t)}-(s-1) {ApoXpo (t) + Bporo (t) + W (t)} = (s-1) [{Ap (t) -Apo} Xp (t) + {Bp (t) -Bpo} r (t) + Apo {Xp (t) -Xpo (t)} + Bpo {r (t) -ro (t)}] ... (10 ) Here, [Ap (t) -Apo: Bp (t) Bpo] = φ (t), φεR ^{5 × 7} [Xp (t) ^T : r (t) ^T ] ^T = Z (t), If ZεR ^{7 × 1} [Apo−BpoF] = Am, then (t) = (s−1) φ (t) Z (t) + AmE (t) (11)

As a candidate of Lyapunov ^{function, V = E T PE + ΣΣγ} ij -1 (φ ij + δ ij θ ij) Consider ² ... (12). However, P: Positive definite symmetric matrix φ = {φ _ij } (i = 1 ... 5, j = 1 ... 7) γ _ij > 0 (const) δ _ij > 0 (const) PEZ ^T = {θ _ij } (12) is differentiated equation at time t, = to become ^{E T PE + E T PE +} ΣΣγ ij -1 2 (φ ij + δ ij θ ij) × {ij + δ ij ij) ··· (13). Here, if φ _ij = −γ _ij θ _ij −δ _{ij ij} (14), then: = −2ΣΣδ _ij θ _ij ² (15), and is negative. Therefore, V becomes a Lyapunov function that becomes V → 0 for t → ∞, and becomes E → 0. That is, e (t) → eo (t).

Than this adaptive law, using e (t) instead of _{E (t), v i =} ΣP ik e k ··· (16) z j = (x p1, x p2, ···, r 1 , r ₂ ) ・ ・ ・ (17) θ _ij = v _i・ z _j・ ・ ・ (18) φ _ij ＝ −δ _ij θ _ij −γ _ij ∫ θ _ij dt ・ ・ ・ (19) Ga = Bo ⁺ { If φ ₁ (t) + Aoo} Gb = −Bo ⁺ {φ ₂ (t)} (20) is calculated, the result is e (t) → eo (t).

Incidentally, if the 0 initial value of phi may be calculated as _{^{Ga = Bo + {φ 1 (}} t)} Gb = -Bo + {φ 2 (t) -Boo}. Bo ⁺ is Boo's Pseudo Inverse,
Boo is Bo in the standard state.

In order to specifically show the effect of the present invention, numerical simulation results are shown in FIGS. 2 (a), (b) and (c). The object is to control the pitch of a special ship that runs at high speed in waves, and control the angle of attack of a hydrofoil attached to the stern to reduce the pitch amplitude. Main feedback is required, assuming that 50 knots is the design standard condition. It shows how the self-referenced adaptive control system of the present invention operates when the ship speed changes from that state to 30 knots.

 The numerical values used for the calculation are as follows.

Bo = [− 0.0007487U ² ] Boo = [− 0.49533] Bo ⁺ = [− 2.0188] F = [− 2.309, −0.8025] P = [0.9762,0.1915] W (t) = {− 0.0004895exp (−0.1178) } × {-509.6cos (1.571) + 324.5sin (1.571)} sin (ωet) ωe = 0.8776 + 0.0785 / U U is the ship speed (m / s). U = 0.5144 Vs, Vs is a knot. Speed 50
U = 25.72m / s for knots, U = 41.16m / s for 80 knots.

And the output is The manipulated variable is U = [α], the attack angle of the hydrofoil. However, α limited the operating range to ± 30 °.

As shown in Fig. 2 (a), the ship speed is 10se
Sail at 50 knots to c, then decelerate to 30 after 20 sec
It is assumed that the vehicle is sailing at knots, and regarding the pitching angle (control amount) and hydrofoil attack angle (manipulation amount) at that time, the case where the adaptive control of the present invention is used and the case where the feedback gain is fixed (there is no adaptive control) Case). Adaptive control was activated after 10 seconds when the ship speed changed.

As can be seen from FIGS. 2 (b) and 2 (c), when there is no adaptive control, the change in the hydrofoil attack angle is small and the pitch angle is large, whereas when there is adaptive control, the hydrofoil is small. The angle of attack of is operated up to the limit (α = ± 30 °), and the pitch angle is reduced.

As described above, it is understood that the self-normative adaptive control of the present invention is an extremely effective means for motion control in waves.

〔The invention's effect〕

As described in detail above, according to the self-reference adaptive control system for hull motion reduction of the present invention, the following effects and advantages are obtained.

(1) It is not necessary to use the reference model, and the reference output can be created based on the output from the controlled object.

(2) Since the reference model is not used, it is not necessary to measure waves, and therefore, it can be applied to control in irregular oblique waves.

(3) Since the reference model does not need to be moved, the time required for control calculation is shortened, and it becomes more suitable for real-time control.

(4) It is possible to realize a non-linear system that minimizes the response to the wave disturbance within the allowable range of the manipulated variable.

(5) The target output coefficient can be adjusted while judging the condition of motion control.

[Brief description of drawings]

FIG. 1 is a block diagram showing a self-reference adaptive control system for reducing ship motion as an embodiment of the present invention, and FIGS. 2 (a), (b) and (c) are adaptations by the system of the present invention. FIG. 3 is a graph showing a control simulation calculation result, and FIG. 3 is a block diagram showing a conventional model reference adaptive control system.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yuu Takahashi 1-1 No. 1 Satinoura-machi, Nagasaki-shi, Nagasaki Mitsubishi Heavy Industries, Ltd. Nagasaki Research Institute (56) Reference JP-A-62-229403 (JP, A) JP 60-116593 (JP, A)

Claims (1)

(57) [Claims] 1. An adaptive control system for reducing the motion of a hull in waves, wherein a response output of a controlled object which operates based on an operation amount from feedback and the response output are multiplied by a coefficient between 0 and 1 Means for creating an error signal from the response value obtained by the above, and a self-normative type that operates by the error signal from the means and outputs an adaptive signal for adjusting the manipulated variable so that the error signal becomes zero. A self-referenced adaptive control system for reducing ship motion, characterized by being provided with an adaptive mechanism.