A hybrid power and dps positioning cooperative control coupling optimization method

Abstract

This invention relates to the field of ship dynamic positioning control technology, specifically to a coupling optimization method for the coordinated control of hybrid power and DPS positioning. The method involves: establishing a multi-input multi-output nonlinear power-positioning mathematical coupling model to achieve precise quantification and conversion of positioning errors into power parameters; analyzing positioning errors and power requirements using a dual-mode positioning sensor; dynamically adjusting the output power of the diesel engine and battery using a model predictive control algorithm to ensure the power response speed reaches a preset value; and correcting closed-loop errors using a Kalman filter algorithm. This invention achieves precise coupling between hybrid power output and DPS positioning requirements, improving the positioning accuracy and stability of ships in complex sea conditions at sea, while simultaneously reducing fuel consumption and equipment wear of the power system.

Description

Technical Field

[0001] This invention belongs to the field of ship dynamic positioning control technology, specifically relating to a coupling optimization method for the coordinated control of hybrid power and DPS positioning. Background Technology

[0002] For vessels operating in the open sea, the synergy between their hybrid propulsion system and DPS positioning system directly determines the safety and efficiency of their operations. In existing technologies, hybrid propulsion control and DPS positioning control are independent control modules, with power output adjustment lagging behind positioning requirements by approximately 0.8 to 1.2 seconds. This results in positioning accuracy being significantly affected by power fluctuations, and in complex sea conditions, positioning errors can easily exceed the industry-permitted range of ±1 meter.

[0003] To address these issues, the industry currently offers two main improvement solutions: First, increasing the rated output power of the diesel engine to improve power response speed and compensate for control lag with superior performance. However, this solution significantly increases fuel consumption and equipment wear, and fails to resolve coupling issues from the control logic perspective, resulting in limited improvement in positioning accuracy under complex sea conditions. Second, using traditional PID algorithms to link the power system and the DPS system to achieve simple parameter linkage adjustment. However, PID algorithms can only perform single-parameter proportional adjustment and cannot perform multi-dimensional collaborative optimization of the output of multiple power sources (diesel engine + battery). Furthermore, they have weak anti-interference capabilities and are prone to parameter overshoot under wind and wave disturbances, leading to poor positioning stability.

[0004] The core flaw of the above-mentioned improvement scheme is that it does not establish a quantitative coupling relationship between power output and positioning accuracy, which makes it impossible to achieve precise matching between power demand and positioning demand, resulting in poor collaborative control. Therefore, there is an urgent need for an optimization method that can achieve precise coupling between hybrid power and DPS positioning. Summary of the Invention

[0005] To address the issues of lag in power regulation and poor coordination among multiple power sources in existing control schemes, this invention provides a coupling optimization method for the coordinated control of hybrid power and DPS positioning. This method achieves precise coupling between hybrid power output and DPS positioning requirements, improving the positioning accuracy and stability of ships in complex sea conditions, while reducing fuel consumption and equipment wear of the power system.

[0006] The technical solution of the present invention is as follows: A coupled optimization method for hybrid powertrain and DPS positioning coordinated control includes the following steps: Step 1: Using the three-dimensional positioning error of the DPS system as the core input, a multi-input multi-output nonlinear dynamic-positioning mathematical coupling model is constructed. This model is based on the ship's hydrodynamic characteristics, propeller thrust characteristics, and the variation law of the three-dimensional positioning error. The least squares method is used to fit the sea trial data, and the basic framework is constructed by combining the ship's kinematic equations and the power system's output characteristic equations. Parameters are iteratively optimized using a Bayesian optimization algorithm. The multi-input multi-output nonlinear dynamic-positioning mathematical coupling model has a two-layer architecture. The upper positioning error analysis layer receives the three-dimensional positioning error and sea state interference parameters, and outputs the total thrust of the propeller and the thrust distribution coefficient. The lower power parameter mapping layer outputs the target power of the propeller, the target power of the diesel engine, and the target power of the battery. Step 2: The actual position of the ship is obtained in real time by GPS + Beidou dual-mode positioning sensor, and the positioning error is obtained by comparing with the target position. Combined with the positioning accuracy requirements, the power required by the thruster and the response speed requirements of the power system are output through the coupled model analysis. Step 3: Using a model predictive control algorithm, the diesel engine output power and battery charging and discharging power are dynamically adjusted based on the analysis results of the power required by the thruster and the response speed requirements of the power system. The prediction time domain, control time domain, and sampling time of the model predictive control algorithm are set. The objective functions are to minimize the power output tracking error and optimize energy consumption. The minimum value is obtained by solving the quadratic programming algorithm to obtain the control quantity, thereby realizing rolling optimization control and ensuring that the power response speed of the thruster meets the response speed requirements of the power system. Step 4: Real-time acquisition of positioning error data and power system output parameters; use Kalman filtering algorithm to perform closed-loop correction of power output parameters; establish a system state vector including three-dimensional positioning error, actual thruster output power, and thruster power change; construct state equation and observation equation; obtain the optimal positioning error estimate through a five-step process of state prediction, covariance prediction, Kalman gain calculation, state update, and covariance update; calculate the error compensation amount and substitute it into the coupled model to obtain the power parameter compensation value; apply it to the diesel engine and battery system in real time to complete the closed-loop correction.

[0007] Furthermore, in step 1, the sea trial data is ≥100,000, covering sea states 0-10 and with a positioning error range of 0-5 meters; the Bayesian optimization algorithm has ≥500 iterations and a fitting residual ≤0.02.

[0008] Furthermore, in step 1, the sea state disturbance parameters include ; The three-dimensional positioning error includes lateral positioning error. .

[0009] Furthermore, the core expression of the multi-input multi-output nonlinear dynamic-positioning mathematical coupling model is: ; ; ; in, This is the positioning error proportionality coefficient. This is the coefficient of the rate of change of positioning error. The sea state disturbance coefficient is... For the basic power of the thruster, This refers to the rated output power of a single diesel engine. Let i be the target power of the i-th thruster. Let i be the target output power of the i-th diesel engine. This refers to the battery's charging and discharging power.

[0010] Furthermore, in step 2, the positioning accuracy of the Beidou dual-mode positioning sensor is ±0.01 meters, the positioning accuracy requirement is set to ±0.5 meters, the power adjustment range of the output thruster is 0-2500kW / unit, and the power system response speed is ≤0.2 seconds.

[0011] Furthermore, in step 3, based on the analysis results of the required power of the thruster and the response speed requirements of the power system, the diesel engine output power is dynamically adjusted within a range of 1000-1850kW / unit, and the battery charging and discharging power is adjusted within a range of -1600kW to 1600kW, with negative for charging and positive for discharging; the model predictive control algorithm is set to predict the time domain N. p =10 steps, control time domain N c =5 steps, sampling time T s =0.02 seconds; The thruster's dynamic response speed is ≤0.2 seconds.

[0012] Furthermore, the objective function expression of the model predictive control algorithm is: ; in, The actual output power of the thruster in the k-th step. Let k be the change in the control quantity at step k. This is the energy consumption weighting coefficient.

[0013] Furthermore, the real-time output power formulas for the diesel engine, battery, and propulsion system are as follows: ; ; ; in, For diesel engine-propeller transmission efficiency, For battery-to-propulsion energy conversion efficiency, Battery power allocated to the i-th thruster This represents the real-time output power of the diesel engine. This refers to the real-time output power of the battery. This refers to the real-time output power of the thruster.

[0014] Furthermore, in step 4, the sampling frequency for real-time acquisition of positioning error data and power system output parameters is 10Hz; the error compensation accuracy of closed-loop correction is ±0.05 meters. The expression for the system state vector is: ; This represents the system state vector at time k. This represents the ship's three-dimensional positioning error at time k. This represents the actual output power of the i-th thruster at time k. This represents the actual output power of the i-th thruster at time k; The expression for the state equation is as follows: ; This represents the system state vector at time k. Let A represent the system state vector at time k-1, and let B represent the state transition matrix and B represent the control input matrix. This represents the control input vector at time k-1. This represents the noise vector at time k-1. The expression for the observation equation is as follows: ; Denotes the system observation vector at time k. Represents the observation matrix. Let k represent the observation noise vector at time k.

[0015] Furthermore, in step 4, the error compensation amount is calculated. The formula is: ; in, This is the optimal positioning error estimate.

[0016] Compared with the prior art, the present invention has the following beneficial effects: (1) Positioning accuracy has been greatly improved: the average positioning error has been reduced from 0.83 meters to 0.49 meters, the accuracy has been improved by 40%, and the positioning error can still be maintained at ≤±0.5 meters under sea state 5, and the anti-interference ability has been significantly enhanced; (2) Optimization of power response and synchronization: The power response time was reduced to 0.2 seconds, and the power-positioning synchronization coefficient was increased from 0.50 to 0.75, with a 50% improvement in synchronization, which completely solved the problem of power adjustment lag; (3) Improved economy and equipment protection: Through multi-source power synergy optimization, fuel consumption is reduced by 12% and diesel engine equipment loss rate is reduced by 18% compared with traditional solutions; (4) Improved operational safety and efficiency: The optimization of positioning accuracy and stability ensures that there are no cases of excessive positioning errors when the ship is operating at sea, and the operational safety and efficiency are improved by more than 40%. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method of the present invention; Detailed Implementation

[0018] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0019] See Figure 1 This invention provides a coupled optimization method for the coordinated control of hybrid powertrain and DPS positioning. Through four core steps—establishing a coupling model, demand analysis, coordinated adjustment, and error correction—it achieves real-time synchronous optimization of the powertrain and DPS positioning system. The specific steps are as follows: 1. Establish a coupling model Using DPS positioning error as the core input, a multi-input multi-output nonlinear dynamic-positioning mathematical coupling model is constructed to ensure that the model fit is ≥0.98.

[0020] Model Construction: Based on the ship's hydrodynamic characteristics, propeller thrust characteristics, and DPS positioning error variation patterns, the least squares method was used to fit ≥100,000 sets of sea trial data (covering sea states 0-10 and positioning error ranges of 0-5 meters). The basic framework was constructed by combining the ship's kinematic equations and the power system output characteristic equations. The model parameters were iteratively optimized ≥500 times using the Bayesian optimization algorithm, with a fitting residual ≤0.02 as the convergence condition to eliminate model bias.

[0021] Model Architecture: A two-layer architecture is adopted, consisting of a positioning error analysis layer and a dynamic parameter mapping layer. The positioning error analysis layer receives 3D positioning errors and sea state interference parameters. The 3D positioning errors include lateral positioning errors. The sea state disturbance parameters include The output propulsion system generates the total thrust and thrust distribution coefficient of the propulsion unit. Based on the above parameters and combined with the characteristics of the hybrid power system, the power parameter mapping layer outputs the target power of the propulsion unit, diesel engine, and battery.

[0022] Core expression: Define the target power calculation expressions for the propeller, diesel engine, and battery to achieve accurate quantitative conversion of positioning error into power parameters.

[0023] The core expression of the multi-input multi-output nonlinear dynamic-positioning mathematical coupling model is: ; ; ; in, This is the positioning error proportionality coefficient. This is the coefficient of the rate of change of positioning error. The sea state disturbance coefficient is... For the basic power of the thruster, This refers to the rated output power of a single diesel engine. Let i be the target power of the i-th thruster. Let i be the target output power of the i-th diesel engine. This refers to the battery's charging and discharging power.

[0024] 2. Requirements Analysis The actual position of the ship is collected in real time by a GPS+BeiDou dual-mode positioning sensor (positioning accuracy ±0.01 meters), and the three-dimensional positioning error is obtained by comparing it with the target position. Combined with the ±0.5 meter positioning accuracy requirement set by this invention, the positioning error is substituted into the coupling model for analysis, and the required power of the thruster (0-2500kW / unit) and the power system response speed requirement of ≤0.2 seconds are output, providing precise parameter targets for power adjustment.

[0025] 3. Coordinated Regulation Model predictive control algorithms are used to achieve coordinated regulation of multi-source power from the diesel engine and battery, specifically as follows: Set the model predictive control algorithm parameters: Set the prediction time domain N of the model predictive control algorithm. p =10 steps, control time domain N c =5 steps, sampling time T s =0.02 seconds, which meets the requirement of dynamic response speed ≤0.2 seconds; Constructing the objective function: With the goals of minimizing power output tracking error and optimizing energy consumption, an energy consumption weighting coefficient is introduced. =0.05 balances dual objectives; The objective function expression of the model predictive control algorithm is: ;

[0026] in, The actual output power of the thruster in the k-th step. Let k be the change in the control quantity at step k. This is the energy consumption weighting coefficient.

[0027] Solution and Control: With the power adjustment range of diesel engine (1000-1850kW / unit) and battery (-1600kW~1600kW) as constraints, the minimum value of the objective function is solved by quadratic programming algorithm to obtain the power adjustment amount; through rolling optimization control, the output power of diesel engine and battery is updated in real time to ensure that the propeller power response is synchronized with the positioning requirements.

[0028] The formulas for the real-time output power of the diesel engine, battery, and propulsion unit are as follows: ; ; ; in, For diesel engine-propeller transmission efficiency, For battery-to-propulsion energy conversion efficiency, Battery power allocated to the i-th thruster This represents the real-time output power of the diesel engine. This refers to the real-time output power of the battery. This refers to the real-time output power of the thruster.

[0029] 4. Error Correction The positioning error and power output parameters are collected in real time at a frequency of 10Hz (0.1s), and a Kalman filter algorithm is used for closed-loop correction, with an error compensation accuracy of ±0.05 meters. This is one of the core improvement points. Establish filtering equations: Define a state vector that includes three-dimensional positioning error, actual thruster output power, and thruster power variation, construct state equations and observation equations, and determine the Gaussian distribution parameters of process noise and observation noise; The expression for the system state vector is: ; This represents the system state vector at time k. This represents the ship's three-dimensional positioning error (lateral, longitudinal, and heading errors) at time k. This represents the actual output power of the i-th thruster at time k. This represents the actual output power of the i-th thruster at time k. This is the state itself, which is a set of physical quantities.

[0030] The expression for the state equation is: ; This represents the system state vector at time k. Let A represent the system state vector at time k-1, and let B represent the state transition matrix and B represent the control input matrix. This represents the control input vector (control quantities such as diesel engine power and battery power) at time k-1. Let $\mathbf{k}$ represent the process noise vector at time k-1 (model uncertainty, sea disturbance, and other unmodeled disturbances), which follows a Gaussian distribution with mean 0 and variance $Q$, $Q = diag0.0010.0050.002$. This is the state evolution equation, not the state itself. They are not the same thing; it's just that the left side of the equation happens to equal the state variable at the next time step.

[0031] The expression for the observation equation is: ; This represents the system observation vector at time k (actually measured by the sensors, representing positioning error, thruster power, etc.). Represents the observation matrix. Let k represent the observation noise vector (sensor noise) at time k, which follows a Gaussian distribution with mean 0 and variance R (R = diag0.00050.001).

[0032] The five-step correction process involves state prediction, covariance prediction, Kalman gain calculation, state update, and covariance update to obtain the optimal positioning error estimate, eliminating observation noise caused by sea state interference. This technology is existing and is detailed below: Step 1: State Prediction Based on the optimal state estimate of the previous moment, predict the system state at the current moment; A represents the prior state estimate (state prediction) at time k; A represents the state transition matrix. Let u(k-1) represent the posterior state estimate at time k-1 (the optimal state estimate at the previous time step); B represents the control input matrix; u(k-1) represents the control input vector at time k-1 (control quantities such as diesel engine power and battery power). Step 2: Covariance Prediction: The covariance of the current state is predicted, reflecting the magnitude of the prediction error; P(k-1) represents the prior estimation error covariance matrix (prediction covariance) at time k; P(k-1) represents the posterior estimation error covariance matrix at time k-1. denoted by , where represents the transpose of state transition matrix A; Q represents the process noise covariance matrix (sea state disturbance, model error, etc.).

[0033] Step 3: Kalman gain calculation: Based on the predicted covariance and observation noise, the optimal gain is calculated, and the weights of the predicted and observed values ​​are balanced. K(k) represents the Kalman gain matrix at time k; H represents the observation matrix; R represents the transpose of the observation matrix H; R represents the observation noise covariance matrix (sensor measurement noise). Representing matrix inversion Step 4: Status Update By combining the actual observations of the DPS system (positioning error, thruster power), the predicted state is corrected to obtain the optimal state estimate for the current moment; Z(k) represents the posterior state estimate at time k (the optimal state estimate at the current time); Z(k) represents the system observation vector at time k (sensor measurements: positioning error, thruster power, etc.). Step 5: Covariance Update: Update the state covariance to prepare for the correction in the next time step; where I is the identity matrix.

[0034] P(k) represents the posterior estimation error covariance matrix at time k; I represents the identity matrix; Power parameter correction: Calculate the error compensation amount and substitute it into the coupling model to obtain the power compensation value of the diesel engine and battery. Apply it to the power system in real time to complete the closed-loop correction and ensure that the positioning error is stable within ±0.5 meters.

[0035] Calculate the error compensation amount The compensation formula is: ; in, This is the optimal positioning error estimate.

[0036] To make the technical solution of this invention clearer and more feasible, the following uses the operation of an 88-meter vessel under the DPS-2 working condition in the open sea as an example, and combines experimental equipment, full process implementation steps, and implementation effects to elaborate on this invention in detail: (I) Experimental Equipment and System Configuration Vessel and Propulsion System: 88-meter vessel, equipped with 4 propellers (2500kW rated power per unit), 4 diesel engines (1000-1850kW output power adjustable range per unit), and 2 sets of power battery packs (total charging and discharging power adjustable range -1600kW~1600kW). DPS positioning system: GPS + Beidou dual-mode positioning sensor (positioning accuracy ±0.01 meters), three-dimensional attitude sensor, sea state monitoring sensor (wind speed, wave height, ocean current speed), sampling frequency 10Hz; Control and transmission system: It adopts a two-layer data transmission architecture of industrial-grade Ethernet + CAN bus. Ethernet realizes the transmission of large data volume between the DPS system and the central controller, and CAN bus realizes the real-time parameter transmission between the central controller and the power system. The transmission delay is ≤0.01 seconds. The central controller is equipped with the coupling model, model predictive control algorithm and Kalman filter algorithm of this invention, with a main frequency of ≥2.0GHz to ensure the real-time operation of the algorithm.

[0037] (II) Implementation Steps of the Entire Process 1. Data Acquisition and Real-time Transmission DPS side: Collects ship's three-dimensional actual position, wind speed, wave height, ocean current speed, and ship attitude data at a frequency of 10Hz, and transmits them to the central controller via Ethernet with a transmission delay of 0.008 seconds; On the power side: diesel engine output power and speed, battery charging and discharging power and remaining power, and thruster thrust and power data are collected at a frequency of 20Hz and transmitted to the central controller via CAN bus with a transmission delay of 0.005 seconds; Data preprocessing: The central controller performs noise reduction and normalization on all data to eliminate sensor noise and the influence of dimensions, and provides a standard data source.

[0038] 2. Analysis of Positioning Error and Power Requirements The central controller compares the actual position with the preset target working position (39°56′N, 122°20′E) and obtains the three-dimensional positioning error: 0.8 meters laterally, 0.1 meters longitudinally, and 0.5° heading. The lateral error exceeds the ±0.5 meter accuracy requirement and is the main target for correction. Substituting the positioning error and sea state data (wind speed 10m / s, wave height 1.5m, ocean current speed 0.8m / s) into the coupled model, the power requirements are analyzed: each of the two right-side thrusters will increase its power by 500kW (target power 2000kW / unit), while the two left-side thrusters will maintain 1500kW / unit, with a power response speed requirement of ≤0.2 seconds.

[0039] 3. Model predictive control algorithm combined with dynamic regulation The power demand parameters are input into the model predictive control algorithm to predict N in the time domain. p =10 steps, control time domain N c =5 steps, sampling time T s =0.02 seconds, construct the objective function The control variables were obtained by solving the quadratic programming algorithm: the two diesel engines on the right increased from 1500kW to 1750kW, the power battery pack discharged 250kW, and the power of the diesel engine on the left remained unchanged. The central controller transmits the adjustment parameters to the power system via the CAN bus. The power system completes the adjustment within 0.15 seconds, increasing the thrust of the right-side thruster from 80kN to 95kN. The response speed meets the design requirements.

[0040] 4. Kalman Filter Algorithm Error Correction After power adjustment, the lateral positioning error is still 0.75 meters, and the central controller synchronously collects positioning and power parameters at a frequency of 10Hz. The Kalman filter algorithm is corrected through a five-step process: state prediction, covariance prediction, gain calculation, state update, and covariance update (existing technology, not described further). This process eliminates 0.03 meters of observation noise and yields the optimal lateral positioning error estimate of 0.72 meters. The error compensation amount is calculated and substituted into the coupling model, and the output diesel engine power compensation value is +50kW / unit, the battery discharge power compensation value is +20kW, the power system completes the secondary adjustment within 0.1 seconds, and the thruster thrust is increased by 3kN. After correction, the ship's lateral thrust was balanced with the interference of ocean currents and the resistance of wind and waves. The lateral positioning error was reduced to 0.3 meters within 3 seconds, and the longitudinal and heading errors were kept at 0.08 meters and 0.3° respectively, all within the accuracy range. Subsequent micro-corrections were made at a frequency of 10 Hz to ensure stable operation of the ship.

[0041] (III) Implementation Results By employing the coupling optimization method of this invention, the vessel achieved the design target of positioning error ≤ ±0.5 meters under DPS-2 operating conditions. The time from positioning error detection to accuracy stabilization was only 3 seconds, data transmission delay ≤ 0.01 seconds, dynamic response speed ≤ 0.2 seconds, and error compensation accuracy ±0.05 meters, fully meeting the design requirements. During 30 days of offshore operations, no positioning errors exceeded the allowable range. Operational safety and efficiency were improved by more than 40% compared to traditional methods, while fuel consumption was reduced by 12.5%, achieving synergistic optimization of positioning accuracy, operational efficiency, and operational economy.

[0042] Table 1 below shows the experimental data and the basis for improvement: Table 1

[0043] The above description is merely an embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A coupled optimization method for hybrid powertrain and DPS positioning coordinated control, characterized in that, Includes the following steps: Step 1: Using the three-dimensional positioning error of the DPS system as the core input, a multi-input multi-output nonlinear dynamic-positioning mathematical coupling model is constructed. This model is based on the ship's hydrodynamic characteristics, propeller thrust characteristics, and the variation law of the three-dimensional positioning error. The least squares method is used to fit the sea trial data, and the basic framework is constructed by combining the ship's kinematic equations and the power system's output characteristic equations. Parameters are iteratively optimized using a Bayesian optimization algorithm. The multi-input multi-output nonlinear dynamic-positioning mathematical coupling model has a two-layer architecture. The upper positioning error analysis layer receives the three-dimensional positioning error and sea state interference parameters, and outputs the total thrust of the propeller and the thrust distribution coefficient. The lower power parameter mapping layer outputs the target power of the propeller, the target power of the diesel engine, and the target power of the battery. Step 2: The actual position of the ship is obtained in real time by GPS + Beidou dual-mode positioning sensor, and the positioning error is obtained by comparing with the target position. Combined with the positioning accuracy requirements, the power required by the thruster and the response speed requirements of the power system are output through the coupled model analysis. Step 3: Using a model predictive control algorithm, the diesel engine output power and battery charging and discharging power are dynamically adjusted based on the analysis results of the power required by the thruster and the response speed requirements of the power system. The prediction time domain, control time domain, and sampling time of the model predictive control algorithm are set. The objective functions are to minimize the power output tracking error and optimize energy consumption. The minimum value is obtained by solving the quadratic programming algorithm to obtain the control quantity, thereby realizing rolling optimization control and ensuring that the power response speed of the thruster meets the response speed requirements of the power system. Step 4: Real-time acquisition of positioning error data and power system output parameters; use Kalman filtering algorithm to perform closed-loop correction of power output parameters; establish a system state vector including three-dimensional positioning error, actual thruster output power, and thruster power change; construct state equation and observation equation; obtain the optimal positioning error estimate through a five-step process of state prediction, covariance prediction, Kalman gain calculation, state update, and covariance update; calculate the error compensation amount and substitute it into the coupled model to obtain the power parameter compensation value; apply it to the diesel engine and battery system in real time to complete the closed-loop correction.

2. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, In step 1, the sea trial data is ≥100,000, covering sea states 0-10 and with a positioning error range of 0-5 meters; the Bayesian optimization algorithm has ≥500 iterations and a fitting residual ≤0.

02.

3. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, In step 1, the sea state disturbance parameters include ; The three-dimensional positioning error includes lateral positioning error. .

4. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, The core expression of the multi-input multi-output nonlinear dynamic-positioning mathematical coupling model is: ； ； ； in, This is the positioning error proportionality coefficient. This is the coefficient of the rate of change of positioning error. The sea state disturbance coefficient is... For the basic power of the thruster, This refers to the rated output power of a single diesel engine. Let i be the target power of the i-th thruster. Let i be the target output power of the i-th diesel engine. This refers to the battery's charging and discharging power.

5. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, In step 2, the positioning accuracy of the Beidou dual-mode positioning sensor is ±0.01 meters, the positioning accuracy requirement is set to ±0.5 meters, the power adjustment range of the output thruster is 0-2500kW / unit, and the power system response speed is ≤0.2 seconds.

6. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, In step 3, based on the analysis results of the required power of the thruster and the response speed requirements of the power system, the diesel engine output power is dynamically adjusted within a range of 1000-1850kW / unit, and the battery charging and discharging power is adjusted within a range of -1600kW to 1600kW, with negative for charging and positive for discharging; the model predictive control algorithm is set to predict the time domain N. p =10 steps, control time domain N c =5 steps, sampling time T s =0.02 seconds; The thruster's dynamic response speed is ≤0.2 seconds.

7. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, The objective function expression of the model predictive control algorithm is: ； in, The actual output power of the thruster in the k-th step. Let k be the change in the control quantity at step k. This is the energy consumption weighting coefficient.

8. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, The formulas for the real-time output power of the diesel engine, battery, and propulsion unit are as follows: ； ； ； in, For diesel engine-propeller transmission efficiency, For battery-to-propulsion energy conversion efficiency, Battery power allocated to the i-th thruster This represents the real-time output power of the diesel engine. This refers to the real-time output power of the battery. This refers to the real-time output power of the thruster.

9. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, In step 4, the sampling frequency for real-time acquisition of positioning error data and power system output parameters is 10Hz; the error compensation accuracy of closed-loop correction is ±0.05 meters. The expression for the system state vector is: ； This represents the system state vector at time k. This represents the ship's three-dimensional positioning error at time k. This represents the actual output power of the i-th thruster at time k. This represents the actual output power of the i-th thruster at time k; The state equation is expressed as follows: ； This represents the system state vector at time k. Let A represent the system state vector at time k-1, and let B represent the control input matrix. This represents the control input vector at time k-1. This represents the noise vector at time k-1. The expression for the observation equation is as follows: ； Let k represent the system observation vector at time k. Represents the observation matrix. Let k represent the observation noise vector at time k.

10. The coupling optimization method for hybrid power and DPS positioning coordinated control according to claim 1, characterized in that, In step 4, the error compensation amount is calculated. The formula is: ； in, This is the optimal positioning error estimate.

Images