Control device of internal combustion engine, in-vehicle electronic control unit, machine learning system, control method of internal combustion engine, manufacturing method of electronic control unit, and output parameter calculation device

A control method of an internal combustion engine, the control method comprising:

• acquiring a plurality of input parameters;
• calculating at least one output parameter based on the input parameters using a neural network model;
• and
• controlling the internal combustion engine based on the at least one output parameter, wherein:
• the neural network model includes a plurality of neural network units and an output layer that outputs the at least one output parameter based on outputs of the neural network units;
• each of the neural network units includes one input layer and at least one intermediate layer;
• and
• the neural network model is configured to input different combinations of input parameters selected from the input parameters to each of the input layers of the neural network units such that a total number of input parameters to be input to the neural network units is larger than the number of the input parameters, wherein:
• the number of the input parameters is n;
• and
• the neural network model includes nCk neural network units to which nCk combinations of input parameters selected from the input parameters are input, where n is three or more and k is two to n−1.

An output parameter calculation device comprising:

• a parameter acquisition unit configured to acquire a plurality of input parameters;
• and
• a calculation unit configured to calculate at least one output parameter based on the input parameters acquired by the parameter acquisition unit using a neural network model, wherein:
• the neural network model includes a plurality of neural network units and an output layer that outputs the at least one output parameter based on outputs of the neural network units;
• each of the neural network units includes one input layer and at least one intermediate layer;
• and
• the neural network model is configured to input different combinations of input parameters selected from the input parameters to each of the input layers of the neural network units such that a total number of input parameters to be input to the neural network units is larger than the number of the input parameters, wherein:
• the number of the input parameters is n;
• and
• the neural network model includes nCk neural network units to which nCk combinations of input parameters selected from the input parameters are input, where n is three or more and k is two to n−1.