Sunday, September 3, 2017

Terahertz in Bulk Food Inspection

                                                                  Andrew Eddleman, Acquire Automation

http://www.powderbulksolids.com/article/4-Key-Insights-to-Bulk-Food-Inspection-08-29-2017

Detection of contaminants in a bulk flow of product has been, traditionally, very difficult. Usually, this type of inspection is for dry bulk food items, such as grains, beans, vegetables or fruits. These inspections are looking for contaminants such as rocks, wood (usually from pallets), pieces of metal, packaging, plastics, and tape. However other extreme cases may also need to be identified. Below are some insights to inspect for these types of contaminants and what to know when spec’ing out a vision system.

1. Environment: Thermal Concerns, Moisture, Lighting
The environment that these types of inspections play is a major concern when designing a bulk food inspection. Creating a system for a full spray washdown environment is usually mandatory. These inspection systems need to be stainless steel and have either IP67 or IP69K enclosures for all electronics (including lighting). For lighting, this can be a challenge as a custom enclosure will need to be designed or a specialty light will need to be accrued (these are not very common for machine vision lights).  Moisture can also play a factor during the inspection if the product is wet. Moisture can cause light to reflect off product in unpredictable ways, which can affect the overall reliability of the results. Mitigating these effects with software or lighting is almost mandatory in most cases. Heat can also play a factor in the longevity of the equipment. Heat can come from an oven, fryer, or just from an unair-conditioned space (usually during summer months). Since most vision equipment has a max temperature threshold of around 115°F, camera and lighting equipment, along with associated electronics must be vented, insulated, or cooled appropriately.

2. Product Differentiation: Defects, Contamination, Etc.
Since most bulk food products have been exposed to an outside environment at one point and usually have been shipped on pallets and trucks, the variability of what the product can be exposed to is usually high. Many times, contamination from shipping materials used, such as a pallet, bag, or tape, is the main culprit of contamination. At other times, wildlife from the surrounding facility may be of concern. Occasionally, contamination from the processing equipment itself (bolts, nuts, Delrin, etc.) is the problem. Whatever the root cause of the issue is, a wide variety of defects can exist and, therefore, multiple strategies for detection need to be deployed. X-rays can be used to look through products for heavier or denser materials, metal detectors can be deployed to look for metals, and machine vision can look for product detects, as well as surface level contaminants, such as wildlife or packaging material. In each of these cases, the technology used is often chosen by the abundance of the type of problem and the relative cost of the equipment. For example, vision systems can be a low-cost alternative to metal detectors or x-rays, but if detecting contaminants under a product is warranted, then traditional machine vision would not be up to the challenge.

3. Analytics: Tracking, Reject, and Data Collection
With the advent of Factory 4.0 and advanced manufacturing, more and more data is being required to detect, reject, and characterize contaminants in product. Whether that is from wildlife for “extermination” purposes or from packaging for transportation improvements. Tracking to a reject point is usually crucial for any inspection equipment. For bulk food, this can be tracking to a gap in the conveyor transport so that a blow-off mechanism can be utilized, a pull nose conveyor or just a simple alarm that stops the transport mechanism. When the product or contaminant is rejected from the bulk-flow the failure is either recorded in a database or alarms an operator to the condition. Data analytics then can be utilized to track the overall progress of operations (an example would be a scada system) or provide more advanced analytics about the process and give insights about future improvements.

4. Advanced Quality Inspections: 3D, Spectral Imaging, Terahertz Imaging, and More
New technologies are being introduced every year to increase the quality of bulk food inspections. 3D technology can be utilized for counting operations as well quality inspections such as volume measurements. Spectral and hyperspectral imagers can detect defects in products like sugary ends in potatoes and bruises in apples. Terahertz cameras can be used to eliminate the need for X-rays in inspection and to detect defects or contaminants hid by packaging or embedded in the product itself. Increasingly, this advanced inspection equipment is being utilized by quality departments to do inline measurements during production or detecting common defects in novel ways.
    

Andrew Eddleman is applications engineer at Acquire Automation, Fishers, IN. He has more than 10 years of experience in machine vision design, developing everything from hardware designs to software programming. In his current role as an, he works with customers across the country on a diverse array of machine vision applications such as 3D technology and Deep Learning. For more information, visit acquireautomation.com.

No comments: