JP7349670B2 - Engine control method, engine control system, and ship - Google Patents

Description

The present invention relates to an engine control method, an engine control system, and a ship that can improve engine performance.

In response to stricter EEDI (Energy Efficiency Design Index) regulations and _CO2 emission regulations, marine engines are becoming smaller compared to the size of ships. As engines become more compact, it will not be possible to avoid the negative effects on the engines due to propeller load fluctuations in actual sea conditions using only the current speed feedback control by governors.
Conventionally, adjustments have been made by changing the governor gain or lowering the set rotation speed in advance depending on the magnitude of load fluctuations, but this method reduces the impact of propeller load fluctuations on the engine and improves fuel efficiency. There is no control to optimize it.

Here, in Patent Document 1, the actual rotation speed of the main shaft connected to the main engine is detected, a control calculation unit performs a PID calculation on the deviation between the rotation speed command and the actual rotation speed, and the governor obtained by the PID calculation is The command is output to the governor to control the amount of fuel supplied to the main engine, and the governor command and actual rotation speed are input to the observer to be controlled to estimate the propeller inflow speed fluctuation, and the calculation unit calculates the propeller inflow speed fluctuation. A marine engine control system has been disclosed that corrects the rotation speed command by multiplying the rotation speed command by a predetermined gain.
Furthermore, in Patent Document 2, propeller inflow speeds are calculated by simulation in consideration of ship motion for various combinations of wave height, wave period, ship speed relative to water, ship weight, etc., and the calculated propeller inflow speeds are Calculate the fluctuation of the main engine rotation speed from the fluctuation and find its standard deviation, use these results as a standard deviation database, and refer to the standard deviation database to calculate the wave height, wave period, speed over water, and weight of the ship during navigation. The standard deviation is determined and the allowable rotation speed deviation is calculated, the control section performs PID control of the main engine, a plurality of control modes with different gains are provided, and the control section calculates the allowable rotation speed deviation based on the comparison between the rotation speed deviation and the allowable rotation speed deviation in the comparison section. Disclosed is a main engine control system for a ship that switches between control modes.
In addition, Patent Document 3 discloses that the deviation between the rotational speed command and the actually measured rotational speed of the main shaft or main engine is input into a PID calculation unit to feedback control the amount of fuel supplied from the fuel injection device to the main engine, and then to the propeller. A marine engine control system is disclosed in which a propeller inflow speed is detected and inputted to a calculation unit, and a rotation speed command is corrected so that a control point moves along an efficiency curve in response to fluctuations in the propeller inflow speed.
Further, Patent Document 4 discloses an engine control method for controlling an engine equipped with an exhaust valve and a fuel adjustment means using an engine state observation device that estimates the engine state using an engine model, and detects at least the engine rotation speed. It is disclosed that the engine state observation device estimates at least an excess air ratio as an engine state, and controls at least an exhaust valve as a control target based on the estimated excess air ratio.

JP2012-57523A Japanese Patent Application Publication No. 2011-214471 Japanese Patent Application Publication No. 2010-236463 JP 2019-19783 Publication

None of Patent Documents 1 to 3 performs feedforward control on the engine.
Further, in Patent Document 4, an excess air ratio is estimated as an engine state, and an exhaust valve is controlled based on the estimated excess air ratio, but no detailed explanation of feedforward control is found.
SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide an engine control method, an engine control system, and a ship equipped with the engine control system that improve engine performance through feedforward control.

The engine control method according to claim 1 includes an engine model setting step for setting an engine model of the engine, a set rotation speed obtaining step for obtaining a set engine speed, and a step for predicting engine load fluctuations. A parameter acquisition step that acquires parameters, a state observation step that applies the acquired parameters to the engine model and observes the state including engine load fluctuations, and a state observation step that applies the acquired parameters to the engine model and observes the state including engine load fluctuations. The present invention is characterized by executing a control parameter deriving step of deriving a feedforward control parameter for controlling the engine based on the engine, and an engine control step of applying the derived feedforward control parameter to engine control.
According to the first aspect of the present invention, it is possible to perform feedforward control on the engine in which load fluctuations are predicted by observing the state, thereby improving engine performance.
Note that setting the engine model involves obtaining the conditions of the engine model from the beginning, further obtaining the variables within the model to build the engine model, and obtaining the variables within the model of the engine model that have already been set. This includes linking with other devices and computers to which model parameters have already been input.

The present invention as set forth in claim 2 is characterized in that the parameters acquired in the parameter acquisition step are engine rotational speed and fuel supply amount.
According to the second aspect of the present invention, it is possible to improve the accuracy of the prediction results of load fluctuations based on state observation, and further improve the accuracy of feedforward control parameters.

The present invention according to claim 3 is characterized in that, in the state observation step, a prediction result of load fluctuation is obtained based on an estimation result of the engine load obtained by applying parameters to the engine model.
According to the third aspect of the present invention, the estimated engine load can be reflected in the prediction result of load fluctuation.

The present invention according to claim 4 is characterized in that, in the control parameter deriving step, a predicted result of load fluctuation and a set rotation speed are applied to a system transfer function model to derive feedforward control parameters.
According to the fourth aspect of the present invention, feedforward control parameters can be derived with higher accuracy by using a system transfer function model.

The present invention according to claim 5 is characterized in that, in the control parameter deriving step, feedforward compensation is performed on the predicted result of load fluctuation and the set rotational speed based on a Kalman filter, and feedforward control parameters are derived.
According to the present invention as set forth in claim 5, by using a Kalman filter, feedforward control parameters can be derived with higher accuracy.

The present invention as set forth in claim 6 is characterized in that in the control parameter deriving step, feedforward compensation is performed on the predicted result of load fluctuation and the set rotational speed based on fuzzy inference, and feedforward control parameters are derived.
According to the present invention as set forth in claim 6, by using fuzzy inference, feedforward control parameters can be derived with higher accuracy.

The present invention as set forth in claim 7 is characterized in that, in the engine control step, a command rotation speed is outputted as a feedforward control parameter to a governor provided in the engine.
According to the seventh aspect of the present invention, control can be performed to improve fuel efficiency by speeding up the response of the engine to load fluctuations and reducing unnecessary movements.

The present invention according to claim 8 is characterized in that the engine load fluctuation is a fluctuation due to disturbance of a propeller connected to the engine.
According to the eighth aspect of the present invention, control can be performed that predicts propeller load fluctuations that have a large influence on engine load fluctuations.
Note that the engine model setting step, the set rotation speed obtaining step, the parameter obtaining step, the state observation step, the control parameter deriving step, and the engine control step in any one of claims 1 to 8 may be executed as a computer program. can. Furthermore, even a computer-readable recording medium on which a program is recorded can have the same effects and effects by operating a computer.

An engine control system according to claim 9 includes an engine, a rotation speed setting means for setting the rotation speed of the engine, a parameter acquisition means for acquiring a parameter for predicting engine load fluctuation, and an engine of the engine. An engine model setting section that sets the model, a state observation section that applies the acquired parameters to the engine model and observes the state including engine load fluctuations, and a rotation speed setting means that uses the predicted results of load fluctuations based on the state observation. The present invention is characterized in that it includes a control means having a control parameter derivation unit that derives feedforward control parameters for controlling the engine based on the set rotation speed, and controls the engine based on the derived feedforward control parameters.
According to the present invention as set forth in claim 9, it is possible to perform feedforward control on the engine in which load fluctuations are predicted by observing the state, thereby improving engine performance.

The present invention as set forth in claim 10 is characterized in that the parameter acquisition means is an engine rotation speed sensor and a fuel supply amount sensor.
According to the present invention as set forth in claim 10, it is possible to improve the accuracy of the prediction result of load fluctuation in the state observation section, and further improve the derivation accuracy of the feedforward control parameter in the control parameter derivation section.

The present invention as set forth in claim 11 is characterized in that the condition observation unit obtains a prediction result of load fluctuation based on an estimation result of the engine load obtained by applying parameters to the engine model.
According to the eleventh aspect of the present invention, the estimated engine load can be reflected in the prediction result of load fluctuation.

The present invention as set forth in claim 12 is characterized in that the control parameter deriving section applies the load fluctuation prediction result and the set rotation speed to the system transfer function model to derive the feedforward control parameter.
According to the present invention as set forth in claim 12, by using a system transfer function model, feedforward control parameters can be derived with higher accuracy.

The present invention as set forth in claim 13 is characterized in that the control parameter deriving section performs feedforward compensation on the predicted result of load fluctuation and the set rotational speed based on a Kalman filter, and derives the feedforward control parameter.
According to the thirteenth aspect of the present invention, by using a Kalman filter, feedforward control parameters can be derived with higher accuracy.

The present invention as set forth in claim 14 is characterized in that the control parameter deriving section performs feedforward compensation on the predicted result of load fluctuation and the set rotational speed based on fuzzy inference, and derives the feedforward control parameter.
According to the fourteenth aspect of the present invention, feedforward control parameters can be derived with higher accuracy by using fuzzy inference.

A fifteenth aspect of the present invention is characterized in that the control means controls a governor provided in the engine using a command rotation speed as a feedforward control parameter.
According to the fifteenth aspect of the present invention, control can be performed to improve fuel efficiency by speeding up the response of the engine to load fluctuations and reducing unnecessary movements.

A ship according to claim 16 is characterized in that the engine control system is mounted on a ship having propeller means driven by the engine.
According to the present invention as set forth in claim 16, it is possible to provide a ship equipped with an engine control system that improves engine performance.

The present invention as set forth in claim 17 is characterized in that the state observation is performed by treating the fluctuations of the propeller means due to disturbances as engine load fluctuations in the state observation section.
According to the seventeenth aspect of the present invention, control can be performed that predicts propeller load fluctuations that have a large influence on engine load fluctuations.

According to the engine control method of the present invention, it is possible to perform feedforward control on the engine in which load fluctuations are predicted by observing the state, thereby improving engine performance.

In addition, if the parameters acquired in the parameter acquisition step are the engine rotation speed and fuel supply amount, it is possible to improve the accuracy of the load fluctuation prediction results based on condition observation and, in turn, improve the accuracy of the feedforward control parameters. can.

In addition, in the condition observation step, when obtaining load fluctuation prediction results based on engine load estimation results obtained by applying parameters to the engine model, the estimated engine load is reflected in the load fluctuation prediction results. be able to.

In addition, in the control parameter derivation step, when deriving feedforward control parameters by applying the load fluctuation prediction results and set rotation speed to the system transfer function model, the feedforward control parameters can be calculated by using the system transfer function model. can be derived with more precision.

In addition, in the control parameter derivation step, feedforward compensation is performed on the load fluctuation prediction result and the set rotation speed based on the Kalman filter, and when the feedforward control parameters are derived, the feedforward control is performed by using the Kalman filter. Parameters can be derived with higher accuracy.

In addition, in the control parameter derivation step, feedforward compensation is performed on the load fluctuation prediction result and set rotation speed based on fuzzy inference, and when feedforward control parameters are derived, feedforward control is performed using fuzzy inference. Parameters can be derived with higher accuracy.

In addition, in the engine control step, when outputting the command rotation speed as a feedforward control parameter to a governor installed in the engine, control is performed to improve fuel efficiency by speeding up the engine's response to load fluctuations and reducing unnecessary movement. It can be performed.

Further, if the engine load fluctuation is caused by a disturbance in the propeller connected to the engine, control can be performed that predicts the propeller load fluctuation that has a large influence on the engine load fluctuation.

Furthermore, according to the engine control system of the present invention, feedforward control that predicts load fluctuations can be performed on the engine by observing the state, thereby improving engine performance.

In addition, when the parameter acquisition means is an engine rotation speed sensor and a fuel supply amount sensor, it improves the accuracy of the load fluctuation prediction result in the condition observation section, and further improves the derivation accuracy of feedforward control parameters in the control parameter derivation section. can be improved.

In addition, when the condition observation unit obtains a load fluctuation prediction result based on the engine load estimation result obtained by applying parameters to the engine model, the estimated engine load is reflected in the load fluctuation prediction result. be able to.

In addition, in the control parameter derivation section, when the feedforward control parameters are derived by applying the predicted load fluctuation results and the set rotation speed to the system transfer function model, the feedforward control parameters can be calculated by using the system transfer function model. can be derived with more precision.

In addition, in the control parameter derivation section, feedforward compensation is performed on the load fluctuation prediction result and the set rotation speed based on the Kalman filter, and when the feedforward control parameters are derived, the feedforward control is performed using the Kalman filter. Parameters can be derived with higher accuracy.

In addition, the control parameter derivation section performs feedforward compensation on the load fluctuation prediction result and set rotation speed based on fuzzy inference, and when deriving feedforward control parameters, feedforward control is performed using fuzzy inference. Parameters can be derived with higher accuracy.

In addition, when the control means controls the governor provided in the engine using the command rotation speed as a feedforward control parameter, the control means improves fuel efficiency by speeding up the response of the engine to load fluctuations and reducing unnecessary movements. It can be performed.

Further, according to the ship of the present invention, it is possible to provide a ship equipped with an engine control system that improves engine performance.

Further, when the condition is observed by using the engine load fluctuation in the condition observation section as a result of the disturbance caused by the propeller means, control can be performed in which the propeller load fluctuation, which has a large influence on the engine load fluctuation, is predicted.

Block diagram of an engine control system according to an embodiment of the invention Flow diagram of the engine control method An explanatory diagram when using a system transfer function model as an example of feedforward control. An explanatory diagram when using a Kalman filter as an example of the same feedforward control An explanatory diagram when using fuzzy inference as an example of the same feedforward control

EMBODIMENT OF THE INVENTION Below, the engine control method, engine control system, and ship by embodiment of this invention are demonstrated.

FIG. 1 is a block diagram of an engine control system according to this embodiment.
The engine control system includes an engine 10 provided with a governor 11, a rotation speed setting means 20 for setting the rotation speed of the engine 10, and a parameter acquisition means 30 for acquiring parameters for predicting load fluctuations of the engine 10. A control means 40 is provided.
The control means 40 includes an engine model setting section 42 that sets an engine model 41 of the engine 10, a state observation section 43 that applies acquired parameters to the engine model 41 and observes the state including load fluctuations of the engine 10, and a state observation section 43 that performs state observation including load fluctuations of the engine 10. It has a control parameter derivation unit 44 that derives feedforward control parameters for controlling the engine 10 based on the prediction result of the load fluctuation and the set rotation speed set by the rotation speed setting means 20.
The engine control system is mounted on a ship having propeller means (propeller) 12 driven by an engine 10.
The engine control system controls the engine 10 based on the feedforward control parameters derived by the control parameter derivation unit 44. By observing the state, feedforward control that predicts load fluctuations can be performed on the engine 10 to improve engine performance.

The parameter acquisition means 30 includes an engine rotation speed sensor 31 that detects the engine rotation speed (engine speed) of the engine 10 and a fuel supply amount sensor 32 that detects the amount of fuel supplied to the engine 10. Note that detection of the fuel supply amount includes detection of the fuel pump rack position, fuel flow rate measurement, and the like.

The rotation speed setting means 20, the parameter acquisition means 30, and the control means 40 are connected via an interface to a computer 50 having an engine control program.
Note that the computer 50 can include part or all of the control means 40. When the computer 50 includes a part of the control means 40, other parts are constructed using other computers and hardware circuits.

FIG. 2 is a flow diagram of the engine control method according to this embodiment.
First, the engine model 41 of the engine 10 is set using the engine model setting section 42 (engine model setting step S1).
The engine model 41 is a combination of physical models representing responses of each component of the engine 10. The physical model includes a physical-mathematical model that mathematically expresses the states of the components of the engine 10, a machine learning (ML) model, a nonlinear regression (NLR) model, a transfer function (TF) model, and the like. Here, the physical/mathematical model can faithfully reproduce the engine 10 if data for creating the model is available. Further, although the machine learning (ML) model has a slightly complicated configuration, it is faithful to the engine 10 if the measurement accuracy of the measuring means 40 is sufficient and data for creating the model is available. Although the nonlinear regression (NLR) model has a simple configuration, its accuracy is somewhat inferior even though there are many measured values by the measuring means 30. Although the transfer function (TF) model has a simple configuration, it may be sufficient depending on the components of the main engine 10 (for example, a cooler, etc.). Although these models have advantages and disadvantages, it is desirable to use them appropriately depending on the data items and amount available.
Here, as a representative example, we will discuss an example in which a physical model is constructed only from a physical-mathematical model of a marine diesel engine.
First, there is a model of a governor 11 for controlling the engine speed. The governor 11 determines the amount of fuel input to generate engine torque according to predetermined control settings, and when a mechanical governor is used, it determines the time constant and proportional gain coefficient that reflect the control settings. It is often a model expressed by a first-order differential equation that includes, and in the case of an electronic governor, it is a model based on the PID control law. The engine torque generation model is a model of engine torque generation due to fuel combustion, and the variables are the fuel input amount output from the governor model, engine rotation speed, and supercharger rotation speed, and the generated power torque and shaft system It is common to use a model that subtracts friction. If the supercharger rotation speed is not measured, calculate the value using the supercharger rotation speed model. This model is often calculated using an axial motion differential equation with the turbocharger's turbine torque and compressor torque as external force terms, and a characteristic equation is calculated that takes into account scavenging and exhaust air from the combustion chamber when calculating the turbine torque and compressor torque. . These calculations include a calculation method in which the combustion problem is handled individually for each cylinder, and a calculation method in which the combustion problem for all cylinders is represented by the average value of one rotation cycle. The engine rotational speed response model is determined by the axial motion differential equation of the propulsion shaft system, with engine torque and external force load torque such as propeller torque as external force terms.
The above configuration is common when a physical model of a marine diesel engine is configured using a physical-mathematical model.
Note that the engine model setting section 42 may not be included in the control means 40, but may be constituted by another computer, etc., and the engine model 41 may be set in the control means 40 in advance using this engine model setting section 42. can.
When the control means 40 is constituted by a computer (including a case where it is constituted by a computer 50), the engine model setting section 42 acquires the input conditions of the engine model 41, further acquires the variables within the model, and sets the engine model. Build and set the model 41, obtain and set the input model variables of the already set engine model 41, and link with the engine model 41 of another computer or device to which model parameters have already been input. Including.
In order to identify the variables (coefficients/constants) within the model, data that can be collected before the ship enters service, such as land operation results of an engine of the same type as the engine 10, is used, or data that can be obtained after the ship enters service is used. By updating the variables in the model using data acquired after the aircraft enters service, it is possible to cope with aging deterioration of the engine control system.

Next, the set rotation speed of the engine 10 set by the rotation speed setting means 20 is acquired (set rotation speed acquisition step S2).
The acquired set rotation speed is transmitted to the control means 40.

Next, parameters for predicting load fluctuations of the engine 10 are acquired using the parameter acquisition means 30 (parameter acquisition step S3).
The parameters acquired in the parameter acquisition step S3 are preferably the engine rotation speed acquired by the engine rotation speed sensor 31 and the amount of fuel supplied to the engine 10 acquired by the fuel supply amount sensor 32. Thereby, it is possible to improve the accuracy of the results of predicting load fluctuations based on state observation, and in turn, it is possible to improve the accuracy of feedforward control parameters. Note that the engine rotation speed sensor 31 may be any of various sensors that directly detect the rotation speed of the engine 10 (photocoupler, rotary encoder, etc.) or indirectly detect it (propeller shaft tachometer, etc.).

Next, the condition observation unit 43 applies the parameters acquired using the parameter acquisition means 30 to the engine model 41, performs calculations, and performs condition observation including load fluctuations of the engine 10 (condition observation step S4).
In the state observation unit 43, it is preferable to obtain a prediction result of load fluctuation based on an estimation result of the engine load obtained by applying the parameters to the engine model 41. Thereby, the estimated engine load can be reflected in the load fluctuation prediction result.
Further, in this embodiment, the load fluctuation of the engine 10 is caused by a disturbance in the propeller means 12 connected to the engine 10. Thereby, control can be performed that predicts propeller load fluctuations that have a large influence on engine 10 load fluctuations.

Next, using the control parameter derivation unit 44, feedforward control parameters for controlling the engine 10 are derived based on the load fluctuation prediction result based on the state observation in the state observation unit 43 and the set rotation speed of the engine 10. (Control parameter derivation step S5).

Next, the control means 40 applies the derived feedforward control parameters to the control of the engine 10 (engine control step S6).
To control the engine 10, for example, the control parameter derivation unit 44 derives a commanded rotation speed as a feedforward control parameter, and the control means 40 controls the governor 11 provided in the engine 10 using the commanded rotational speed as the feedforward control parameter. . Thereby, it is possible to perform control that improves fuel efficiency by speeding up the response of the engine 10 to load fluctuations and reducing unnecessary movements.
Note that deriving the command rotation speed as a feedforward control parameter and controlling the governor 11 means performing predictive control by replacing the set rotation speed of the engine 10 set by the rotation speed setting means 20 with the command rotation speed. .

FIG. 3 is an explanatory diagram when a system transfer function model is used as an example of feedforward control according to this embodiment.
FIG. 3(a) shows the configuration of the engine control system. Note that illustration of the propeller means 12, rotation speed setting means 20, parameter acquisition means 30, engine model setting section 42, and computer 50 is omitted.
The engine model 41 and the state observation unit 43 include parameters of the engine 10 acquired by the parameter acquisition means 30 (engine rotation speed n _e , fuel supply amount h _p , supercharger rotation speed n _TC , scavenging pressure P _s , and average The effective pressure P _e ) is input. Note that the supercharger rotation speed n _TC , scavenging pressure P _s , and average effective pressure P _e may be estimated using the engine model 41.
The condition observation unit 43 observes the condition by applying the acquired parameters to the engine model 41, and obtains an estimated value u _p of the propeller inflow speed (propeller disturbance) as a prediction result of the load fluctuation of the engine 10 ("u" is written at the top). ) is output.
The control parameter deriving unit 44 applies the estimated value of the propeller inflow speed u _p ("u" has "~" at the top) and the set rotation speed to the system transfer function model, and derives the feedforward control parameters. By using a system transfer function model, feedforward control parameters can be derived with higher accuracy, leading to improvements in fuel efficiency, etc.

FIG. 3(b) is a diagram showing a system transfer function model. In FIG. 3(b), "n _sp " is the set rotation speed of the engine 10, "FF" is the feedforward filter, "W _G " is the governor response (transmission) function, and "W _h " is the fuel supply output by the governor. ``W u'' is the transfer function from the quantity h <sub>p</sub> to the engine speed <sub>ne</sub> , ``W _u '' is the transfer function from the disturbance u _p (``u'' is marked with ``~'' above) to the engine speed _ne , and ``W _nt '' is the A transfer function from the charger to the engine speed _ne , "W _ne " is a transfer function from the engine speed _ne to the supercharger, and "W _th " is a transfer function from the governor 11 to the supercharger.
The output Y is obtained by multiplying the transfer function W _ss by the state X. Further, the control value Δn _sp of the set rotational speed n _sp is obtained by multiplying the reciprocal of the transfer function W _ss by the estimated value u _p of the propeller inflow speed (“u” has “~” attached above).

The control means 40 transmits the command rotation speed n _order as a feedforward control parameter to the governor 11 . The governor 11 adjusts the fuel supply amount _hp based on the command rotation speed n _order . This compensates for disturbances caused by fluctuations in the propeller inflow speed _up and stabilizes the speed of the supercharger. This has a significant impact on fuel consumption. Furthermore, the control means 40 can adjust the gain of the feedforward filter of the control parameter deriving section 44 to reduce the negative influence on the engine speed _ne . Here, the gain is a value (proportional gain) that determines whether the control value Δn _sp , which is a control parameter, is to be changed significantly.

FIG. 4 is an explanatory diagram when a Kalman filter is used as an example of feedforward control according to this embodiment. Note that illustration of the propeller means 12, rotation speed setting means 20, parameter acquisition means 30, engine model setting section 42, and computer 50 is omitted.
The engine model 41 and the state observation unit 43 include parameters of the engine 10 acquired by the parameter acquisition means 30 (engine rotation speed n _e , fuel supply amount h _p , supercharger rotation speed n _TC , scavenging pressure P _s , and average The effective pressure P _e ) is input. Note that the supercharger rotation speed n _TC , the scavenging pressure P _s , and the average effective pressure P _e may be estimated using calculation results by the engine model 41.
The condition observation unit 43 applies the acquired parameters to the engine model 41 to observe the condition by calculation, and obtains an estimated value u _p of the propeller inflow speed (propeller disturbance) as a prediction result of the load fluctuation of the engine 10 ("u is the upper (with “～”) is output.
The control parameter deriving unit 44 performs feedforward compensation on the estimated propeller inflow speed u _p (“u” is marked with “~” above) and the set rotation speed n _sp based on a Kalman filter, and derives feedforward control parameters. do. By using a Kalman filter, feedforward control parameters can be derived with higher accuracy, leading to improvements in fuel efficiency, etc. Note that as the Kalman filter, an extended Kalman filter (EKF), an unscented Kalman filter (UKF), or the like can be used.

The feedforward control parameters are derived from the following equation (1).
Here, h' _p is the fuel supply amount correction value, K _G is the Kalman gain, and P _i is the state covariance.

The control means 40 transmits the command rotation speed n _order as a feedforward control parameter to the governor 11 . The governor 11 replaces the set rotation speed set by the rotation speed setting means 20 with the predicted control command rotation speed n _order and adjusts the fuel supply amount h _p . This makes it possible to compensate for disturbances caused by fluctuations in the propeller inflow speed _up .

FIG. 5 is an explanatory diagram when fuzzy inference is used as an example of feedforward control according to this embodiment.
FIG. 5(a) shows the configuration of the engine control system. Note that illustration of the propeller means 12, rotation speed setting means 20, parameter acquisition means 30, engine model setting section 42, and computer 50 is omitted.
The engine model 41 and the state observation unit 43 include parameters of the engine 10 acquired by the parameter acquisition means 30 (engine rotation speed n _e , fuel supply amount h _p , supercharger rotation speed n _TC , scavenging pressure P _s , and average The effective pressure P _e ) is input. Note that the supercharger rotation speed n _TC , the scavenging pressure P _s , and the average effective pressure P _e may be estimated using calculation results by the engine model 41.
The condition observation unit 43 applies the acquired parameters to the engine model 41 to observe the condition by calculation, and obtains an estimated value u _p of the propeller inflow speed (propeller disturbance) as a prediction result of the load fluctuation of the engine 10 ("u is the upper (with “～”) is output.
The control parameter deriving unit 44 performs feedforward compensation on the estimated propeller inflow speed u _p (“u” is marked with “～” above) and the set rotation speed n _sp based on fuzzy reasoning, and derives feedforward control parameters. do. By using fuzzy inference, feedforward control parameters can be derived with higher accuracy, leading to improvements in fuel efficiency, etc.

FIG. 5(b) is a diagram showing derivation of feedforward control parameters based on fuzzy inference. In FIG. 5(b), "&" is an AND operation, and "||" is an OR operation.
The control parameter deriving unit 44 combines engine torque and propeller torque, and derives feedforward control parameters using fuzzy inference based on the imbalance between propeller torque and engine torque.

The control means 40 transmits the command rotation speed n _order as a feedforward control parameter to the governor 11 . The governor 11 replaces the set rotation speed set by the rotation speed setting means 20 with the command rotation speed n _order and adjusts the fuel supply amount _hp . By changing the command rotation speed n _order , it is possible to compensate for disturbances caused by fluctuations in the propeller inflow speed _up .
Note that in the above example, the estimated value u _p of the propeller inflow speed (propeller disturbance) was used as the prediction result of the load fluctuation of the engine 10, but the propeller inflow speed u _p is directly measured and used to derive the feedforward control parameters. You can also do that.

The above description is illustrative of exemplary embodiments according to the present disclosure and is not intended to be limiting. The present disclosure may be implemented in forms different from those expressly described herein, and various modifications, optimizations, and variations may be realized by those skilled in the art to the extent consistent with the claims. .

[Additional notes]
Note that the present invention can also be expressed as follows.
(Additional note 1)
to the computer,
an engine model setting step for setting an engine model of the engine;
a set rotation speed obtaining step of obtaining a set rotation speed of the engine;
a parameter acquisition step of acquiring parameters for predicting load fluctuations of the engine;
a state observation step of applying the acquired parameters to the engine model and observing the state of the engine including the load fluctuation;
a control parameter deriving step of deriving a feedforward control parameter for controlling the engine based on the prediction result of the load fluctuation based on the state observation and the set rotation speed of the engine;
An engine control program comprising: an engine control step of applying the derived feedforward control parameters to control of the engine.
(Additional note 2)
The engine control program according to appendix 1, wherein the parameters acquired in the parameter acquisition step are engine rotation speed and fuel supply amount.
(Additional note 3)
The engine according to appendix 1 or 2, wherein in the state observation step, the prediction result of the load fluctuation is obtained based on the engine load estimation result obtained by applying the parameter to the engine model. control program.
(Additional note 4)
Any one of Supplementary Notes 1 to 3, wherein in the control parameter deriving step, the predicted result of the load fluctuation and the set rotation speed are applied to a system transfer function model to derive the feedforward control parameter. The engine control program described in section.
(Appendix 5)
Supplementary Notes 1 to 3, wherein in the control parameter deriving step, feedforward compensation is performed on the prediction result of the load fluctuation and the set rotation speed based on a Kalman filter, and the feedforward control parameter is derived. The engine control program according to any one of the items.
(Appendix 6)
Supplementary notes 1 to 3, wherein in the control parameter deriving step, feedforward compensation is performed on the prediction result of the load fluctuation and the set rotation speed based on fuzzy inference, and the feedforward control parameter is derived. The engine control program according to any one of the items.
(Appendix 7)
The engine control program according to any one of Supplementary Notes 1 to 6, wherein in the engine control step, a command rotation speed is outputted as the feedforward control parameter to a governor provided in the engine.
(Appendix 8)
The engine control program according to any one of appendices 1 to 7, wherein the load fluctuation of the engine is a fluctuation due to disturbance of a propeller connected to the engine.
(Appendix 9)
A recording medium for an engine control program, characterized in that the engine control program according to any one of Supplementary Notes 1 to 8 is recorded thereon.

INDUSTRIAL APPLICABILITY The present invention can improve performance and fuel efficiency of marine engines or other engines by predicting load fluctuations through feedforward control. Furthermore, the present invention can be developed as an engine control method and system, as well as a program and a recording medium on which a program is recorded.

10 Engine 11 Governor 12 Propeller means (propeller)
20 Rotation speed setting means 30 Parameter acquisition means 31 Engine rotation speed sensor 32 Fuel supply amount sensor 40 Control means 41 Engine model 42 Engine model setting section 43 Condition observation section 44 Control parameter derivation section 50 Computer S1 Engine model setting step S2 Setting rotation speed Acquisition step S3 Parameter acquisition step S4 Condition observation step S5 Control parameter derivation step S6 Engine control step h _p Fuel supply amount n _e Engine rotation speed n _Order command rotation speed n _sp Set rotation speed

Claims (17)

an engine model setting step for setting an engine model of the engine;
a set rotation speed obtaining step of obtaining a set rotation speed of the engine;
a parameter acquisition step of acquiring parameters for predicting load fluctuations of the engine;
a state observation step of applying the acquired parameters to the engine model and observing the state of the engine including the load fluctuation;
a control parameter deriving step of deriving a feedforward control parameter for controlling the engine based on the prediction result of the load fluctuation based on the state observation and the set rotation speed of the engine;
An engine control method characterized by executing an engine control step of applying the derived feedforward control parameter to control of the engine. The engine control method according to claim 1, wherein the parameters acquired in the parameter acquisition step are an engine rotation speed and a fuel supply amount. According to claim 1 or claim 2, in the state observation step, the prediction result of the load fluctuation is obtained based on the engine load estimation result obtained by applying the parameter to the engine model. engine control method. 4. In the control parameter deriving step, the feedforward control parameter is derived by applying the predicted result of the load fluctuation and the set rotation speed to a system transfer function model. The engine control method according to item 1. In the control parameter deriving step, feedforward compensation is performed on the prediction result of the load fluctuation and the set rotation speed based on a Kalman filter, and the feedforward control parameter is derived. 3. The engine control method according to any one of 3. In the control parameter deriving step, feedforward compensation is performed on the prediction result of the load fluctuation and the set rotation speed based on fuzzy inference, and the feedforward control parameter is derived. 3. The engine control method according to any one of 3. 7. The engine control method according to claim 1, wherein in the engine control step, a command rotation speed is outputted as the feedforward control parameter to a governor provided in the engine. 8. The engine control method according to claim 1, wherein the load fluctuation of the engine is a fluctuation due to disturbance of a propeller connected to the engine. an engine, a rotation speed setting means for setting the rotation speed of the engine, a parameter acquisition means for acquiring parameters for predicting load fluctuations of the engine, an engine model setting section for setting an engine model of the engine; a state observation unit that applies the acquired parameters to the engine model and observes the state of the engine including the load fluctuation; and a set rotation set by the rotation speed setting means based on the prediction result of the load fluctuation based on the state observation. and a control means having a control parameter deriving unit that derives a feedforward control parameter for controlling the engine based on the number of feedforward control parameters, and the engine is controlled based on the derived feedforward control parameter. Engine control system. The engine control system according to claim 9, wherein the parameter acquisition means is an engine rotation speed sensor and a fuel supply amount sensor. 11. The state observation unit obtains the prediction result of the load fluctuation based on the engine load estimation result obtained by applying the parameter to the engine model. engine control system. 12. The control parameter deriving section applies the predicted result of the load fluctuation and the set rotational speed to a system transfer function model to derive the feedforward control parameter. The engine control system according to item 1. The control parameter deriving section performs feedforward compensation on the prediction result of the load fluctuation and the set rotational speed based on a Kalman filter, and derives the feedforward control parameter. 12. The engine control system according to any one of Item 11. The control parameter deriving unit performs feedforward compensation on the prediction result of the load fluctuation and the set rotation speed based on fuzzy inference, and derives the feedforward control parameter. 12. The engine control system according to any one of Item 11. The engine control system according to any one of claims 9 to 14, wherein the control means controls a governor provided in the engine using a command rotation speed as the feedforward control parameter. A ship, characterized in that the engine control system according to any one of claims 9 to 15 is mounted on a ship having propeller means driven by the engine. 17. The marine vessel according to claim 16, wherein the condition of the propeller means is observed as the load variation of the engine in the condition observation section.