Piezoelectric Inkjet Drive Waveform Creation Using Flight Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing drive waveforms for liquid ejection heads, particularly those using piezoelectric elements, are inefficient and require high-level knowledge and experience, making it difficult for technicians without such expertise to create suitable waveforms for varying ink properties.
Innovation Solution
A method utilizing machine learning models trained on actual flight shape data to predict and determine optimal drive waveforms for liquid ejection, incorporating techniques like autoencoders and variogram analysis to efficiently find suitable waveforms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physical simulation techniques (equivalent circuit model or CFD) are used to predict flight shape, then prediction capability is achieved, but high-level knowledge and experience in fluid dynamics and computation are required
Solution Approach 1:
The patent replaces complex physical simulation systems (CFD, equivalent circuit models) with a machine learning model that has been trained on flight shape data. This substitution eliminates the need for users to possess high-level knowledge in fluid dynamics and computation, while maintaining accurate prediction capability through data-driven learning patterns.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between the drive waveform input and flight shape prediction output. This intermediary has been trained to capture the complex relationships between waveforms and flight characteristics, allowing users to obtain accurate predictions without needing to understand the underlying physical simulations.
2Productivity
If a drive waveform group is prepared in advance and evaluation is performed, then optimization can be conducted, but an enormous amount of time is required due to trial and error
Solution Approach 1:
The patent performs preliminary training of the machine learning model using a comprehensive set of drive waveform data before actual optimization is needed. This preliminary action allows the model to learn optimal waveform patterns in advance, so that during actual use, technicians can quickly obtain optimized waveforms without time-consuming trial and error evaluation of multiple candidates.
Solution Approach 2:
The patent creates a virtual model (machine learning model) that copies and simulates the complex relationship between drive waveforms and flight shapes. Instead of performing repeated physical trials and evaluations, the system uses this copied model to predict outcomes instantly, dramatically reducing optimization time while maintaining accuracy.
3Productivity
If a pre-defined drive waveform group is used, then optimization can be performed, but it is impossible to search for a completely unknown drive waveform
Solution Approach 1:
The patent implements a dynamic machine learning model that can adapt to and generate completely unknown drive waveforms. Unlike static pre-defined waveform groups, the model learns from training data and can predict flight shapes for any novel waveform input, enabling both optimization of known patterns and discovery of entirely new effective waveforms that were not previously in any predefined set.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables technicians without specialized knowledge to efficiently create drive waveforms suitable for ejecting liquids, reducing the time and effort required for optimization.
Implementation Method 1
a liquid ejection head that ejects liquid by driving a piezoelectric element
Data Source
AI summary
A drive waveform creation method, an information processing apparatus, and a program that enable even a technician not having professional knowledge to efficiently create a drive waveform suitable for ejecting liquid to be used.A method of creating a drive waveform to be used for driving a piezoelectric element of a liquid ejection head including the piezoelectric element includes, via one or more processors, predicting flight of liquid to be ejected by the liquid ejection head in a case of inputting an unknown drive waveform using a machine learning model that is trained through machine learning using data related to an actual flight shape of the liquid in a case where each of a plurality of drive waveforms is applied to the piezoelectric element using the liquid and the liquid ejection head, and determining a drive waveform suitable for ejecting the liquid based on the prediction of the flight.