My original concept of wireless has changed a lot. I always realized that wireless could potentially eliminate most of the installation design and construction costs particularly for inline, pipeline and vessel mounted devices where you don’t have the cost of sensing or sample lines. What is new to me is the additional savings in maintenance costs from the automatic transmission of alerts without interrogation, the smarter diagnostics in the new wireless transmitters, and the elimination of the inevitable questions about the integrity of wiring and terminations whenever there is a problem. Additionally, I became aware how the built-in security, minimization of interference, and automated self-organization and optimization of the network for reliability and performance makes the installation and maintenance cost and time comparison between wired and wireless more dramatic.
I thought back to a visit to the main lab of a leading biopharmaceutical company where all of the equipment, control systems, and analyzers were portable. The lab manager firmly believed the ability to mix and match was exceptionally important in the dynamic research environment. I then asked myself, what if the instruments were wireless?
There is a beta test in progress for wireless control of single use bioreactors (SUB) at Broadley-James Corporation. The poster Wireless-Devices-in-Single-Use-Bioreactors.pdf describes the beta test and the potential benefits of wireless for SUBs. Even more important to me since I am a control type is the test of the enhanced PID for wireless control called PIDPLUS. The poster and the PIDPLUS_Results.pdf presentation describe how exception reporting and PIDPLUS can reduce communications enough to make battery life a non-issue. The set point response is the same as a wired traditional PID. The load response is not quite as good for the resolution setting of 1% but if the resolution is decreased, the load rejection improves. The optimum resolution setting should be larger than the maximum amplitude of measurement noise and larger than the resolution of the control valve multiplied by the process gain but less than 1/5 of the allowable deviation from set point. For many flow loops, the resolution setting could be about 0.5%. For tight temperature and pH control and narrow spans, the setting might need to be 0.1%. The limit cycles caused by resolution limits in a measurement from exception reporting or in a control valve from stick-slip may not be as significant as one might expect. The cycles are often smoothed by process volumes (via process lags) to the degree where the amplitude is inconsequential and difficult to distinguish from noise.
Finally, what surprised me the most was how the PIDPLUS improved the stability for any control system whenever there was a significant measurement time delay. This potential improvement was mentioned at the end of Advanced Application Note 5 and briefly described in the Control magazine article “Unlocking the Secret Profiles of Batch Reactors” . The presentation “IFPAC2009_Techniques_Wireless_pH_Control.pdf goes into more detail. This presentation shows how a wireless PIDPLUS does about as well as wired traditional PID for bioreactor pH control for a resolution of 0.01 pH (0.1%). More impressive in the presentation is the improvement shown by the PIDPLUS compared to the traditional PID for glucose control and for generic self-regulating (e.g. continuous) and integrating (e.g. batch) processes where significant analyzer measurement delay has been introduced. This technological advancement in the PID for all loops is a fall out of maximizing the utility of the new wireless technology.