CN102508960A - Design method for filters for dynamically positioning drag suction dredger in variable-draft operational state - Google Patents

Abstract

The invention discloses a design method for filters for dynamically positioning a drag suction dredger in a variable-draft operational state. The design method includes steps of acquiring data in a measuring system of the drag suction dredger by the aid of information acquisition and processing software based on OPC (OLE (object linking and embedding) for process control) standards, and transferring the data to an upper computer in real time; building a ship motion mathematical model of the drag suction dredger in a state-space form, and respectively designing filters in the degree of freedom of surging, the degree of freedom of swaying and the degree of freedom of heading after the ship motion mathematical model is in discretization processing; designing filters by the aid of improved Sage adaptive filtering algorithm and strong tracking Kalman filtering algorithm; and acquiring the position (in a surging direction and a swaying direction) and the heading angle of the drag suction dredger after data are filtered by the filters. The design method for the filters for dynamically positioning the drag suction dredger in the variable-draft operational state has the advantages that high-frequency interference and noise can be effectively removed in a filtration manner, and the low-frequency position (in the surging direction and the swaying direction) and the heading angle of the drag suction dredger can be estimated.

Description

Become the filter design method of trailing suction hopper dredger dynamically positioning under the drinking water job state

Technical field

The present invention relates to the dynamic positioning of vessels technology, particularly the filter design method of the Large Trailing Suction Hopper Dredger dynamically positioning in the special engineering boats and ships.

Background technology

Receive the influence of marine environment apoplexy, wave, stream in the time of trailing suction hopper dredger operation at sea, mainly show as the low frequency wonder that can make boats and ships produce 0-0.25rad/s and the high frequency reciprocating vibration of 0.3-1.6rad/s.In the process of control, fixed point control position and flight path control, need overcome the influence of above-mentioned factor at the trailing suction hopper dredger bow, traditional way is the accurate operation that relies on mooring system or driver, and this way is time-consuming, effort and efficient are not high yet.

The dynamically positioning of trailing suction hopper dredger is exactly in order to address this problem, to realize automation of operation, intellectuality, improving efficiency of operation.The displacement that the high frequency reciprocating vibration produces can reset voluntarily, and dynamic positioning system need not respond, but can make the position-measurement device run-off the straight, the precision that influence is measured.Therefore need to pass through the wave filter filtering, estimate the low frequency displacement, realize accurately control through the controller of dynamic positioning system from position and the high frequency motion component and the noise of bow in sensor information.

As special engineering ship---trailing suction hopper dredger, become amplitude that drinking water interior draft of following unit interval of job state changes greatly, comprise DT dredging, fixed point dredging, DT and throw mud and four kinds of operator schemes of DP bow spray.Throw the mud operator scheme at DT, trailing suction hopper dredger is constantly thrown mud to light condition along the trajectory of setting from full load condition; Under DP bow spray operator scheme, to locating through bow spray operation from full load condition to light condition, bow spray this moment is to the huge opposition of boats and ships generation at desired location and bow for trailing suction hopper dredger; Under the dredging operator scheme of dredging, fix a point at DT, trailing suction hopper dredger passes through the drag head operation from light condition to full load condition along trajectory of setting or point, and operation drag head and seabed can produce huge dredging power at this moment.The present invention mainly solves the Design of Filter problem under this kind state.

Through retrieval to the prior art document; One Chinese patent application number: 201110080330.2; Proprietary term: the dynamic face hog barge power-positioning control system and the method thereof of adaptive disturbance compensation; Though described hog barge power-positioning control system and method thereof, do not had the concrete design of detailed description power-positioning control system median filter.One Chinese patent application number: 200910052768.2; Proprietary term: a kind of dynamic localization method for ship of revising in real time based on noise matrix; In dynamic positioning of vessels, adopted the von kormon number character filter of revising in real time based on noise matrix; But what be directed against is the common ship of fixing drinking water, and the trailing suction hopper dredger for becoming under the drinking water job state does not have applicability.

Summary of the invention

The technical matters that the present invention will solve is to propose a kind of filter design method that becomes trailing suction hopper dredger dynamically positioning under the drinking water job state.

The present invention becomes the filter design method of trailing suction hopper dredger dynamically positioning under the drinking water job state, it is characterized in that comprising following steps:

A) use data acquisition and process software to obtain the data in the trailing suction hopper dredger measuring system and be sent to host computer in real time based on the OPC standard;

B) set up the trailing suction hopper dredger ship motion mathematical model of state space form, after handling through discretize, at surging, swaying, bow designing filter respectively on three degree of freedom;

C) adopt improved Sage adaptive filter algorithm and strong tracking card Kalman Filtering algorithm design wave filter;

D) position (surge direction, swaying direction) and the bow that behind filter filtering, obtain trailing suction hopper dredger are to angle.

In the said step a), the data of obtaining comprise: the vessel position that DGPS records position and bow that to be the boats and ships bow that records of surge direction and swaying direction, gyro compass promptly be converted into boats and ships central point under the earth coordinates to angle to; The wind speed and direction angle; The power of main thruster, pitch number percent, the pitch number percent of side propeller, the rudder angle of steering wheel; The drinking water of boats and ships midship, FORE DRAFT, aft draft, the water discharge of boats and ships, molded volume, boats and ships speed over the ground; Boats and ships are to the speed of water; About the PIN sensor records between the rake pipe tension force, drag head porch, left and right sides pressure differential, stem jet mud density, flow.

In the said step b); Set up the trailing suction hopper dredger ship motion mathematical model of state space form, structure is as follows: state equation:

Measurement equation: Y=HX+v

Wherein, A, B, E, H are matrix of coefficients, and X is a state vector, and Y throws under the mud operator scheme at DT for measuring vector, and U is making a concerted effort and the wind-force and the wind moment item sum of resultant moment Xiang Yufeng to the boats and ships generation of producing of propulsive mechanism; Under DT dredging, fixed point dredging operator scheme, U is wind-force and wind moment item and the drag head drag item three sum that making a concerted effort of propulsive mechanism generation and resultant moment item, wind produce boats and ships; Under DP bow spray operator scheme, U is wind-force and wind moment item and the bow spray reaction thrust item three sum that making a concerted effort of propulsive mechanism generation and resultant moment item, wind produce boats and ships; W is the process noise item, and v is the observation noise item.

In the said step c), adopt improved Sage adaptive filter algorithm and strong tracking card Kalman Filtering algorithm design wave filter.

The invention has the beneficial effects as follows:

Become under the drinking water job state at trailing suction hopper dredger, the amplitude that drauht changes in the unit interval except that wind, wave, stream, also has the influence of other factors in the suffered external environmental interference greatly.The method that adopts this moment improved Sage adaptive filter algorithm and strong tracking Kalman filtering algorithm to combine; Improved Sage adaptive filter algorithm can carry out match to the variance of observation noise in the variance of process noise in the state equation and the measurement process, and estimates in real time and revise, filtering accuracy height but the limited stability and convergence that is difficult to guarantee system filter of adaptive ability; The strong Kalman filtering algorithm of following the tracks of has very strong mutation status tracking power; Adaptive ability is strong, and reliability is high, but filtering accuracy can decrease; Both are combined; Can give full play to both advantages, not only can suppress filtering divergence but also can guarantee filtering accuracy, well solve the filtering problem when trailing suction hopper dredger becomes the drinking water operation.

Description of drawings

Fig. 1 is a concrete realization block diagram of the present invention;

Fig. 2 is improved Sage adaptive filter algorithm and strong tracking card Kalman Filtering algorithm combination process flow diagram;

Fig. 3 filter effect synoptic diagram.

Embodiment

To combine accompanying drawing below, detailed explanation will be carried out in concrete realization of the present invention.

As shown in Figure 1, a kind of filter design method that becomes trailing suction hopper dredger dynamically positioning under the drinking water job state may further comprise the steps:

1) use the data in the measuring system that information acquisition and process software based on the OPC standard obtain trailing suction hopper dredger also to be sent to host computer in real time.

The data of obtaining comprise: the boats and ships bow that vessel position that DGPS records (surge direction, swaying direction) and gyro compass record is to angle position and the bow of boats and ships central point under the earth coordinates (be converted into to); The wind speed and direction angle; The power of main thruster, pitch number percent, the pitch number percent of side propeller, the rudder angle of steering wheel; The drinking water of boats and ships midship, FORE DRAFT, aft draft, the water discharge of boats and ships, molded volume, boats and ships speed over the ground; Boats and ships are to the speed of water; About the PIN sensor records between the rake pipe tension force, drag head porch, left and right sides pressure differential, stem jet mud density, flow etc.

Vessel position (surge direction, swaying direction) and bow mainly consist of the following components to angle: low frequency amount, high frequency content and noise are formed, and need to estimate the low frequency amount through wave filter filtering high frequency content and noise.These three data are as the input data of wave filter, and remainder data then is used for setting up and upgrading the ship motion mathematical model of trailing suction hopper dredger, calculates the correlation parameter in the ship motion mathematical model.

2) set up the trailing suction hopper dredger ship motion mathematical model of state space form, after discretize is handled, at surging, swaying, bow designing filter respectively on three degree of freedom.Set up the trailing suction hopper dredger ship motion mathematical model of state space form, structure is as follows:

State equation:

Measurement equation: Y=HX+v

Wherein, A, B, E, H are matrix of coefficients, and X is a state vector, and Y throws under the mud operator scheme at DT for measuring vector, and U is making a concerted effort and the wind-force and the wind moment item sum of resultant moment Xiang Yufeng to the boats and ships generation of producing of propulsive mechanism; Under DT dredging, fixed point dredging operator scheme, U is wind-force and wind moment item and the drag head drag item three sum that making a concerted effort of propulsive mechanism generation and resultant moment item, wind produce boats and ships; Under DP bow spray operator scheme, U is wind-force and wind moment item and the bow spray reaction thrust item three sum that making a concerted effort of propulsive mechanism generation and resultant moment item, wind produce boats and ships.Above data can obtain from information acquisition and process software.W is the process noise item, and v is the observation noise item.

Because the dimension of ship craft integrated motion state equation is higher, directly be used for designing filter and can cause calculated amount excessive, therefore at surging, swaying and bow designing filter respectively on three directions, and mentality of designing, method are all consistent.

Do not consider three couplings on the direction, set up three state equation and measurement equations on the direction, all adopt following version:

State equation: X _k=A _{K, k-1}X _K-1+ B _{K, k-1}U _K-1+ E _{K, k-1}W _K-1

Measurement equation: Y _k=H _{K, k-1}X _K-1+ v _K-1

Wherein, A _{K, k-1}, B _{K, k-1}, E _{K, k-1}, H _{K, k-1}Be matrix of coefficients, X _K-1, Y _k, U _K-1, w _K-1, v _K-1Definition and top X, Y, U, w, v is identical.

Discretize is handled:

State equation: X _k=Φ _{K, k-1}X _K-1+ Δ _{K, k-1}U _K-1+ Γ _{K, k-1}W _K-1

Measurement equation: Y _k=H _{K, k-1}X _K-1+ v _K-1

Wherein, Φ _{K, k-1}=I+hA _{K, k-1}, Δ _{K, k-1}=hB _{K, k-1}, Γ _{K, k-1}=hE _{K, k-1}, I is corresponding unit matrix, h is the sampling time.

3) adopt improved Sage adaptive filter algorithm and strong tracking card Kalman Filtering algorithm design wave filter, show that like Fig. 2 the concrete performing step of algorithm is following:

(1) given initial value

P ₀, Q ₀, R ₀, adjustability coefficients γ (γ>=1), forgetting factor ρ (0＜ρ≤1);

(2) $Z_k=Y_k-H_{k,k-1}\·{\hat{X}}_{k-1};$

(3) if

sets up (Tr is for asking matrix trace); Then adopt improved Sage adaptive filter algorithm; Change top step (4)---(12) over to; Otherwise adopt the strong Kalman filtering algorithm of following the tracks of, change following step (13) over to;

(4) $P_{k/k-1}={\mathrm{\Φ}}_{k,k-1}{P}_{k-1}{\mathrm{\Φ}}_{k,k-1}^T+{\mathrm{\Γ}}_{k,k-1}{Q}_{k-1}{\mathrm{\Γ}}_{k,k-1}^T;$

(5) ${\hat{X}}_{k/k-1}={\mathrm{\Φ}}_{k,k-1}{\hat{X}}_{k-1}+{\mathrm{\Δ}}_{k,k-1}{U}_{k-1};$

(6) $K_k=P_{k/k-1}{H}_{k,k-1}^T{(H_{k,k-1}{P}_{k/k-1}{H}_{k,k-1}^T+R_{k-1})}^{-1};$

(7) ${\hat{X}}_k={\hat{X}}_{k/k-1}+K_k(Y_k-H_{k,k-1}{\hat{X}}_{k/k-1});$

(8)P _k＝(I-K _kH _k，k-1)P _k/k-1；

(9) $V_k=H_{k,k-1}{\hat{X}}_k-Y_k;$

(10)

(11) $R_k=\frac{1}{N}\underset{i=0}{\overset{N-1}{\mathrm{\Σ}}}{V}_{k-i}{V}_{k-i}^T+H_{k,k-1}{P}_k{H}_{k,k-1}^T,$ N is a window width, can set as the case may be;

(12)

forwards step (15) to;

(13) $P_{k/k-1}={\mathrm{\λ}}_k{\mathrm{\Φ}}_{k,k-1}{P}_{k-1}{\mathrm{\Φ}}_{k,k-1}^T+{\mathrm{\Γ}}_{k,k-1}{Q}_{k-1}{\mathrm{\Γ}}_{k,k-1}^T;$

Wherein, λ _k=diag [λ ₁(k), λ ₂(k), K, λ _n(k)],

${\mathrm{\λ}}_i\left(k\right)=\left\{\begin{array}{cc}{\mathrm{\α}}_i{C}_k& {\mathrm{\α}}_i{C}_k>1\\ 1& {\mathrm{\α}}_i{C}_k\≤1\end{array}\right.$

$M_k={\mathrm{\Φ}}_{k,k-1}{P}_{k-1}{\mathrm{\Φ}}_{k,k-1}^T{H}_{k,k-1}^T{H}_{k,k-1},$

$C_k=\mathrm{Tr}[V_0\left(k\right)-R_k-{\mathrm{\Γ}}_{k,k-1}{Q}_{k-1}{\mathrm{\Γ}}_{k,k-1}^T]/\mathrm{Tr}\left[{\mathrm{\λ}}_k{M}_k\right],$

$V_0\left(0\right)=Z_0{Z}_0^T;$ $V_0\left(k\right)=[\mathrm{\ρ}{V}_0(k-1)+Z_k{Z}_k^T]/(1+\mathrm{\ρ});$

α _iValue confirm by priori, if no priori, desirable α _i=1;

(14) change step (5)---(12) over to;

(15) if continue filtering, k=k+1 changes (2) and gets into circulation next time, otherwise stops.

Wherein,

Be the initial set value of state vector, the measured value in the time of can be according to filtering carries out initial setting; P ₀For estimating the initial set value of square error battle array, can be set at P ₀=10 ⁴I, I is and P ₀Corresponding unit matrix, Q ₀Be the initial set value of process noise matrix, can set Q ₀=10 ²I, I is and Q ₀Corresponding unit matrix; R ₀Be the initial set value of measurement noise matrix, can set R ₀=I, I is and R ₀Corresponding unit matrix.

4) position (surge direction, swaying direction) and the bow that behind filter filtering, obtain trailing suction hopper dredger are to angle.

Fig. 3 is the filter effect synoptic diagram:

At mean wind speed is under the sea situation about 10.5m/s; Boats and ships midship drinking water from 9.61m change to 9.73m, water discharge from 36508.23 tons change to 37185.6 tons, be under the dredging state about 1.2 joints to the water speed of a ship or plane, adopt this wave filter, trailing suction hopper dredger surging, swaying and bow are as shown in Figure 3 to the filter effect of three directions; Wherein, Before the curve that jitter amplitude is bigger is filtering, after smooth curve is filtering, the evaluated error average, estimate that the square error average is as shown in table 1.

Direction	Evaluated error average/m	Estimate square error average/m
			Surging	0.0913	1.4625
Swaying	0.1023	1.7265
			Bow to	0.0811	0.9347

Table 1 evaluated error average, estimation square error average table

In table 1, the evaluated error average is used for assessing filtering accuracy, estimates that the square error average is used for assessing the stability of filter effect.At trailing suction hopper dredger dynamically positioning accuracy requirement surging, swaying direction ± 2m; Bow can be found out by table 1 and Fig. 3 under the situation of direction ± 1 degree, is becoming under the drinking water job state; This wave filter filtering high frequency interference effectively and noise, estimate low frequency position (surging, swaying) and bow to.

Claims (4)

1. one kind becomes the filter design method that absorbs water trailing suction hopper dredger dynamically positioning under the job state, it is characterized in that comprising following steps:

A) use information acquisition and process software to obtain the data in the trailing suction hopper dredger measuring system and be sent to host computer in real time based on the OPC standard;

B) set up the trailing suction hopper dredger ship motion mathematical model of state space form, after handling through discretize, at surging, swaying, bow designing filter respectively on three degree of freedom;

C) adopt improved Sage adaptive filter algorithm and strong tracking card Kalman Filtering algorithm design wave filter;

D) position that behind filter filtering, obtains trailing suction hopper dredger is that surge direction and swaying direction, bow are to angle.

2. the filter design method of trailing suction hopper dredger dynamically positioning under the change drinking water job state according to claim 1; It is characterized in that in the said step a), the data of obtaining comprise: the vessel position that DGPS records position and bow that to be the boats and ships bow that records of surge direction and swaying direction, gyro compass promptly be converted into boats and ships central point under the earth coordinates to angle to; The wind speed and direction angle; The power of main thruster, pitch number percent, the pitch number percent of side propeller, the rudder angle of steering wheel; The drinking water of boats and ships midship, FORE DRAFT, aft draft, the water discharge of boats and ships, molded volume, boats and ships speed over the ground; Boats and ships are to the speed of water; About the PIN sensor records between the rake pipe tension force, drag head porch, left and right sides pressure differential, stem jet mud density, flow.

3. the filter design method of trailing suction hopper dredger dynamically positioning under the change drinking water job state according to claim 1 is characterized in that in the said step b) that set up the trailing suction hopper dredger ship motion mathematical model of state space form, structure is as follows:

A state equation: mistake! Do not find Reference source.

A measurement equation: mistake! Do not find Reference source.

Wherein, A, B, E, H are matrix of coefficients, and X is a state vector, and Y throws under the mud operator scheme at DT for measuring vector, and U is making a concerted effort and the wind-force and the wind moment item sum of resultant moment Xiang Yufeng to the boats and ships generation of producing of propulsive mechanism; Under DT dredging, fixed point dredging operator scheme, U is wind-force and wind moment item and the drag head drag item three sum that making a concerted effort of propulsive mechanism generation and resultant moment item, wind produce boats and ships; Under DP bow spray operator scheme, U is wind-force and wind moment item and the bow spray reaction thrust item three sum that making a concerted effort of propulsive mechanism generation and resultant moment item, wind produce boats and ships; A mistake! Do not find Reference source.Be process noise item, mistake! Do not find Reference source.Be the observation noise item.

4. the filter design method of trailing suction hopper dredger dynamically positioning is characterized in that in the said step c) under the change drinking water job state according to claim 1, adopts improved Sage adaptive filter algorithm and strong tracking card Kalman Filtering algorithm design wave filter.

Images