42 lines
5.2 KiB
TeX
42 lines
5.2 KiB
TeX
\chapter{Conclusion and Outlook}\label{chap:sum}
|
|
|
|
The state of this project upon completion can be analyzed by recalling the research question and requirements set out in \autoref{chap:intro}. The research question being: \textit{How can a cheaper and easier to install measurement system for bulk material flow on a conveyor belt be designed?}
|
|
|
|
A breakdown of the various factors that determine the suitability of the implementation presented in this project:
|
|
\begin{description}
|
|
|
|
\item[Sensor Suitability] \hfill \\ The wavelength of the infrared laser used in this project of \SI{860}{\nano\meter} was shown to be unsuitable for use with the conveyor belt during the on-site testing. This is most likely due to the absorption spectrum of the belt material that had very flow reflectivity at this infrared wavelength. The similarly black colored belt used in laboratory testing however was visible to the LIDAR sensor. A further study of belt materials commonly deployed in the field is necessary.
|
|
|
|
\item[Temperature Suitability] \hfill \\ On the higher end of the temperature range, the LIDAR sensor used in this project is the limiting factor. The maximum temperature of \SI{30}{\celsius} is easily exceeded in particularly hot weather or even in direct sunlight. Design of the housing must account for adequate cooling, as well as reflectivity, should the system be deployed in view of direct sunlight.
|
|
|
|
\item[Hardware Suitability] \hfill \\ The Raspberry Pi provided sufficient processing power in order to develop, test and deploy the prototype software. The flexibility of the Linux platform also grants sufficient flexibility in order to easily add further functionality---i.e. a web server or other interface---or modify existing functionality.
|
|
|
|
The netHAT was also shown to be performant and stable during testing. Combined with the Raspberry Pi, it provides a low cost platform to bring IoT to Industrial Networking.
|
|
|
|
\item[Process Suitability] \hfill \\ The process flow which was developed with ease-of-use in mind was shown to be beneficial during both laboratory and field-testing. Once the sensor and Pi units were set up on the belt, the engineer could continue working on the configuration of the system remotely---or even externally through the use of a VPN.
|
|
|
|
\item[Software Suitability] \hfill \\ The software architecture as laid-out in \autoref{chap:design} was shown to fulfill the requirements of the process. The individual software elements and libraries such as ZeroMQ and Qt were shown to be robust enough to carry out the configuration process. The software was shown to successfully combine the separate process elements into a single workflow which simplified the configuration process.
|
|
|
|
\item[Cost Suitability] \hfill \\ At a development cost of just under €600---even at a profit margin of \SI{500}{\percent}---the system is still able to remain competitive with conventional systems in use in the industry today\footnote{See \autoref{table:cost}}.
|
|
|
|
\item[Housing Suitability] \hfill \\ The housing designed for the field-testing stage of this project is only suitable as a prototype. A more robust housing must be developed out of more durable materials, and account for weather and vibration.
|
|
\end{description}
|
|
|
|
\section{Project Status and Feasibility}
|
|
Although a commercially viable product was not realized during the duration of this project, great strides were nevertheless made towards this goal.
|
|
|
|
This work has shown that commercially available products do have the necessary performance to carry out the operations required to deliver the intended results.
|
|
|
|
It has also shown, that the software development methods used, methods that are more common in commercial software development rather than industrial development, are highly flexible and powerful. This was naturally enabled by the availability of the Linux kernel on smaller and more performant platforms. This work shows the strong potential of merging the industrial world with that of more conventional software development, also known Industrial IoT.
|
|
|
|
Discounting the setbacks faced during field-testing due to sensor and material issues, laboratory testing has shown that a fully-functioning result of this project is in theory, feasible.
|
|
|
|
\section{Work Remaining}
|
|
One or two more iterations of development are required in order to fully realize this project. The fundamental operations have already been developed and tested, namely the analysis of the cross-sectional area and belt velocity.
|
|
|
|
The issues at this stage are only that of signal acquisition and signal pre-processing. The field-testing has shown that the expectation of the signal was slightly different from reality due to the optical properties of the conveyor belt. New methods and operations need to be developed to circumvent these issues.
|
|
|
|
Once these signal issues have been overcome, all that remains is testing the system for accuracy, stability and robustness. Future work must deal with the questions of environment-proofing and housing.
|
|
|
|
In order to study the commercial viability of this product to its end, future work must also investigate the potential sourcing and supply chains of the hardware used. As mentioned earlier in this work, the RealSense L515 has been discontinued, and other suitable hardware must be sourced and integrated.
|