Wavelet-based artificial neural net combustion sensing

A method for determining a combustion characteristic of an internal combustion engine, comprising

• concurrently sampling an ionization signal and determining at least one engine operating parameter during a combustion event;
• transforming the ionization signal with a wavelet filter;
• extracting at least one critical signal feature from the ionization signal that has been transformed by the wavelet filter;
• and
• processing the at least one critical signal feature and the at least one engine operating parameter through an artificial neural network.

A system to control an internal combustion engine, comprising controller operably connected to:

• at least one output device of the internal combustion engine,
• at least one ion sensor operable to measure an ionization signal from a combustion event of the internal combustion engine, and
• at least one engine sensor operable to measure at least one engine operating parameter;
• wherein said controller is operable to:
• concurrently sample an ionization signal using the at least one ion sensor and determine at least one engine operating parameter using the at least one engine sensor during a combustion event,
• transform the ionization signal with a wavelet filter,
• extract at least one critical signal feature from the ionization signal that has been transformed by the wavelet filter, and
• process the at least one critical signal feature and the at least one engine operating parameter through an artificial neural network to determine at least one combustion characteristic of the engine;
• and
• wherein the controller controls the at least one output device of the internal combustion engine based upon the at least one combustion characteristic of the engine.

A system to determine a combustion characteristic of an internal combustion engine, comprising:

• a controller operably connected to:
• at least one output device of the internal combustion engine,
• at least one ion sensor operable to measure an ionization signal from a combustion event of the internal combustion engine, and
• at least one engine sensor operable to measure at least one engine operating parameter;
• wherein said controller is operable to:
• concurrently sample an ionization signal using the at least one ion sensor and determine at least one engine operating parameter using the at least one engine sensor during a combustion event,
• transform the ionization signal with a wavelet filter,
• extract at least one critical signal feature from the ionization signal that has been transformed by the wavelet filter, and
• process the at least one critical signal feature and the at least one engine operating parameter through an artificial neural network.