Eaton Intelligent Power Limited

Johannes Van De Belt

Keith Nolan

Niall Cahill

Daniel Zucchetto

Salma Abdulaziz

Ruth Quinn

James Ryle

Maebh Larkin

A method for time alignment of sensor having a plurality of sensors for use in sensor data integration. A reference signal, with timestamp, is provided to each sensor. Output signals in response to the reference signal are identified and used to determine a time delay for response for the sensors. These time delays are stored applying time corrections to sensor data during sensor data integration used in event detection. An associated method of detecting events from sensor data from a plurality of sensors is disclosed. A threshold is determined for each of the sensors, such that a signal exceeding the threshold is identified as a potential sensor event. Potential sensor events from each sensor are detected and when potential events from at least two sensors fall within a predetermined time window this is identified to be a likely actual event.

Sensor data integration and event detection

It is known for a product to be mass produced by way of a manufacturing process. Typically, a quality control step is used in a manufacturing process to monitor the quality of manufactured products. However, quality control procedures in manufacturing are typically labour intensive. A technician or other person must inspect the product and carry out any necessary tests. The present disclosure provides a surface roughness measurement system and method for determining a surface roughness of a product with an imaging system, a coherent light source, a light sensor and several trained machine learning algorithms.

Surface roughness measurement system

A data transmission and compression method arranged to compress and transmit data between an edge device and a remote server. The method includes collecting data at the edge device, wherein the data is attributed to a plurality of signal signatures, generating a data matrix at the edge device, and transforming the data matrix, wherein transforming the data includes using a wavelet transform. The method further includes compressing the data, wherein compressing the data includes utilising an autoencoder, and transmitting the encoded compressed data to the remote server via a communication channel. The method further includes, at the remote server, decompressing the data utilising an autoencoder, reconstructing the signal signatures using an inverse wavelet transform, and storing the reconstructed data signatures in a datastore on the remote server.

Compression of multimodal sensed signals

A method characterizes a noise event, and includes locating microphones in an environment and generating a training event at a location in the environment as a reference for a noise event. A sound sample is recorded at each microphone and phase differences between the sound samples are used to establish a noise event signature for an event at that location. A noise event may subsequently be identified by taking sound samples from the microphones associated with the noise event and using phase differences between them to identify the noise event by matching against noise event signatures.

Growing community of inventors

Ballsbridge, Ireland

Johannes Van De Belt