“But What Happens When…” Using Dynamic Simulation to Test Solids Handling Control Strategies

San Francisco Public Utilities Commission (SFPUC) is currently in the design phase of updating its solids handling facilities at its Southeast Water Pollution Control Plant (SEP) as part of its Biosolids Digester Facilities Project. SEP is a 57 MGD treatment facility which treats about 80 percent of the City and County of San Francisco wastewater. The new solids handling facilities will include gravity belt thickeners (GBTs), sludge screens, centrifuges, a batch-style thermal hydrolysis process (THP), anaerobic digesters, and belt filter press (BFP) dewatering.The upgrade to the solids handling facilities is designed to process up to 230 dry tons per day of solids while incorporating thermal hydrolysis and anaerobic digestion to produce a Class A biosolids product. The project will replace the existing solids treatment process in its entirety, and the new solids process requires a series of solids pretreatment steps prior to thermal hydrolysis and digestion. Although SFPUC operations staff is already experienced with many of the individual solids processing steps, the sequential nature of these pretreatment steps and the continuous flow requirements of the thermally hydrolyzed solids requires robust control strategies for reliable operation. Operation staff have provided insight and recommendations for how each of the upgraded facilities should be controlled to optimize operations and reduce risk of failure. The joint design team of CH2M and Brown and Caldwell have used a dynamic simulation software program that integrates both hydraulics and process controls during the design phase so that the proposed control strategies for each facility could be modeled and evaluated. This has allowed SFPUC staff to witness how the future facilities would operate during the design phase so that changes in strategy and any additional instrumentation could be added as needed. It also provided a platform to test the process control strategies while they were being developed so that the programmed control strategy would be aligned with SFPUC expectations.The solids handling system has several tightly coupled processes and control loops that are required to feed each of the treatment process and keep solids moving in pipes to avoid solids accumulation and pipe blockage. The dynamic simulation modeling allows the design team to test different process control strategies by changing the overall process control strategy including specific control loop functionality. The resultant simulation model that was presented to SFPUC team has allowed the SFPUC team to visualize how these control loops will perform by tuning the proportional-integral-derivative (PID) control loops and witnessing the simulation for a variety of scenarios. During process control workshops, the SFPUC team was able to ask to see the results and consequences of various operational scenarios and mishaps, including turning on all belt filter presses at the same time, failing valves open or closed, and failing pumps. This allowed the operations team to understand the capabilities of the system, and allowed the design team to make informed decisions as to the levels of failure that the designed system should accommodate.The dynamic simulation model was set up to appear similar to a SCADA operations graphic so that the performance of processes could be witnessed from one central location in the model that the operations team would be comfortable with. Figure 1 provides an example of the digester feed loop pump station and digester feed control valves. At the current stopping point in the simulation, all four digesters are online and each receiving the maximum allowable flow of 113 gpm. A most-open-valve control loop is maintaining the digester feed control valves at 60 percent open to reduce the potential for unnecessary head loss in the system if the valves drifted to a more closed position. The digester feed loop pump station is being operated to achieve a constant flow of 130 gpm in the loop so that the solids are not stagnant in the pipe. Figure 1 shows that this is achieved with one pump online operating at 97 percent speed. Figure 1 is typical of the simulation screens that were set up for each major facility to evaluate performance.One specific challenge that SFPUC wanted to evaluate was the flow rate in the digester feed loop as the influent system flow rate was changed. Figure 2 provides an example where two of the three pumps that feed the digester feed loop are failed in a 15 minute period. The total digester feed flow is reduced from 360 gpm to 120 gpm with all four digesters online. The digester 1 and 2 flow rates are shown in red and green, respectively, and are shown to decrease from 90 gpm to 60 gpm to 30 gpm during the 30 minute simulation (1,800 seconds). The digester feed loop, meanwhile, is able to quickly maintain the flow setpoint in the loop of 130 gpm. During each pump failure, the loop flow spikes up close to 160 gpm before the control loop reduces it back down to the setpoint. Figure 2 provides one example of the many that were simulated to verify how the system would operate under various scenarios.Using combined dynamic hydraulic and control logic simulation during the design phase can be a powerful tool to align the team on how the system should be designed so that it meets the operational goals and expectations of the operations staff that will be expected to start up and operate the system. The dynamic simulation modeling allows the design team to test different process control strategies by changing the overall process control strategy including specific control loop functionality. System performance can be visualized in a consequence-free environment so that various failure scenarios can be simulated. Ultimately, this modeling effort increases system understanding which leads to informed decision making and reduced risk.Figure 1: Digester Feed Loop Pump Station and Control Valve Simulation ScreenFigure 2: Digester Feed Loop Performance While Digesters Feed Flow is Reduced