10174003AcouBatt LTD MVP DevelopmentClosedFast Start ResponseInnovate UKThis project is the fruit of my PhD research, focused on using acoustic technology with integrated bespoke machine learning algorithms to improve lithium-ion battery (LIB) manufacturing, diagnostics, and failure detection. This technology utilises Acoustic Emission (AE) piezoelectric sensors strategically placed on batteries to detect elastic waves generated by localised energy releases, such as those caused by chemical reactions, gas evolution, structural changes, or potential failure events within the cell. The sensors, connected to a filter, preamplifier, and computer, utilise a tailored custom algorithm to process acoustic signals and identify events during battery usage. By recognising reaction-specific acoustic waveforms and patterns, manufacturing protocols can be refined, significantly reducing processing times and costs. The machine learning component enhances this analysis by classifying signals in real time, identifying anomalies, and detecting early signs of failure. This predictive capability improves quality control, enhances battery safety, and ensures the long-term reliability of LIBs. The technology is non-invasive, scalable, and cost-effective, making it highly suitable for industrial adoption. It integrates acoustic signal analysis with electrochemical performance insights and is validated through advanced complementary techniques such as synchrotron radiography, gas volume measurements, and microscopy. By streamlining formation cycles, enhancing cell consistency, and reducing defect rates, this technology addresses a bottleneck in the battery industry. Beyond optimising the formation process to improve performance and reduce costs, the real-time detection of potential failures provides an additional layer of safety assurance.