Signature image processing

From Wikipedia, the free encyclopedia

The steel box that houses WeldPrint, which uses SIP for the industrial analysis of weld quality
The steel box that houses WeldPrint, which uses SIP for the industrial analysis of weld quality

Signature Image Processing (SIP) is a technology for analysing electrical data collected from welding processes—usually automated, robotic welding. In developed countries, some form of welding is used in more than 50% of manufactured products. Acceptable welding requires fine tuning and exact conditions; even minute variations in conditions and control can cause a weld to be unacceptable. Thus, there has been a need for a robust, reliable, real-time welding fault detection, especially in safety-critical applications in automotive manufacture. SIP allows welding faults to be identified in real time, measures the stability of welding processes and enables welding processes to be optimised. The advance was possible only through the advent of more powerful PCs. The quality monitoring of automatic welding can save production downtime, reduce the need for product reworking and recall, and is becoming increasingly important as car makers increase the number of robots and reduce the number of human operators in their plants. The technology is used by eight auto component manufacturers, including leading companies in Australia and the EU, in which it has transformed welding practices, significantly improving the industrial process and the quality of the products.

Contents

[edit] Development

The idea of using sophisticated algorithms to assess the quality of the welds produced in robotic manufacturing emerged in 1995 from research by Associate Professor Stephen Simpson at the University of Sydney on the complex physical phenomena that occur in welding arcs.

Associate Professor Stephen Simpson, the inventor of the signature image processing technology
Associate Professor Stephen Simpson, the inventor of the signature image processing technology

The critical realisation was that a sophisticated way of determining the quality of a weld could be developed without a definitive understanding of those phenomena. This led to a new way of analysing the data: SIP.[1][2][3] The development involved five key innovations:

  • new methods for handling sampled data blocks by treating them as phase-space portrait signatures with appropriate image processing
  • the recognition that quantitative comparison of signatures was necessary and possible for weld-quality monitoring
  • novel techniques for analysing welding signatures based on statistical methods from the social sciences, such as principal component analysis
  • the development of algorithms and mathematics appropriate for real-time welding analysis on personal computers
  • the multidimensional optimisation of fault-detection performance using experimental welding data.

Unlike conventional systems, which log information for later study or use X-rays or ultrasound to check samples, the new technology looks at the electrical signal and detects faults when they occur. Data blocks of 4,000 points of electrical data are collected four times a second and converted to signature images. After image processing operations, statistical analyses of the signatures provide quantitative assessment of the welding process, revealing its stability and reproducibility, and providing fault detection and process diagnostics.[4]

SIP is the scientific basis for the WeldPrint system (owned by the University of Sydney), which was developed for arc welding with the assistance of an Australian government R&D Start grant (1999–2001), after support by the Australian Research Council for the fundamental research (1997–2001). The system consists of a front-end interface and software based on the SIP engine. From the user’s perspective, a key advantage is that the system relies on electrical signals alone. It is non-intrusive, simple to set up, and sufficiently robust to withstand harsh industrial welding environments. The first major purchaser of the technology, GM Holden[5][6][7] provided feedback that allowed the system to be refined in ways that increased its industrial and commercial value. Improvements in the algorithms, including multiple parameter optimisation with a server network, have led to an order-of-magnitude improvement in fault-detection performance over the past five years. At the same time, the simplicity of the software user interface belies the complexity of the underlying mathematics of SIP.

[edit] Industrial use

The technology in use on the shop floor of Melbourne firm Unidrive, which alone has used WeldPrint to monitor the quality of steering-column component welds in more than half a million Australian vehicles in the period 2001–06
The technology in use on the shop floor of Melbourne firm Unidrive, which alone has used WeldPrint to monitor the quality of steering-column component welds in more than half a million Australian vehicles in the period 2001–06

Globally, more than 70 million passenger vehicles are built each year, each containing typically around 50 metres of continuous arc welding in its subassemblies. WeldPrint for arc welding became available in mid-2001. The innovation has enabled companies to significantly improve the quality, durability and safety of their vehicles, with considerable cost savings in the avoidance of recalls to fix the large proportion of systemic quality problems that arise from suboptimal welding. About 70 units have been deployed since 2001; of these, about 90% are used on the shop floors of automotive manufacturing companies and their suppliers. The industrial users include Lear (UK), Unidrive, GM Holden, Air International and QTB Automotive (Australia). Units have been hired to Australian companies such as Rheem, Dux, and OneSteel for welding evaluation and process improvement.

The WeldPrint software received the Brother business software of the year award (2001); in 2003, the technology received the A$100,000 inaugural Australasian Peter Doherty Prize for Innovation;[8] and WTi—the University's original spin-off company—received an AusIndustry Certificate of Achievement in recognition of the development.

SIP has opened opportunities for researchers to use it as a measurement tool both in welding[9] and in related disciplines, such as structural engineering.[10] Research opportunities have opened up in the application of biomonitoring of external EEGs, where SIP offers advantages in interpreting the complex signals[11]

[edit] See also

[edit] References

[edit] Notes

  1. ^ Simpson SW and Gillespie P (1998) ‘In-process monitoring of welding processes—a commercial success’, Australasian Welding Journal, 43, 16–17
  2. ^ Simpson SW, Weld quality measurement, WIPO PCT WO9845078 (1998); US 6288364 (2001); Australia 741965 (2002); Europe (14 countries) 1007263 (2003); Canada 2285561 (2004); South Korea 0503778 (2005)
  3. ^ Simpson SW, Welding assessment, WIPO PCT WO0143910 (2001); Australia 763689, US 6660965 (2003); Canada 2393773 (2005); PAs: Japan 2001-545030 (2001); China 00817251.X, S. Korea 2002-7007624, India IN/PCT/2002/00740 2002), Brazil PI0016401-1, EU 00984649.4 (2002)
  4. ^ Simpson, SW (2007) ‘Statistics of signature images for arc welding fault detection’, Science & Technology of Welding and Joining, 12(6), 557–64
  5. ^ ‘Holden orders award-winning weldprint welding technology’, Techwatch, Price Waterhouse Coopers, 12(6), 2002,
  6. ^ ‘Holden purchases award winning weldprint welding technology’, Australian Technology Showcase http://www.techshowcase.nsw.gov.au/ News and Events (2002)
  7. ^ ‘University weld checker to be used by Holden’, Australian Innovation Magazine, 3–5/02, 29
  8. ^ ‘Bright sparks join forces to take out Doherty Prize’, The Australian (national newspaper)—Higher Education Supplement, 2 April 2003
  9. ^ Nguyen NT, Mai Y-W, Simpson SW and Ohta A (2004) “Analytical approximate solution for double-ellipsoidal heat source in finite thick plate”, Welding J, 83, 82s
  10. ^ The LH and Hancock GJ (2005) ‘Strength of welded connections in G450 sheet steel’, J Struct Eng, 131, 1561
  11. ^ Car plant technology has medical spin-off, UniNews, USyd, 34(1), 1 (2002)