PROJECT ON OPTIMIZATION AND PRODUCTION QUALITY CONTROL BY USING ELECTRICAL CAPACITANCE TOMOGRAPHY AND OPTICAL DETECTION TECHNIQUES
Modern production processes are increasingly complex, while customers are demanding higher quality products at the lowest price. This situation increases the importance of process optimization. Products and requirements are changing faster and faster, technologists have less and less time to learn the process and to optimize it solely on the basis of their own experience. On the other hand, very often the processes are automated and metered in detail, so we have plenty of data describing it. In such a situation we cannot give up the chance of decision support. Analysis of the data may be used for process optimization in different ways and it can affect many aspects of process management. We can detect disturbances in the process, find the causes affecting the problems with quality, and choose the optimal settings for the process, comparing different preparation procedures and many others. It is worth noting that in the processes, there are many people involved in with different tasks and permissions. Therefore, data analysis tool should provide access control. The research project assumes creation of two measurement platforms: a set of multiphase flow system and a mini production line. In the first case we will be analysing two-phase flows of liquid (water) and air. That type of flow is commonly used in chemical reactors where air is mixing substances. In the second case, we will be analysing moving objects on the mini production line using RFID tags and image detection techniques. The system will be also using biometric reader’s witches will provide access control.
Electrical Capacitance Tomography
Electrical Capacitance Tomography (ECT) is an imaging technique invented in the eighties of the last century. The common approach to tomographic measurements data analysis is their processing in order to obtain a reconstructed image of the distribution of a required property in the examined cross section area. In the case of ECT, this property is the electric permittivity. The visualisation of the inside of the investigated body is obtained on the basis of the recorded capacitance measurements. This permittivity distributions could be directly related to the material concentration present in the sensor (probe) space. In a typical measurement ECT sensor use between 8 and 16 electrodes symmetrically spaced around cylindrical container (tube, pipe, etc.), as shown in the following figure (fig. 1).
Fig. 1. Schematic diagram of the probe in the electrical capacitance tomography.
The main problem to obtain good resolution images in the ECT, is measuring the capacity in the order of femto Farads. In modern electric capacitance tomography system to obtain such accurate measurements, method of “charge-discharge” is being used. It is based on cyclic charging and discharging of the capacitor, with the registration of the charging current, which is directly proportional to the measured capacitance. It is mainly used to control the quantity and quality of materials transported in pipelines (petrochemical industries). Its main advantages include: high speed tomographic imaging (about 200 scans per second), ease of installation on the measuring system, as well as the low level of complexity. Alternatively, there are used techniques based on X-rays or ultrasonic waves, however, this type of solution is expensive and troublesome to implement. Development of electric capacitive tomography is related to the development of algorithms and methods for image reconstruction. Despite the many successes in application, possibilities of this method are not fully utilized.
The hardware concept of multiphase flow system
The basic measurement platform is set up to study multiphase flows. Schematic set is shown in the figure below (Fig. 2).
Fig. 2. Sketch of multiphase flow system. Elements marked on the sketch are: 1 – expansion tank, 2 – protection container, 3 – throttle, 4 – pump, 5 – transparent Plexiglas elements (for ECT electrodes), 6 – steel frame, 7 – no transparent pipes (PVCU), 8 – compressed gas (air), 9 – valve.
The design system allows the study of vertical and horizontal flows. Application of control elements such as throttle, pump or valves (air), allow precise repeatable measurement conditions. This configuration admits the study of counter current flows, which are very important in real industrial installations. The transparent elements provides use of the optical detection system. At the same time, multiphase flow system can be used for correlative ECT measurements using two sets of electrodes. This measurement system is used, not only to study different types of multiphase flows, but mainly to develop and test new process tomography algorithms. One of the main objectives of this project is to establish a suitable system to control and optimize production process. The following figure (fig. 3) shows a schematic of the system, which consists of an optical detection system based on industrial IP cameras and electrical capacitance tomography system.
Fig. 3. Schematic of the measurement system to study the multiphase flows.
Simultaneous use of two research techniques (tomography and optical detection) allows for precise analysis of the flow processes.
The hardware concept of the mini production line
The main part of the measurement system is closed (oval) production line with adjustable speed. The following figure shows a diagram of the measuring system (Fig. 4).
Fig. 4. Sketch of the mini production line system.
The measurement system includes IP cameras, RFID readers, biometric readers and vision recording system (DVR). Control access to the production control panel line is protected by the biometric system.
Optical detection system as in the previous system is based on the IP cameras. The program will be analysing images from the cameras in real time. The analysed objects will recognized by their characteristic features such as: shape, size, colour, etc. At the same time the image will be saved on the disk array. Additional identification of objects will be performed by the programmable RFID tags placed on their surface. Such a system will provide not only accurate modelling in micro scale production facility, but will also be a base for the development of fast vision analysis algorithms.