Thesis for the Degree of Licentiate of Technology


Year 1995
Arto Tolvanen

Control System Development for a Pressure Filter

In this thesis a vertical pressure filter (PF) is studied in order to make the PF operation more efficient and adaptive against the disturbations arising from the variations in the fees slurry quality.

The PF operation takes place in a sequential manner where one filtration cycle consists of three main stages i.e., fees of filtration, pressing and air drying stages. The basic control system is included in a programmable logic control (PLC) environment and it remains as the preferable implementation platform for the developed control strategies. However, during the project a PC based automation software is utilised due to the inflexibility of the PLC environment.

A short survey on the theory concerning the pressure filter modelling is presented.

An empirical modelling utilising statistical test design and regression analysis is carried out. The test runs are performed with an industrial size pressure filter. The development models for the capacity of the filter and cake moisture content are used to optimise the operation of the PF. An adaptive type control configuration, that maximises the capacity when the upper limit of the moisture content is set, is derived. The control configuration is a look-up type table from where the values for the control variables can be fetched according to the moisture setpoint and measured feed density value (measurable disturbance).

A fuzzy controller for the PF is developed in order to derive a more general type control strategy for the PF. The operator control actions and expert knowledge are utilised to build the rulebase for the fuzzy controller.

This info last modified 23 Sep 2017 - contact webmaster