Lisses, France--Laser technology has proven successful when cleaning statues and historical monuments--in-doors. Outside on the street is a different matter. Changing environmental conditions in general, and temperature variations in particular, can cause havoc with laser beam stability.
An open-loop correction system, based on neural network technology, promises to remedy this situation. A joint effort between Neural Computer Sciences (NCS, Southampton, England) and B.M. Industrie of France, the system "learns" the relationships present in data through a training process and, once trained, responds accurately to new input data.
Neural network technology is appropriate, says Chris Isbell, NCS project manager, because the relationship between ambient conditions and laser output are imperfectly understood. Temperature variations, for example, can influence laser beam power, quality, and most importantly, pulse beam profile.
Isbell explains: "If power intensity is plotted across the laser beam from edge to edge, a near Gaussian curve will be seen. Ideally, that bell-shaped curve should be symmetrical for maximum effect. If it appears lopsided, the power or internal optics must be corrected."
In order to implement corrective action, B.M. Industries has defined the system's performance-related parameters and has collected training data. NCS is developing the neural net software, its associated embedded control board, and supervisory system software. The resulting system is built into the laser to maintain its performance under conditions common to building cleaning operations.
Funded, in part, by the European Union's Eureka program, the international RESTOR project hopes to offer a system suitable for cleaning 10m(super2) of building facade per hour. Benefits: no erosion from sandblasting; no discoloring from chemical cleaners; no recycling or purification of water jets.