Fluid handling – Liquid level responsive or maintaining systems – Control of both inflow and outflow of tank
Reexamination Certificate
2000-09-05
2002-07-23
Drodge, Joseph W. (Department: 1723)
Fluid handling
Liquid level responsive or maintaining systems
Control of both inflow and outflow of tank
C137S395000, C137S558000, C137S571000, C700S281000, C700S282000, C340S618000
Reexamination Certificate
active
06422263
ABSTRACT:
CROSS-REFERENCE TO RELATED APPLICATIONS
Not Applicable
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
Not Applicable
REFERENCE TO A MICROFICHE APPENDIX
Not Applicable
BACKROUND-FIELD OF INVENTION
The technical field on this invention relates to fluid handling in a plurality of reservoirs connected in series. This invention involves innovative changes to the uses and configurations of the traditional proportional only (P), and proportional/integral/derivative control algorithms (PID) as they have been applied to maintain system demand and minimize disturbances to the critical levels within a process. Although this invention has been birthed out of needs specifically attributable to the water treatment industry as recognized by the applicant pursuant to over 20 years of process control experience in the field, the invention can be applied to all other liquid handling processes where demand fluctuates and reservoir or vessel level is desired to be maintained at a specific level.
BACKROUND-PRIOR ART
Please refer to
FIG. 1
Prior Art for a general simplified layout of a typical water treatment facility. This drawing shows pumped raw water influent to sedimentation basins, then flowing by gravity to a common flume which hydraulically links multiple sand filters. Each filter effluent then flows by gravity to an effluent clearwell from which water is pumped to maintain system demand.
The water treatment industry has long recognized the importance of minimizing filter flow rate fluctuations to minimize the breakthrough of turbidity (i.e. trapped sediment within the filter media bed) to the effluent clearwell to insure a clean water supply. Thus, different control schemes have been utilized by the industry to minimize water level (i.e. flume level) fluctuations in gravity flow sand filters. This is due to the fact that head pressure across the media bed varies with fluctuating level and since flow is proportional to the square root of the differential pressure across the filter, flow rate also fluctuates. Referring to
FIG. 2
Prior Art depicting the relationship of flow rate to differential pressure, for a given effluent valve position the effect of pressure differential (as related to filter water level) is even more significant at lower flow rates.
To further aggravate the situation, as trapped sediment builds up within the media bed during the filter run period (filter run period defined as the period of time filtering occurs between filter backwashes) the hydraulic head pressure decreases, thus requiring the filter effluent valve to open further to maintain a set flow rate. Since filters are backwashed one at a time on a staggered basis to minimize disruption to the water plant's water production capability, the head loss for a given filter will not be the same as any other and can vary by as much as four to five feet or more.
The scenario in which the quality of filtered water could be maximized would be where plant effluent demand would remain constant, filter level would remain constant, and the positions of the filter effluent valves would gradually increase over the course of the filter run to maintain a constant filter flow with increasing head loss across the filter media bed during the filter run. This scenario however is totally unrealistic since plant effluent demand does change and filters must be backwashed. Thus, the challenge is one where filtered water quality must be optimized with plant demand and other system disturbances, which requires filter flow rates to change as little as and as smoothly as possible. Based on fluctuating flume level with varying degrees of head loss as these variables relate to flow as described previously, if filter flow rates can be made to follow system demand while maintaining constant flume level, not only is filtered water quality optimized with plant demand, but the water production calculation data required by the regulatory agencies such as filter load rates are much more consistent, realistic and accurate.
The problem, however, becomes more complicated because at this point, only the first phase of water production and associated levels and flows has been discussed. Ultimately, filtered water must pass to the effluent clearwell, which is used as a storage medium for plant effluent pumping. As such, the water treatment industry has long recognized the need to minimize clearwell water level fluctuations in order to be responsive to system demand and various control schemes have been utilized to accomplish this. Further, the trend of state governments responding to the increased restrictions mandated by the Federal EPA for water quality has been to consider the effluent clearwell as the chemical contact chamber for chemical post treatment. This consideration requires the effluent clearwell to be maintained within certain set levels, below which fines may be imposed. This is so because it is recognized that contact time of the water to be treated with the chemicals is essential to the bonding process and the ultimate effectiveness of the chemical treatment, and that contact time increases with higher and reasonably steady state set levels. Similarly as for flume level, the water production calculation data required by the regulatory agencies such as chemical contact time is much more consistent, realistic, accurate, and accepted with a reasonably steady state clearwell level.
Further to the issue of responding to the increased requirements of the regulatory agencies, the credibility of the computer-based historical data collection and reporting systems concerning the provision of meaningfully accurate and consistent report data often times is a problem for municipalities. Since all computer-based data collection systems monitor process variable signal data by sampling techniques, accuracy decreases as these variables fluctuate, and this is again another reason to optimize the responsiveness of the water plant to system demand while minimizing the fluctuations in process level and flow variables. Inherent inaccuracies with these data collection systems are further aggravated when there are calibration errors and discrepancies amongst the various flow measuring devices that go undetected for indeterminate amounts of time.
Various control schemes have been implemented over the years in an attempt to control both flume level and clearwell level for the reasons mentioned previously. These schemes have attempted to address the challenge of maintaining these levels in a situation where the hydraulic design of the water plant requires the control aspects of these levels to be in competition with one another. These control schemes are herein described with their flaws.
Refer to the control scheme for flume level control depicted on
FIG. 3
Prior Art. This configuration for flume level control utilizes dedicated proportional/integral/derivative (PID) flow controllers for each filter. A single proportional only (P) direct acting level controller output is cascaded to the set point inputs of the various PID reverse acting filter effluent flow controllers.
Referring to
FIG. 4
Prior Art, shows the output responses of the direct acting proportional only flume level controller shown in
FIG. 3
based on example gain values of 1 (waveform “D”) and 2 (waveform “C”), a bias constant of approximately 64% (waveform “E”), and a flume level set point of approximately 84% (waveform “B”). The typical proportional only filter influent flume level controller is set up to measure over a 0 to 7 foot range where the desired control set level is approximately
6
feet and the control action is such that the output will swing from 0 to 100% when the level swings from minus 6 inches to plus 6 inches (control band) around the 6-foot level set point. This correlates to a gain of approximately 7, or a proportional band of approximately 14%. The bias constant input is selected to shift the control band to the 6-foot “null” point of the controller. With this configuration, although very stable, flume level is maintained at set point only at the “
Cecil Terry K.
Drodge Joseph W.
Spicer Guy K.
LandOfFree
Nested and cascaded variable bias feedfoward and feedback... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Nested and cascaded variable bias feedfoward and feedback..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Nested and cascaded variable bias feedfoward and feedback... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2818700