In a new development, researchers at Penn College of engineering have developed an algorithm in a bid to address difficulties faced by human to recognize and analyze patterns in dynamic systems. This could be patterns of both natural as well as human made systems such as factory machinery or human heart.
To develop the algorithm, the researchers focused on to understand patterns in non-linear dynamic systems. This is because these systems are difficult to analyze due to their nature – they fluctuate on being subject to multiple dimensions such as space and time. And, for these reasons are unable to be understood via human observation.
The algorithm developed as a joint effort of researchers and academicians from some prestigious institutions is published in the Chaos journal of American Institute of Physics.
Algorithm serves to decipher Man-made and Natural Pattern
“The methodology analyzes different kinds of recurrences in data in order to provide a better understanding of the world around us,” said one of the researcher. The algorithm helps to build a bridge between biological patterns such as human anatomy, and in man-made patterns such as manufacturing.
To create the algorithm, the team examined spatial data in complex, microscopic images generated by ultra-precision machining. For example, UPM, a manufacturing process is widely used in modern industries such as aerospace and semiconductors to generate highly precise cuts or polishing. The process employs single- crystal diamond tools to refine metal workpieces at atomic scale.
During observation, the spatial data showed a number of surfaces over UPM images, from flat to rough to severely rugged. This led to the conclusion, god quality products feature a similar surface, and bad quality products might feature different textures on the surface.