Process Controls

Dennis P. Nolan , in Handbook of Fire and Explosion Protection Engineering Principles, 2011

10.1 Introduction

Process controls play an important role in how a plant process upset can be controlled and subsequent emergency actions executed. Without adequate and reliable process controls, an unexpected process occurrence cannot be monitored, controlled, and eliminated. Process controls can range from simple manual actions to computer logic controllers, remote from the required action point, with supplemental instrumentation feedback systems. These systems should be designed to minimize the need to activate secondary safety devices. The process principles, margins allowed, reliability, and the means of process control are mechanisms of inherent safety that will influence the risk level at a facility.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9781437778571000100

Process Control

Pratima Bajpai , in Biermann's Handbook of Pulp and Paper (Third Edition), 2018

24.1 Introduction

Introduction

The objective of process control is to keep key process-operating parameters within narrow bounds of the reference value or setpoint. The importance of this, and some aspects of this, has been discussed in Chapter 22, Volume 2. This section will describe the theory behind control circuits to maintain automatic control over a process.

Automatic Control

The basis of automatic control (where a machine or electronic circuit is the control as opposed to a human being) is the control loop. A control loop for a given process must have at least one sensor, a controller (that decides what to do with the information that is collected), and a control element to which the results are applied.

A good example of automatic control is electric heating of a house in a cold environment, a process where a single variable is controlled. A thermostat (sensor) monitors the temperature. When the temperature drops below the setpoint a switch (controller) is closed and the heater (control element) comes on. When the temperature rises above the setpoint, the switch opens and the heater is turned off.

The difference between the actual value and the setpoint is called the error. The controller reads the error and makes a decision. The action taken can be very simple to very complicated, depending on the system and the size of the acceptable error. When heating a house the actuator turns the furnace on or off if the error is below or above the setpoint, respectively. It is a very simple control, but the error ranges to 2°C, not very precise by many standards.

This system is called a process control loop. This is a feedback loop because the desired result is determined and the information (the error) is fed back to the controller for appropriate action. In a feedforward loop, it is generally not convenient to use the desired result to control the process; therefore, another variable must be substituted.

For example, we refine pulp to influence the properties of the final sheet; however, it is not convenient to use any paper properties to control refining level because the dead time is too high. The dead time is a measure of the length of time before a change in the actuator (refining level) causes a change at the sensor (paper properties). Changes in refining level may not be seen in the paper for several minutes to several days if stock storage is used. In this case, the refining level is controlled by monitoring the energy input to the refiners for a given amount of pulp. Two sensors (stock flow rate and energy consumption) are combined into one variable (refining intensity). One must know the relationship between refining intensity and paper properties.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128142387000246

Process Control

A. Gupta , D.S. Yan , in Mineral Processing Design and Operation, 2006

18.9.4 Thickener Control

The control of thickener operation is directed to obtaining a clear overflow as rapidly as possible. The sedimentation rate is usually accelerated by additions of flocculants. Flocculants are added in the feed pipe (or feed launder) and maximum dispersion attempted by appropriate design of its entry to the thickener tank. The object is to entirely cover all the surface of mineral particles. The choice of flocculent and its concentration vary. It depends on the minerals present in the slurry, their composition and their surface characteristics. It is necessary not to have too much turbulence at the entry point of the tank‥ For this purpose, the overflow level is kept sufficiently high above the feed level to ensure acceptable solid concentration in the overflow.

Fig.18.23 is a schematic diagram of a thickener showing the different parameters. From a process control point of view the design parameters and major variables are:

Fig. 18.23. Schematic diagram of thickener.
1.

height of the overflow clear fluid, H,

2.

height of the bed level from discharge end, HB,

3.

solids inventory, S,

4.

sold mass inflow, SI,

5.

solids mass outflow, S0,

6.

average bed solid volumetric fraction, vS

It can be seen that the solid inventory S = A HB vS ρS, where ρS is the solids density.

At the bed level the solids residence time, tR will be:

(18.61) t R = S S I

The dynamic solids mass balance is:

(18.62) d S d t = ( S 1 -S 0 )

During operation, if the feed flow changes, i.e., increases or decreases, the flocculent input changes proportionately. A control loop in the flocculent charging device involving a pump is required to follow these changes at an appropriate level of control.

The power, P, required by the pump, which is assumed to be connected to a horizontal pipe with no bends, is given by:

(18.63) P = k μ ( F D ) 2 kW/h

where k = constant (proportional to pipe length),
μ = viscosity,
FD = underflow discharge flow rate.

The pump pressure obviously varies as the solids mass outflow (kg/m3) and according to Elber [10], is given by:

(18.64) P = k ρ 2 μ B s 2 S 2 0.96 Pa

Control Strategy

The control of the solid contents in the overflow and underflow streams is the basis of thickener control. The average bed density (solids inventory) has to be controlled by the underflow flow rate and the flocculent additions to the slurry. To attain target overflow solids concentration the underflow density should be sufficiently high. This is obtained by longer residence time of treated slurry in thickener.

The underflow flow rate is measured by magnetic or ultrasonic flow meters.

The control scheme can now be summarized:

Level 1: Control Loops

Two main loops are placed in Level 1. The first main loop (# 1) is for underflow control. The second main loop (#2) is for the control of flocculent flow.

At level #1, the essential process measurements for underflow control are:

1.

rake torque with a torque meter fixed to the rakes,

2.

bed level by using a simple float or vertical position sensor,

3.

thickener bed pressure, by measuring the pulp pressure on the floor by a sensor.

At level #2 for flocculent loop control, measurements are chiefly flow rates of fluids by standard flow meters and power draft measurements for variable speed positive displacement pump. Other measurements at this level include:

1.

pump speed,

2.

underflow density measurement (γ-ray density gauge),

3.

pump discharge pressure by standard pressure gauge.

Level 2: Control Loops

The aim of Level 2 is to keep the underflow bed level close to target. The set point of the bed level is therefore controlled by the bed level controller. In practice it is found that the bed level can be disturbed by high bed density which could result in high torque on the rakes. To avoid such situations the basic flow control is designed to be over-ridden. This is achieved by providing a high-selector that outputs the flow set point [10]. Such an arrangement is shown in Fig. 18.24 where it can be seen that the bed density and rake torque with maximum limiting values are connected to the high selector. The output from high selector or the flow controller set point has a set low limit. A safe flow is therefore maintained from the underflow. The pump that pumps the underflow is set to high and low speed limits taking signals from the output of the flow controller The advantage of this system is that in the event the thickener operation ceases due to say, stoppage of mill operation, and therefore feed to the thickener, the underflow pump continues to operate till the thickener is empty and chances of clogging is remote.

Fig. 18.24. Underflow control setup [10].

The pump speed controller incorporates limiting power draft so that the pump does not trip at high power

For controlling the flocculent flow signals are taken from the bed density controller. Controlling the flocculent flow is difficult by this method as it takes time for the flocculent to properly mix with the rest of the inventory. The speed of response varies with the rate of change of bed density. Elber [10] suggests using underflow flow to "control inventory in conjunction with bed level-flocculent dosage cascade."

Level 3: Control Loops

Level 3 control involves optimisation of thickener operation (and pipe lines). This includes cost function based on:

1.

flocculent consumption,

2.

pump power,

3.

discharges to tailings.

All these factors depend on the underflow density set point. Optimum conditions are usually ascertained by trial and error method by taking signals from the underflow density, pump discharge pressures and pump power drafts and estimating the corresponding cost functions.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B978044451636750019X

Process Control

James R. Couper , ... Stanley M. Walas , in Chemical Process Equipment (Third Edition), 2012

On typical grass roots chemical processing facilities, as much as 10% of the total capital investment is allocated to process control equipment, design, implementation and commissioning. Process control is a very broad topic with many distinct aspects. The following list of possible sub-topics gives some idea of the full breadth of this topic:

In the field, the topic includes the selection and installation of sensors, transmitters, transducers, actuators, valve positioners, valves, variable-speed drives, switches and relays, as well as their air supply, wiring, power, grounding, calibration, signal conditioning, bus architecture, communications protocol, area classification, intrinsic safety, wired interlocks, maintenance, troubleshooting and asset management.

In the control room, the topic encompasses the selection and installation of panel mounted alarms, switches, recorders and controllers, as well as Program Logic Controllers (PLC) and Distributed Control Systems (DCS), including analog and digital input/output hardware, software to implement control strategies, interlocks, sequencing and batch recipes, as well as display interfaces, alarm management, and Ethernet communication to networked computers, which are used to provide supervisory control, inferential measures, data historians, performance monitoring, and process optimization.

Also, the design practice includes P&ID documentation, database specification and verification of purchased equipment, control design and performance analysis, software configuration, real-time simulation for DCS system checkout and operator training, reliability studies, interlock classification and risk assessment of safety instrumented systems (SIS), and hazard and operability (HAZOP) studies.

Books have been written about each of these sub-topics and many standards exist to specify best practices or provide guidance. The Instrumentation, Systems and Automation Society (ISA) is the primary professional society that addresses many of these different aspects of process control. The focus of this chapter will be on control loop principles, loop tuning and basic control strategies for continuous processes.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780123969590000033

Pearl GTL

Peter J. de Kok , Rob R.M. Overtoom , in Proceedings of the 3rd Gas Processing Symposium, 2012

3.2 Advanced process control (APC)

Advanced Process Control (APC) is the term used within Shell for "multivariable model-based predictive control". It is used on top of the regulatory control to enhance the stability and operability of the plant and to locally optimize parts of the plant. Typically advanced control is applied on a single reactor or a single distillation column, although there is a tendency nowadays to enlarge the scope of model-based controllers to control ever larger parts of a plant with a single APC controller.

Although APC is not considered necessary to start-up and operate the plant, studies always reveal that the use of advanced control has a substantial economic incentive due to improved plant stability and hence less shutdowns, maximization of production and reduction of quality give-away. Using APC as early as possible, once the plant has been started, is therefore seen as an important driver to Operational Excellence.

APC resides in a separate computer/server, linked to the DCS system. The platform used in Pearl GTL for APC is the Shell proprietary package SMOCPro (Shell Multivariable Optimising Control) in combination with RQEPro (Robust Quality Estimator). In total approximately 70 SMOC controllers and 100 Quality Estimators are being developed for Pearl GTL.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780444594969500242

Process Monitoring and Process Control

Mark Berry , Nick Schott , in Applied Plastics Engineering Handbook, 2011

20.2.4 Early Twenty-first Century: More Powerful Computers; More Technical Approach to Processing

Process monitoring and control technology have continued to develop, with better utilization of computer hardware and software to provide improved ease of use and reliability. The 16-bit and 32-bit systems provide more speed, more data storage, and more complex control algorithms. Systems are becoming easier to implement and use with the introduction of more advances such as advanced user interfaces, and wireless transducers and transmitters. Processors have generally accepted that to be able to manage and optimize any process one must understand the process at the most fundamental levels. In injection molding, closely monitoring what is actually happening in the mold cavity itself is gaining wider acceptance, with continued improvements in pressure sensor technology as well as in melt temperature measurement. While cavity pressure is the most commonly used in-mold sensor, Priamus® has emphasized the utilization of melt temperature in the cavity as an important control characteristic, using a temperature to capture the arrival of the melt front to a specific location, to determine when to instruct the molding machine to execute the switchover from ram velocity control to pressure-based control [8] . There is also work being done which combines the output of both cavity pressure and melt temperature to characterize an acceptable process window. All of the above approaches are based on the notion that the primary machine process parameters as well as material flow properties and environmental conditions all vary over time and are interrelated in ways that are far too complex to provide a basis for a simple process adjustment. It is generally agreed that the cavity filling experience of the melt is the best determinant of the quality of the product. While more processors believe that the best process control model for injection molding should include measurements made in the mold during the cavity filling step, there is still considerable debate as to how this is best accomplished.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9781437735147100200

Supporting technologies for the processing of metals and alloys

In Smithells Metals Reference Book (Eighth Edition), 2004

40.4.3 Process control techniques 32, 35

Process control plays a critical role in metals production and treatment. Metallurgical processes, both in batch and in continuous operations, normally exhibit time delays with large time constants, uncertainties, non-linearities, unmodelled dynamics, etc. Taking this into account, different control techniques have been developed since 1960's. In the early days, the process control was based on mechanical, electrical or pneumatic analogue controllers. Later on, the broad use of digital computers came to the process control thanks to the rapid development of control theory, from single-input single-output linear system to a wide range of multivariable non-linear systems. Nowadays, the extensive application of expert systems, neural networks, fuzzy control, process monitoring and on-line diagnostics has become important parts of a modern process control system in metallurgical processes.

Figure 40.8 illustrates a general control structure for a typical metallurgical process, 35 which indicates different levels of a modern control system and the importance of database central to the whole system. For the benefit of the discussion below 5 levels are defined, however, various levels have been defined in the past by various vendors and providers. The figure explains the significance of each level. From Figure 40.8, discussed with reference to a submerged arc furnace for ferroalloy production, various levels can be discerned, each containing a mixed bag of the methodologies discussed above. 32 This section will dwell on the process control structure and will specifically attempt to explain the structure and function of the database, which is central to the industrially implemented system.

Figure 40.8. Control structure for a metallurgical furnace 35

Levels 1 & 2: Mintek's patented low-level control system (Minstral™) for submerged-arc furnaces 31 has an international proven track record, having been implemented in various plants all over the world. This control system, packaged in different forms, i.e. from a stand-alone unit to a Windows based network version, has various proven features (a) control based on a proven resistance based algorithm, which de-couples the movement of electrodes, (b) optimisation of power input into the furnace as a function of MVA, resistance, MW and current set-points as well as a function of various dead bands and feedback loop settings, (c) differential tapping of transformers, (d) balancing of the electric circuit and hence minimisation of the asymmetry of the electric circuit, (e) the TCP/IP Windows based network version of the Minstral™ which permits plant wide maximum demand control, and (f) control of electrode baking. Level 3: The SCADA system creates the man-machine-interface (MMI) between the control system, the user and subsequently between all higher-level control and modelling applications. The networked SCADA system (Table 40.8) provides all the standard features of such state-of-the-art packages e.g. graphical interfaces, data logging, historical trending, and SQL-features, intelligent multiple symptom alarming e.g. through the command language, statistical process control (SPC), on-line documentation, checking of trends and setting of corresponding alarms, checking instrument readings for failure and setting alarms, and monitoring furnace pressures, temperatures, etc. and setting of alarms. In addition to the basic SCADA software customised for an application, various modular Windows utilities could also be accessed from the user-SCADA's user-interfaces. These model-based utilities provide the user with simulation tools with which various furnace modes of operation can be simulated hence providing very useful operator guidance. Level 4: A good database (for Mintek's Furnstar™ system 31 —MS Access® front end to SQL Server®) residing on the server of a network is the nucleus of a supervisory control system. It links databases containing data originating from different sources such as real-time data from a SCADA, production, planning and human resources data from e.g. SAP®, etc. It accommodates all furnaces in a plant and collates data in a metallurgically meaningful way in a metallurgically useful database. This structure ensures that all data can be considered simultaneously during decision making, be it by management, metallurgist, operator, fundamental process models, the control system or even an expert system. Since the furnace, its feeds and products, are only poorly defined, rather accurate process data sets are required in order to be able to calibrate (semi-)empirical models, create statistical models, or provide inputs to expert systems, etc. It can, therefore, be stated that: the more poorly a system is defined, the more heavily the (feed-forward) control system should make use of a well-structured and metallurgically useful database. Level 5: As is clear from Figure 40.8 various types of on-line models can be developed for this level, which could include fundamental models, expert systems, neural nets, process scheduling, ecological simulation and plant wide optimisation models. Level 6: Although not indicated, man (still) plays a critical part in supervisory control systems in extractive metallurgy. For submerged-arc furnaces man's involvement is still large. What this in fact also implies is that, if everyone on a plant does not buy into the control philosophy of a plant, it can be stated that optimal results will only be achieved with difficulty. Man is often still really the learning machine on an extractive metallurgical plant!

Table 40.8. COMMONLY USED LOWER LEVEL (1 TO 3) CONTROL SOFTWARE FOR METALS PRODUCTION AND PROCESSING

Name of software Functionality
PI system (OSI software) (www.osisoft.com) PI is a universal production data acquisition system with an integrated analysis system. The data from various systems of your production plant (PCS, PLC, laboratory data, manual inputs) is stored in a long-term archive on a data server and can be visualised and evaluated on a PC within the Windows environment. In other words, a set of server and client based software programs, designed to fully automate the collection, storage and presentation of plant information, also permitting real-time data reconciliation.
G2 (Gensym) (www.gensym.com) A graphical, object-oriented, customisable software platform for rapidly building expert manufacturing applications. G2 applications model and simulate operations, intelligently monitor and control processes and diagnose faults.
Emerson process management (www.emersonprocess.com) Process management software.
FIX DMACS™ iFIX (www.intellution.com) SCADA (Supervisory Control and Data Acquisition) software, iFIX, the HMI/SCADA component of the Intellution Dynamics family of automationsoftware, is a Windows NT-based industrial automation solution for monitoring and controlling manufacturing operations.
Fisher-Rosemount systems (www.emersonprocess.com) Part of Emerson Process Management. Provides software for process management; Process control, automation, and optimisation; Asset management, monitoring, maintenance, and optimisation.
Honeywell (www.honeywell.com) Provides advanced software applications for home/building control and industrial optimisation; sensors, switches, controls systems and instruments for measuring pressure, air flow, temperature, electrical current and more.
InTouch™ and FactorySuite (www.wonderware.com) InTouch, a HMI, provides a single integrated view of all the control and information resources. InTouch enables engineers, supervisors, managers and operators to view and interact with the workings of an entire operation through graphical representations of their production processes. FactorySuite is an integrated suite of industrial automation software.
iBaan for metals (www.baan.com) Business software for the metals industry e.g. management of inventory to optimise supply and value chain.
ABB IndustrialIT Solutions for Metals (www.abb.com) ABB Industrial IT integrates diverse automation and information technologies in real time to provide better business decision support, standardisation of global processes.
Minstral™ (www.mintek.ac.za) Mintek's Submerged Arc Furnace Controller, an integrated electrode and tap-changer control system.
CSense™ Crusader Systems (www.crusader.co.za) CSense™, interfaces with SCADAs, DCSs and historians, for diagnosis of process deviations: Real-time analyzes of deviations, Alarming and messaging in HMI environment, Automatic reporting. Real-time soft sensor with a variety of artificial intelligence tools such as neural nets, fuzzy as well as classical statistics.
MDC Technology (www.mdctech.com) MDC Technology is a major specialist provider of Production Optimisation and Performance Monitoring software solutions to the oil & gas, hydrocarbon processing, chemicals and power/utility industries. Part of Emerson Process Management.

In Table 40.7 standard control techniques are summarised and listed for various operations of metals processing. For extensive applications in different minerals and metals processing industries, please read the paper by Jämsä-Jounela. 5

Table 40.7. STANDARD CONTROL TECHNIQUES USED FOR METALS PRODUCTION AND PROCESSING 5

Control technique Application examples
Classical methods
P, PI, PID control Submerged-arc furnace (e.g. FeMn, FeCr, SiMn, SiCr, FeSi, Si), rotary kiln, hot and cold rolling mills
Feed forward control Submerged-arc furnace
Model-based control and model predictive control Thermal treatment, reheat furnace, extrusion, ferrosilicon production, hot and cold rolling
Multivariable control Hot rolling mill, Al production cell,
Adaptive control Al production cell, BOF, ingot/slab/continuous casting, reheat furnace, cold rolling
Artificial intelligence (AI) control
Expert system Blast furnace, BOF, hot/cold rolling
Fuzzy logic Sintering machine, BOF, Zn roaster, continuous casting
Neural networks Reheat furnace, EAF, BOF, ladle furnace, continuous casting, hot rolling
Genetic algorithms
Diagnostics and monitoring
Fault diagnostics and process monitoring Zn hydrometallurgy plant, Outokumpu flash smelting, blast furnace, ferroalloy furnace, steelmaking, continuous casting, hot rolling
Quality monitoring Steel surface inspection, continuous casting, hot rolling
Modelling techniques
Thermodynamic (ChemSage, Metsim) Equilibrium calculations for metallurgical reactors
Dynamic modeling Various ARMAX type models to quantify the dynamics of metallurgical reactors
Statistical:

Data reconciliation

Neural networks

Principal component analysis

 Blast furnaces, converters, arc-furnaces, Sn/Cu/Zn smelters
CFD Casting, modelling of flow in various metallurgical furnaces, off-gas systems, rotary kilns, waste incinerator

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780750675093500439

Modelling and Control of Temper Rolling and Skin Pass Rolling

Olof Wiklund , Fredrik Sandberg , in Metal Forming Science and Practice, 2002

15.6 DEVELOPMENT TRENDS

15.6.1 Modelling and control

Modelling and process control will be more focused on temper rolling and skin pass rolling. This is partially due to a "saturation" of the efforts in the upstream processes and partially due to increased customer demands on the mechanical and surface properties as well as the flatness of the finished product.

15.6.2 Temper rolling and tension levelling

Temper rolling is not likely to be replaced by tension levelling, since it is better for the material properties and the surface, but in some cases a tension leveller is added in-line after the temper rolling mill in order to improve the flatness.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780080440248500151

Process control instrumentation

In Fluid Catalytic Cracking Handbook (Fourth Edition), 2020

3.2 Process control instrumentations

Process control instrumentation controls the FCC unit in a safe, monitored mode with limited operator intervention. Two levels of process control are used:

•

Basic supervisory control.

•

Advanced process control (APC).

3.2.1 Basic supervisory control

The primary controls in the reactor-regenerator section are flow, temperature, pressure, and catalyst level.

The flow controllers are often used to set desired flows for the fresh feed, recycle, air rate, stripping steam, dispersion steam and etc. Each flow controller usually has three modes of control: manual, auto, and cascade. See Fig. 3.1 for a typical process flow diagram (PFD). In manual mode, the operator manually opens or closes a valve to the desired percent opening. In auto mode, the operator enters the desired flow rate as a set-point. In cascade mode, the controller set-point is an input from another controller.

Fig. 3.1. Typical FCCU process flow diagram (PDF). Note: FV, flow control valve; FT, flow transmitter; KO, knock out; LI, level indicator; LV, level control valve; MF, main fractionator; OVHD, overhead; PDT, pressure differential transmitter; PT, pressure transmitter; TV, temperature control valve.

The reactor temperature is controlled by a temperature controller that regulates the regenerated catalyst slide valve. The regenerator temperature is not automatically controlled but depends on its mode of catalyst regeneration. In partial combustion, the regenerator temperature is controlled by adjusting the flow of air to the regenerator. In full burn, the regenerator temperature is a function of several variables, including feedstock quality, catalyst properties, use of recycle, stripping steam rate and mechanical conditions of the feed injection system and the catalyst stripper.

The reactor pressure is not directly controlled; instead, it floats on the main column overhead receiver. A pressure controller on the overhead receiver controls the wet gas compressor and indirectly controls the reactor pressure. The regenerator pressure is often controlled directly by regulating the flue gas slide or butterfly valve. (In some cases, the flue gas slide or butterfly valve is used to control the differential pressure between the regenerator and reactor.)

The reactor or stripper catalyst level is maintained with a level controller that regulates the movement of the spent catalyst slide valve. The regenerator level is manually controlled to maintain catalyst inventory.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780128126639000035

Unit Monitoring and Control

Reza Sadeghbeigi , in Fluid Catalytic Cracking Handbook (Second edition), 2000

PROCESS CONTROL INSTRUMENTATION

Process control instrumentation controls the FCC unit in a safe, monitored mode with limited operator intervention. Two levels of process control are used:

•

Basic supervisory control

•

Advanced process control (APC)

Basic Supervisory Control

The primary controls in the reactor-regenerator section are flow, temperature, pressure, and catalyst level.

The flow controllers are often used to set desired flows for the fresh feed, stripping steam, and dispersion steam. Each flow controller usually has three modes of control: manual, auto, and cascade. In manual mode, the operator manually opens or closes a valve to the desired percent opening. In auto mode, the operator enters the desired flow rate as a set-point. In cascade mode, the controller set-point is an input from another controller.

The reactor temperature is controlled by a temperature controller that regulates the regenerated catalyst slide valve. The regenerator temperature is not automatically controlled but depends on its mode of operation. In partial combustion, the regenerator temperature is controlled by adjusting the flow of combustion air to the regenerator. In full burn, the regenerator temperature is a function of operating conditions such as reactor temperature and slurry recycle.

The reactor pressure is not directly controlled; instead, it floats on the main column overhead receiver. A pressure controller on the overhead receiver controls the wet gas compressor and indirectly controls the reactor pressure. The regenerator pressure is often controlled directly by regulating the flue gas slide or butterfly valve. In some cases, the flue gas slide or butterfly valve is used to control the differential pressure between the regenerator and reactor.

The reactor or stripper catalyst level controller is controlled with a level controller that regulates the movement of the spent catalyst slide valve. The regenerator level is manually controlled to maintain catalyst inventory.

Regenerated and Spent Catalyst Slide Valve Low Differential Pressure Override

Normally, the reactor temperature and the stripper level controllers regulate the movement of the regenerated and spent catalyst slide valves. The algorithm of these controllers can drive the valves either fully open or fully closed if the controller set-point is unobtainable. It is extremely important that a positive and stable pressure differential be maintained across both the regenerated and spent catalyst slide valves. For safety, a low differential pressure controller overrides the temperature/level controllers should these valves open too much. The shutdown is usually set at 2 psi (14 Kp).

The direction of catalyst flow must always be from the regenerator to the reactor and from the reactor back to the regenerator. A negative differential pressure across the regenerated catalyst slide valve can allow hydrocarbons to back-flow into the regenerator. This is called a flow reversal and can result in an uncontrolled afterburn and possible equipment damage. A negative pressure differential across the spent catalyst slide valve can allow air to back-flow from the regenerator into the reactor with equally disastrous consequences.

To protect the reactor and the regenerator against a flow reversal, pressure differential controllers are used to monitor and control the differential pressures across the slide valves. If the differential pressure falls below a minimum set-point, the pressure differential controller (PDIC) overrides the process controller and closes the valve. Only after the PDIC is satisfied will the control of the slide valve return to the process.

Advanced Process Control

To maximize the unit's profit, one must operate the unit simultaneously against as many constraints as possible. Examples of these constraints are limits on the air blower, the wet gas compressor, reactor/regenerator temperatures, slide valve differentials, etc. The conventional regulatory controllers work only one loop at a time and they do not talk to one another. A skilled operator can "push" the unit against more than one constraint at a time, but the constraints change often. To operate closer to multiple constraints, a number of refiners have installed an advanced process control (APC) package either within their DCS or in a host computer.

The primary advantages of an APC are:

•

It provides more precise control of the operating variables against the unit's constraints and, therefore, obtains incremental throughput or cracking severity.

•

It is able to respond quickly to ambient disturbances, such as cold fronts or rainstorms. It can run a day/night operation, taking advantage of the cooler temperatures at night.

•

It pushes against two or more constraints rather than one single constraint. It can maximize the air blower and wet gas compressor capacities.

As mentioned above, there are two options for installing an APC. One option is to install an APC within the DCS framework, and the other is to install a multivariable modeling/control package in a host computer. Each has advantages and disadvantages, as indicated below.

Advantages of Multivariable Modeling and Control

The multivariable modeling/control package is able to hold more tightly against constraints and recover more quickly from disturbances. This results in an incremental capacity used to justify multivariable control. An extensive test run is necessary to measure the response of unit variables.

In APC on DCS framework, the control structure must be designed, configured, and programmed for each specific unit. Modifying the logic can be an agonizing process. Wiring may be necessary. It is difficult to both document the programming and to test.

With a host computer framework, the control package is all in the software. Changing the program can still be agonizing, but the program can be tested off-line. There is more flexibility in the computer system, which can be used for many other purposes, including on-line heat and weight balances.

Disadvantages of Multivariable Modeling and Control

A multivariable model is like a "black box." The constraints go in and the signals come out. Operators do not trust a system that takes the unit away from them. Successful installations require good training and continual communication. The operators must know the interconnections in the system.

The model may need expensive work if changes are made during a turnaround. If the feed gets outside the range the unit was modeled for, results can be at best unpredictable. An upset can happen for which the system was not programmed.

The DCS-based APC is installed in a modular form, meaning operators can understand what the controlled variable is tied to more easily.

The host computer-based system may have its own problems, including computer-to-computer data links.

In any APC, the operators must be educated and brought into it before they can use it. The control must be properly designed, meaning the model must be configured and properly "tuned." The operators should be involved early and all of them should be consulted since all four shifts may be running the unit differently.

Read full chapter

URL:

https://www.sciencedirect.com/science/article/pii/B9780884152897500068