Cavity Detection System

Refrigerators are insulated with polyurethane insulating foam. During the manufacturing of refrigerators, polyurethane is injected between the inner cabinet and outer housing of the refrigerator as a liquid, and it expands over 30 times its original volume in the short period of time to fill the cav

Project Title

Cavity Detection System

Project Area of Specialization

Electrical/Electronic Engineering

Project Summary

Refrigerators are insulated with polyurethane insulating foam. During the manufacturing of refrigerators, polyurethane is injected between the inner cabinet and outer housing of the refrigerator as a liquid, and it expands over 30 times its original volume in the short period of time to fill the cavity between the inner cabinet and outer housing and provides insulation which is extremely airtight. This process is called the foam filling process. However, during this foam filling process, the air is trapped inside the polyurethane foam which causes heat conduction and decreases the efficiency of the refrigerator. The industries use destructive testing methods to detect the cavities formed during the foam filling process. These processes are cost-inefficient, time-consuming, and cause a great loss of produced units. The goal of this project is to develop a suitable system for the detection of these air pockets formed within the walls of the refrigerator efficiently so that they can be treated. To avoid the loss of produced units in bulk, active thermography, which is a non-destructive testing method, is proposed in this research. The active thermography is performed on the produced units to obtain thermograms. These thermograms are inputted to our proposed system to get the localized coordinates of the hot spots, indicating the air pockets.

Project Objectives

Presently, the industry makes use of the Acceptable Quality Limit (AQL) which corresponds to the ISO 2859 standard. A certain limit is set, subsequently, samples from the produced lot of units are tested destructively. This gives rise to two scenarios, the first involving failure of tests in which the whole batch is rejected, while the second relates to the possibility that a passed batch may contain faulty units. This is due to the fact that only a small ratio of the total units is checked.

The objective of the project is to design a solution for the refrigerator industry and replace the destructive testing methods used in quality assurance with a non-destructive method that makes use of contemporary technologies. Our project will not only decrease the loss due to the AQL method but also allow each unit to be individually checked for faults further increasing the efficiency and performance of the refrigerator units.

Project Implementation Method

Our methodology comprises three steps discussed below. 

1. Active Thermography: Infrared radiation (IR), or infrared light, is electromagnetic radiation that is not visible to the human eye but can be felt as heat. Infrared active thermography is a versatile tool and is an efficient method to determine the temperature of an object as well as to detect any anomalies in the object under inspection. Active thermography is a technique that makes use of an external energy source to create temperature variance in the component or surface which is influenced by interior materials and defects. This external energy source may comprise a heat element or source such as ovens, halogen lamps, and laser heat sources. This work has adopted the use of halogen lamps as they tend to heat a broader surface in a cheaper and safer way.

2. Experimental Setup: The experiment consisted of an external heat source comprising a load bank. The load bank made use of three halogen rods each rated for 500 Watts. The initial experiments were done on the unit without altering the wall insulation. However, in the subsequent experiments, the polyurethane insulation within the refrigerator cabinet was removed from certain areas. The refrigerator was then heated externally to capture the thermograms. The thermal imaging module utilized to capture the displayed thermograms in this experiment was the FLIR E8 infrared camera.

3. Image Processing: The thermogram of the refrigerator is imported into the image processing algorithm. Image enhancement techniques is performed on the thermogram. This image is then converted into the HSV color space to define the color range of the hot spot since images in the HSV color space are able to detect colors more accurately as compared to RGB color space. An example is an image where casted shadows are treated as a separate entity in the RGB color space. HSV color-space on the other hand would treat shadows on a certain color as a variation of the “value” (brightness) component while keeping the “Hue” (color) component constant. As shown in Figure-5. The lower and upper values of the colors in consideration are then set in accordance with their HSV counterparts. A mask is created from the thermogram and is used to detect the hot spots in the original thermogram. Finally, the contour around the segmented region is measured and a rectangular box is overlayed on top indicating the presence of a detected hot spot. This is done by first calculating moments and the centroid of these moments in the detected image. Image moments are defined as a set of statistical parameters which include calculating the area of a region to measure the distribution pixels and their intensities.

Benefits of the Project

Cost-Effective:  Solves the common problem of rejected lot faced by the industry using conventional methods. Time and cost-efficient for the industry.

Standalone Solution: Compact form factor and efficient in both power usage and anomaly detection.

Versatility: Not limited to a certain IR Imaging Module. Can be used with a number of Thermal Imaging Cameras.

Integration: Can be integrated with the existing system for interaction with local records/database.

Improved Testing: Testing Improving the industrial standard of testing. Introduction and application of technology-based solutions.

Applications: Real-time solution using Thermography and Image processing for similar problems (Crop Protection from Wild Animals).

Technical Details of Final Deliverable

The Final Deliverable will be in the form of a plug-and-play solution. We will provide an Integrated Hardware-Sofware solution that includes the following components:

1. Hardware Box: This includes the Raspberry Pi computer, Interactive buttons, Input/Output port, and internal wiring.

2. I/O: The I/O consists of the display monitor, mouse, and easily accessible USB-A ports.

3. Interactive GUI: An easy-to-use Graphical User Interface with the detection algorithm working in the background.

4. Thermal Camera: A thermal camera that can easily be used with the controller for capturing thermograms.

Final Deliverable of the Project

HW/SW integrated system

Core Industry

Manufacturing

Other Industries

Core Technology

Others

Other Technologies

Sustainable Development Goals

Industry, Innovation and Infrastructure, Responsible Consumption and Production

Required Resources

If you need this project, please contact me on contact@adikhanofficial.com