The modern applications of electrostatics range from precipitation of dust, electro-photography, separation of granular material, atomisation, and spraying, to charge elimination, micro-fluidics, and DNA manipulation [4, 6, 13, 16, 45, 51, 74, 76, 88, 101]. All these electrostatic processes require the effective control of various electrical and mechanical factors, which are often strongly interrelated. In addition to this, the high sensitivity to environmental perturbations increases the overall variability of electrostatic processes, turning their optimisation into a difficult task.
This work has been driven by the belief that thorough observation and enhanced measurement capabilities can provide the appropriate answers to the problems faced by the scientists and the engineers involved in the research and development of electrostatic processes [32, 78, 82, 83]. The accurate sampling procedures and signal analysis techniques required for rigorous control of industrial processes cannot be reasonably deployed without integrating computer-enhanced instrumentation [86, 92]. By fusion of several measurement data fluxes, virtual instrumentation is the solution for monitoring the input variables and the responses of complex electrostatic processes. Such statistical methods like the "Design of Experiments", which is aimed at rising experimental efficiency [36-37, 47, 84, 90, 104-106, 115-117, 124-125], and the "Statistic Process Control" technique, which is focused on reducing process variation [43, 80, 83, 87, 100], can be used - in conjunction with reliable measurement techniques - to detect, define, and isolate perturbation sources.
The interest industry presently shows in electrostatic processes justifies the orientation of this thesis towards the research and development of novel measurement methods and virtual instrumentation that can empower users with a better understanding of the complex phenomena the new applications are based on. Adapting measurement techniques to match the high-voltage / low-current characteristics of electrostatic processes requires particular attention when the purpose is to identify perturbation sources. On the other hand, getting together statistic methods and virtual instrumentation is likely to enhance the capability of electrostatic processes, by reducing their sensibility to uncontrolled variations of the operating conditions.
The first chapter of the thesis attempts to give a brief state-of-the-art of the measurement techniques that are or could be employed for the study of electrostatic processes. In this way, the novel measurement methods and instruments developed by the author are put in the right perspective.
The first virtual instrument introduced in Chapter 2 of the thesis was developed for the charge measurement of finely divided matter. The instrument facilitated the use of statistic process control techniques to improve the operation of an electrostatic painting gun.
Aimed at monitoring the discharge of granular nonconductive materials, the surface potential measurement technique described in chapter 3 is based on the use of an electrostatic voltmeter leaded towards defining an analytical method for outcome determination of an electrostatic separation process in Chapter 3. The primary experimental results on a PE/rubber mixture allowed us to partially challenge a common belief that only tribo-electrostatic methods can insure the separation of nonconductive mixtures.
Analysing interactions between process factors like metal content, high voltage AC and DC characteristics, humidity etc. allowed developing a new method for metal content measurement in Chapter 4. The use of computer based Virtual Instrumentation was the catalyst of this development that would not have been observed without.
Finally, Chapter 5 approaches modelling separation process while it studies linear interaction models, robust control, and environmental influence on the metal/insulator electrostatic separation.
The conclusion of the thesis, its contributions to the field, and the further perspectives makes for the final part.