Keeping pH in Its Place

Industrial process water treatment and recycling have become important social and economic issues, especially with freshwater becoming a scarce resource. In a large iron and steel complex, significant amounts of wastewater are produced and need to be treated. If not done properly and efficiently, wastewater treatment can be costly and time consuming. Poor pH control remains the key problem because it causes excessive dosing of chemicals, resulting in high water treatment costs and low efficiency.

In industrial plants, most pH loops are in a “bang-bang” type of control, with pumps turning on and off and resulting in large oscillations. Because acid and caustic neutralize one another, overdosing acid and caustic is prohibitively expensive. Statistics show that a poorly controlled pH process can cost tens of thousands of dollars in chemical usage each month, not to mention the penalties imposed by violating U.S. Environmental Protection Agency or local government discharge codes.

In this article, we will discuss strategies for effectively controlling challenging pH loops and describe how model-free adaptive (MFA) control is enabling Baosteel to improve its wastewater treatment efficiency and cost savings.

The pH Process & Control Challenges

Shanghai Baosteel Group Corp. is the largest Chinese iron and steel conglomerate. It is the third-largest steel producer in the world, ranked by crude steel output, with a crude steel production capacity of about 30 million tons. Baosteel’s products are primarily carbon, stainless and specialty steel, which it forms as billets, tubes, pipes, bars and plates. The company also manufactures iron and tin products. Its markets include the appliance, auto, construction, oil and shipbuilding industries, both in China and abroad.

In Baosteel’s Shanghai chemical plant, coking wastewater, a highly toxic effluent, is treated in an ammonia still process to remove fixed ammonium (NH4+) and volatile ammonia, as shown in Figure 1. Sodium hydroxide (NaOH) is the chemical reagent used to decompose fixed ammonium. To assure efficient and sufficient decomposition reaction, closed-loop pH control in the outlet ammonia wastewater is critical.

A strong-acid-strong-base pH process is highly nonlinear. Its characteristics can be seen from the Titration curve, as shown in Figure 2, where pH is on the Y axis and the reagent flow (acid or caustic) is on the X axis. The pH value vs. the reagent flow has a logarithmic relationship. Away from neutrality, the process gain is relatively small. Near neutrality, where pH equals 7, the process gain can be a few thousand times higher.

When there is a large time delay in the pH process, it will make the control problem much worse. Not only must the user deal with the large gain changes, but also the varying time delays. Dealing with a large gain change requires precise control action, and dealing with a varying time delay requires the controller to “foresee” what is going to happen to avoid overdosing or underdosing.

The coking wastewater treatment process at Baosteel Chemical has a delay time of 15 minutes. Due to these difficulties, the pH loop had been left on manual control from 1985 to 2007.

MFA Control

MFA control is an adaptive control method that does not require process models. A set of MFA controllers is developed to address various control problems, including:

• Single-input-single-output (SISO) MFA to replace PID so that manual controller tuning is eliminated;
• Multi-input-multi-output (MIMO) MFA to control multivariable processes;
• Nonlinear MFA to control highly non- linear processes;
• MFA pH controller to control pH processes;
• Feed-forward MFA controller to deal with measurable disturbances;
• Anti-delay MFA to control processes with large time delays; and
• Robust MFA to protect the process variable from running outside a boundary.

Anti-Delay MFA

In industrial applications, many processes have large time delays due to the delay in the transformation of heat, materials and signals, etc. From a control point of view, a time delay causes the process variable to not respond to the control signal promptly; it is equivalent to disabling the feedback for a period of time. As we know, the feedback loop is essential to automatic control.

The measure of the significance of the delay time is related to the T-Tc ratio, where T is the delay time and Tc is the time constant. A PID controller usually works if the process T-Tc ratio is smaller than 1. If the T-Tc ratio is larger than 2, the process is difficult to control.

A Smith Predictor is a useful control scheme to deal with processes with large time delays. A precise process model, however, usually is required to construct a Smith Predictor. Otherwise, its performance may not be satisfactory. The anti-delay MFA controller is specially designed to control processes with significant and varying time delays.

The coking wastewater pH process at Baosteel has a T-Tc ratio of 2.5. Although it is a pH process, its main problem is the large time delay. Therefore, an anti-delay MFA controller was selected and launched to control the process. CyboCon MFA control software runs in an industrial PC interfacing to a distributed control system (DCS) through analog inputs/outputs. A heartbeat signal was configured in CyboCon and sent to the DCS to confirm that the PC and communication is in normal condition. In case the DCS detects the loss of the heartbeat signal, it will switch to manual control. Many MFA control systems are implemented in this way, and this safety feature makes the system very reliable.

The MFA pH control trends are shown in Figure 3, where SP in blue is the set point, PV in green is the measured process variable and OP in red is the controller output. The bottom trend shows the heartbeat signal that changes from 0 to 100 and back at every sample interval.

The MFA pH control system has been running since its installation in 2007. Baosteel has achieved good benefits, including smoother and safer plant operation and a 10% reduction in chemical reagent consumption, which equals to annual savings of $45,000 USD.

Dr. George Cheng is chief technical officer for CyboSoft. Cheng can be reached by e-mail at gscheng@cybosoft.com. Zhe-Gang Wang is senior engineer for Baosteel Chemical. Wang can be reached by e-mail at wangzhegang@baosteel.com.

For more information, write in 1113 on this issue’s Reader Service Card.In industrial plants, most pH loops are in a “bang-bang” type of control, with pumps turning on and off and resulting in large oscillations. Because acid and caustic neutralize one another, overdosing acid and caustic is prohibitively expensive. Statistics show that a poorly controlled pH process can cost tens of thousands of dollars in chemical usage each month, not to mention the penalties imposed by violating U.S. Environmental Protection Agency or local government discharge codes.
In this article, we will discuss strategies for effectively controlling challenging pH loops and describe how model-free adaptive (MFA) control is enabling Baosteel to improve its wastewater treatment efficiency and cost savings.

The pH Process & Control Challenges
Shanghai Baosteel Group Corp. is the largest Chinese iron and steel conglomerate. It is the third-largest steel producer in the world, ranked by crude steel output, with a crude steel production capacity of about 30 million tons. Baosteel’s products are primarily carbon, stainless and specialty steel, which it forms as billets, tubes, pipes, bars and plates. The company also manufactures iron and tin products. Its markets include the appliance, auto, construction, oil and shipbuilding industries, both in China and abroad.
In Baosteel’s Shanghai chemical plant, coking wastewater, a highly toxic effluent, is treated in an ammonia still process to remove fixed ammonium (NH4+) and volatile ammonia, as shown in Figure 1. Sodium hydroxide (NaOH) is the chemical reagent used to decompose fixed ammonium. To assure efficient and sufficient decomposition reaction, closed-loop pH control in the outlet ammonia wastewater is critical.
A strong-acid-strong-base pH process is highly nonlinear. Its characteristics can be seen from the Titration curve, as shown in Figure 2, where pH is on the Y axis and the reagent flow (acid or caustic) is on the X axis. The pH value vs. the reagent flow has a logarithmic relationship. Away from neutrality, the process gain is relatively small. Near neutrality, where pH equals 7, the process gain can be a few thousand times higher.
When there is a large time delay in the pH process, it will make the control problem much worse. Not only must the user deal with the large gain changes, but also the varying time delays. Dealing with a large gain change requires precise control action, and dealing with a varying time delay requires the controller to “foresee” what is going to happen to avoid overdosing or underdosing.
The coking wastewater treatment process at Baosteel Chemical has a delay time of 15 minutes. Due to these difficulties, the pH loop had been left on manual control from 1985 to 2007.

MFA Control
MFA control is an adaptive control method that does not require process models. A set of MFA controllers is developed to address various control problems, including:
• Single-input-single-output (SISO) MFA to replace PID so that manual controller tuning is eliminated;
• Multi-input-multi-output (MIMO) MFA to control multivariable processes;
• Nonlinear MFA to control highly non-
linear processes;
• MFA pH controller to control pH processes;
• Feed-forward MFA controller to deal with measurable disturbances;
• Anti-delay MFA to control processes with large time delays; and
• Robust MFA to protect the process variable from running outside a boundary.

Anti-Delay MFA
In industrial applications, many processes have large time delays due to the delay in the transformation of heat, materials and signals, etc. From a control point of view, a time delay causes the process variable to not respond to the control signal promptly; it is equivalent to disabling the feedback for a period of time. As we know, the feedback loop is essential to automatic control.
The measure of the significance of the delay time is related to the T-Tc ratio, where T is the delay time and Tc is the time constant. A PID controller usually works if the process T-Tc ratio is smaller than 1. If the T-Tc ratio is larger than 2, the process is difficult to control.
A Smith Predictor is a useful control scheme to deal with processes with large time delays. A precise process model, however, usually is required to construct a Smith Predictor. Otherwise, its performance may not be satisfactory. The anti-delay MFA controller is specially designed to control processes with significant and varying time delays.
The coking wastewater pH process at Baosteel has a T-Tc ratio of 2.5. Although it is a pH process, its main problem is the large time delay. Therefore, an anti-delay MFA controller was selected and launched to control the process. CyboCon MFA control software runs in an industrial PC interfacing to a distributed control system (DCS) through analog inputs/outputs. A heartbeat signal was configured in CyboCon and sent to the DCS to confirm that the PC and communication is in normal condition. In case the DCS detects the loss of the heartbeat signal, it will switch to manual control. Many MFA control systems are implemented in this way, and this safety feature makes the system very reliable.
The MFA pH control trends are shown in Figure 3, where SP in blue is the set point, PV in green is the measured process variable and OP in red is the controller output. The bottom trend shows the heartbeat signal that changes from 0 to 100 and back at every sample interval.
The MFA pH control system has been running since its installation in 2007. Baosteel has achieved good benefits, including smoother and safer plant operation and a 10% reduction in chemical reagent consumption, which equals to annual savings of $45,000 USD.

Dr. George Cheng is chief technical officer for CyboSoft. Cheng can be reached by e-mail at gscheng@cybosoft.com. Zhe-Gang Wang is senior engineer for Baosteel Chemical. Wang can be reached by e-mail at wangzhegang@baosteel.com.

Leave A Comment

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <code> <ul> <ol> <li> <dl> <dt> <dd>
  • Lines and paragraphs break automatically.

More information about formatting options

By submitting this form, you accept the Mollom privacy policy.