Transtron Inc.

Kogakuin University

Isuzu Motors Limited

Takuma Degawa

Masatoshi Ogawa

Masakazu Mukai

Shigeaki Kurita

Satoshi Hanawa

Yoshiaki Ikai

Hidekazu Fujie

A control device calculates a first estimate value of a first monitored amount of a mobile object, from sensor information of a controlled object, using a transformed first model obtained by transformation of a first model having a multi-layered neural structure and activation functions of a ReLU structure, and predicting dynamic behavior of the mobile object. The transformed first model is a model for which weight coefficients, biases, and upper and lower limit values of input/output variables of neurons are set based on the first model and in which activation functions of the neurons are transformed using a linear inequality function having a binary variable. The control device calculates a second estimate value of a second monitored amount of the controlled object from a candidate value of an operation amount for controlling the second monitored amount. The control device determines a value of the operation amount based on a target value of the second monitored amount, the second estimate value, and the first estimate value.

Control device, control method, and recording medium

An engine intake system control device configured to control an intake system of an engine, having a map function that inputs at least a fuel injection pressure of the engine, a fresh air flow, and a compressor outlet temperature of a supercharger, and outputs a control gain; and a control unit that inputs the control gain and a deviation between a controlled variable of the intake system of the engine and a target value thereof, and controls a manipulated variable of the intake system of the engine.

Engine intake system control device and control method thereof

An engine control device includes a model that, based on engine operating condition and first-type operation amount, reproduces at least one index from among various indexes of combustion state of engine, and a processor that executes a process including deciding on second-type operation amount, by optimization using the model so as to treat at least one of the indexes, which are reproduced by the model, as estimated value of control amount, and ensure that the estimated value of the control amount follows control target value, associating the second-type operation amount with the control target value and the engine operating condition, rewriting a learning control table in which operation amount corresponding to the control target value and the engine operating condition is registered, and calculating operation amount according to the learning control table based on the control target value and the engine operating condition.

Engine control device, and engine control method

A method including acquiring a first test pattern in which an operation variable for an engine test is changed in time series, inputting a first operation variable to a mathematical model representing a time series response obtained by inputting a test pattern as a simulation of the engine test, monitoring a first monitoring parameter obtained by inputting the first operation variable to the mathematical model, holding the first operation variable until the first monitoring parameter is less than a first threshold value, creating a history of the first operation variable in the simulation as a second test pattern, monitoring a second monitoring parameter obtained by inputting a second operation variable to a real engine based on the second test pattern, holding the second operation variable until the second monitoring parameter is less than a second threshold value, acquiring time series data of the second operation variable and a controlled variable.

Engine test method, computer-readable recording medium, and engine test apparatus

A control device includes a controller configured to: obtain a driving condition and an actual value of a controlled amount of an engine; use a transformed engine model to calculate an estimate value of a future controlled amount, from the obtained driving condition and a candidate value of an operation amount for controlling the controlled amount, the transformed engine model in which an activation function of each neuron is transformed using a linear inequality function having a binary variable, the transformed engine model for which a weight coefficient, bias, and upper and lower limits of an input/output variable of the each neuron are set based on an engine model that has a multilayer neural structure and the activation function of a ReLU structure, receives the driving condition and operation amount as input, and predicts dynamic characteristics of the engine; determine a value of the operation amount, based on error between the calculated estimate value and a target value of the controlled amount identified from the driving condition.

An engine model construction method includes generating test patterns in which a plurality of manipulated variables used for an engine test are changed with time, correcting the test patterns based on first coverage of a first space of manipulated variables are allowed to take and second coverage of a second space of change rate values of the manipulated variables are allowed to take, acquiring pieces of time series data of operation amounts of the manipulated variables and controlled amounts with respect to the manipulated variables by performing an engine test using the corrected test patterns, and constructing a first engine model by performing machine learning on training data in which the corrected test patterns are adopted as input and the pieces of time series data are adopted as correct answers, by a processor.

Engine model construction method, engine model constructing apparatus, and computer-readable recording medium

A control parameter computation method includes, by a processor, identifying, for each time segment, a plurality of mathematical models of different structures. The method also includes, computing an adaptiveness representing a level of adaptation between a time series of the computed control variable predicted values and time series data of the control variable in the paired data corresponding to the time segments different from the time segment employed in the identification. The method also includes, selecting, as a mathematical model for the paired data of the time segment. The method also includes, configuring a PID controller, generating a transfer function represented by a product of the PID controller and the selected mathematical model, and generating a function representing a gain margin of the transfer function and a function representing a phase margin of the transfer function. The method also includes computing a parameter Lambda.

Control parameter computation of a proportional-integral-derivative (PID) controller

In order to adequately suppress both an increase in NOx emission amount and deterioration of fuel consumption rate, the present invention provides a fuel injection control information generation device equipped with: a test point information storage unit for holding test point information including a plurality of test points constituted by sets of engine speed, fuel injection amount, and oxygen concentration; and a control information generation unit for generating, for each test point included in the test point information, fuel injection control information in which the engine speed, fuel injection amount, and oxygen concentration of the test point are associated with an optimum fuel injection timing at which an index pertaining to the total of a fuel consumption rate and an NOx emission amount in the test point becomes the smallest.

Growing community of inventors

Yokohama, Japan

Takuma Degawa