557 e2 fm Guide to Advanced Control Systems API RECOMMENDED PRACTICE 557 SECOND EDITION, OCTOBER 2013 Special Notes API publications necessarily address problems of a general nature With respect to pa[.]
Introduction
This Recommended Practice (RP) addresses the implementation and ownership of advanced control systems for petroleum processing facilities The major sections of this RP are described in 1.2.2 through 1.2.7
Figure 1 highlights the key functions essential for the efficient and cost-effective operation of a refinery, demonstrating the integral role of advanced control systems within this framework These advanced control systems serve as a crucial foundation for various other operational functions, and similar functional diagrams can be applied to other continuous processing facilities.
Scope
This RP outlines best practices for identifying opportunities, justifying, managing, implementing, and maintaining advanced control systems It does not favor any specific techniques or applications but serves as a foundation for defining the necessary work processes and common functions essential for the effective implementation and upkeep of these systems.
This document outlines practices relevant to all advanced control systems applications Experienced users may have their own effective methods, and this guidance is not meant to replace established practices that are deemed acceptable or to mandate adherence to these recommendations if they are unsuitable for specific situations.
Selection of a specific hardware platform, software platform, or application software is not within the scope of this RP
Product, intermediate, and raw material planning and operating objectives Multiplant integration
Yield accounting and process analysis
Blend property control and storage management
Blending and oil movements Regulatory control system
Field measurements, control elements, and communications
Operation planning and optimization, online optimization
Maintenance, engineering, and administration Corporate management/enterprise resource planning systems
1.2.2 Control Systems Functions and Types
The functions and characteristics of commonly used advanced control systems applications are described in Section 3.
General procedures for identification of advanced control systems applications that may provide economic or operational benefit to a facility are described in Section 4.
General concepts for planning and management of an advanced control project are described in Section 5.
The technical issues that should be considered in selecting advanced control system hardware and software are described in Section 6
Application design features needed to support control functions, operator interfaces, and engineer interfaces are described in Section 7.
Ongoing maintenance recommended practices for advanced control systems are described in Section 8.
For the purposes of this document, the following definitions apply Also refer to API 554, (all parts) Process Control
Systems for definitions of related terms.
Personnel
An advanced control engineering specialist is a trained and experienced professional skilled in designing and implementing sophisticated control systems They possess expertise in process engineering, process control theory, and relevant computer applications These specialists can work for refining companies, control systems manufacturers, or as independent consultants and contractors.
An advanced control application requires dedicated oversight and maintenance, typically performed by a knowledgeable professional such as a unit process engineer or a plant control system engineer These individuals play a crucial role in ensuring the application's optimal performance and functionality.
The ultimate user of an advanced control system application can be either a process operator or an engineer responsible for its operation.
A person or persons that is responsible for day-to-day operation of a process unit and its advanced control applications.
An advanced control project manager oversees the execution of projects, handling various responsibilities based on the project's specific nature and scope Key tasks include managing project resources, budget, and schedule In smaller projects, this specialist may also take on the role of project engineer.
A process engineer is responsible for managing engineering tasks linked to the daily operations of a process unit or area, ensuring efficiency and effectiveness Additionally, this role may encompass responsibilities as an advanced control support specialist, enhancing operational performance through specialized technical support.
Controller Types
An advanced control system goes beyond traditional regulatory control by managing multiple variables to achieve specific operational goals It incorporates complex calculations that surpass standard algorithms typically found in process control systems Additionally, it often integrates a multitude of standard algorithms in a sophisticated manner, enhancing its functionality and effectiveness in controlling processes.
An advanced control system can be deployed on a high-level computing resource, like a process control computer, or within a programming environment at lower control levels, regardless of the complexity involved in the computations.
These types of applications are referred to by terms such as advanced process control, model-based predictive control, matrix control, and multivariable control.
The term "controller" in this document refers specifically to an advanced control system, encompassing a range of functions related to regulatory and advanced control systems.
Multivariable control systems are advanced applications that maintain multiple control variables (CVs) at desired levels through complex interrelationships These systems adjust several manipulated variables (MVs) simultaneously to achieve economic or operational objectives Typically, multivariable controllers operate at frequencies ranging from one to five minutes, although they can also function at faster rates.
An optimization function in process control identifies the ideal operating conditions to enhance economic benefits while adhering to specific constraints This scheme can focus on various objectives, including maximizing product output, minimizing operational costs, or extending the lifespan of equipment Optimization programs generally update operational targets every few hours to days, reflecting the frequency of changes in economic factors at each site.
A control application maintains a specific variable at a desired value by adjusting a single manipulated variable (MV) This regulatory control can also involve applications that use common calculations or predictions, such as steam drum level controls, combustion controls, and mass flow calculations.
Controller Terminology
The combination of the hardware platform, software platform, and application software necessary to implement an advanced control system application.
An advanced control system can fully or partially deactivate, transferring control back to the regulatory control scheme This situation may arise due to invalid input values, the controller's inability to deliver outputs, or failure to achieve its objectives.
In any process or equipment, it is crucial to adhere to established limits, often known as "limit variables." These constraints can manifest as physical parameters, including design temperature and pressure, or predefined process limits such as maximum feed rates and composition values They can represent either maximum or minimum thresholds, affecting factors like flow pressure, temperature, and the quality of process streams It is essential that these constraints align with the HAZOP (Hazard and Operability Study) and the alert/alarm strategy for the controlled area, ensuring safety and operational efficiency.
Process values that are maintained by the control system by making appropriate adjustments to MVs.
An advanced control system application processes input values that are measured but not directly controlled, such as ambient temperature and feed from other units These applications actively take control actions to maintain control objectives in response to changes in disturbance variables (DVs).
Linear programming (LP) is an algebraic optimization technique that utilizes multiple linear equations to relate various economic or process variables By solving these relationships, LP aims to either maximize or minimize an objective function, which typically represents an economic measure of operational efficiency.
Process values that are adjusted by the advanced control system application to meet operating targets and desired values of controlled variables.
Refers to open data interchange functions that are based on standards developed and maintained by the OPC Foundation
Process performance is assessed through direct measurements obtained from instrumentation sensors and transmitters, as well as values derived from laboratory testing and other methods.
The effectiveness of an advanced control system application is assessed through a complex calculation that goes beyond a simple on or off measure This evaluation takes into account the various subfunctions within the application, the operational time percentage of these functions, and the economic significance of each subfunction.
An input variable determines the target value of a controlled variable and can be set manually, automatically, or through programming Its value is represented in the same units as that of the controlled variable.
3 Control System Functions and Types
General
This section describes functions common to advanced control and related systems Figure 2 illustrates basic control system functions and how they relate to this and other RPs.
Regulatory Control System Functions
Regulatory control systems are essential for managing process variables like pressure, temperature, flow, and level by adjusting final control elements such as control valves and electric motors The primary control algorithm employed in these systems is the proportional-integral-derivative (PID) method, although other algorithms may also be utilized The functions and standards of regulatory control systems are outlined in API 554, Part 1.
Complex regulatory control systems typically combine a number of functions used in regulatory control systems to meet a control objective Typical applications include:
— dynamic compensation (e.g filtering or time-shifting of variables),
— variable gain adaptive controllers or other nonlinear algorithms,
— calculated variables such as pressure and temperature compensation of flow or other simple computations,
— override control using output selectors,
The regulatory control system must be designed for remote access to its setpoints, enabling the advanced control system to make necessary adjustments Additionally, it should incorporate a feature to automatically disable these remote setpoints or revert to a predefined status in the event of a failure or shutdown of the advanced control system It is essential to document the reason for any failure or discontinuation of use at the time it occurs.
Model-based Control Systems
Model-based control systems enhance process performance by utilizing a mathematical representation of the process, which can be derived from fundamental engineering principles or empirical data gathered from the process itself.
3.3.2 First Principle Model-based Advanced Control Systems
First principle models are based on essential mass and energy balances, along with thermodynamics, equilibrium, and reaction kinetics When the process characteristics are well understood, these models enhance the accuracy and range of control systems Common applications include optimizing industrial processes and improving system performance.
— computation of heat and material balances or reactor yields and conversions,
Figure 2—Control and Automation Functions
Tank gauge and valve signals
Tank gauges, sensors, valves, and actuators
Blending flow meters and controllers
Safety and logic sensors, transmitters, valves, and actuators
Blending and oil movement controls
Alarm and abnormal situation management—
Lab operations and data management
— computation of inferred process variables that are not measured,
— improvement of control system operating range by including nonlinear effects.
First principle models often represent steady-state relationships, making them challenging to apply in processes with complex or variable dynamics A key difficulty lies in accurately determining the values needed to establish these relationships Additionally, process unit meters are not designed for custody transfer quality, highlighting the need to consider methods for inferring flows and qualities.
3.3.3 System Response Based Advanced Control Systems
System response models leverage real-world observations of dynamic performance to manage complex processes that involve multiple interacting variables These models are characterized by their ability to execute multivariable control, predict the impact of control adjustments, manage constraints, and optimize performance according to the parameters defined in the process model.
The system model enables controllers to link several manipulated variables (MVs) with multiple disturbance variables (DVs) and control variables (CVs) This approach allows for the simultaneous adjustment of all MVs to optimize the performance of all CVs, ensuring they reach their ideal operating points.
Process models identify the optimal operating point by considering user-defined constraints and economic factors Generally, this optimal point corresponds to the maximum number of achievable system constraints As illustrated in Figure 3, the relationships between operating points and constraints are complex, as real-world applications often involve multiple operating values and constraints.
Figure 3—Operating Conditions vs Constraints
Operating range without advanced controls
Controlled Variable Minimum stability constraint
Operating range with advanced controls
Empirical models derived from process testing are utilized by matrix controllers to establish relationships between manipulated variables (MVs), disturbance variables (DVs), and controlled variables (CVs) These controllers often include transformations for inputs and outputs to manage process nonlinearities, with some advanced versions employing nonlinear models Having been in industrial use for many years, matrix controllers have significantly evolved, enhancing the console operator's span of control to its maximum horizontal limits.
Neural networks function similarly to matrix controllers, but they incorporate a learning engine that creates a process model from observations This learning engine acts like an optimizer, integrating user-provided rules and observations to develop an effective process model.
Neural net controllers require sufficient “good” process historical data to create a model If such data is not available, a data acquisition period will be required.
Neural net controllers are capable of generating nonlinear models and adapting them to handle changes without the need for formal process testing These adaptations are usually executed offline, utilizing parameters derived from the online neural net application.
Optimizers
Optimizers utilize a computational approach to identify the optimal operating point based on user-defined economic objectives and process models These optimization programs typically refresh operational targets every few hours to days, factoring in the frequency of economic changes at each site The results generated by optimizers serve as steady-state objectives for other controllers, including multivariable controllers responsible for dynamic control.
Optimizers can function in two modes: offline and online In offline mode, the model does not receive dynamic data from the process, while in online mode, it is connected to key real-time process data.
An online mode application can operate in either open-loop or closed-loop configurations In closed-loop mode, outputs are automatically directed to dynamic controls, ensuring seamless adjustments Conversely, in open-loop mode, the optimizer's outputs are relayed to the operator, who must then manually implement changes to the dynamic control system.
In online applications, it is essential to perform thorough data validation and reconciliation to ensure accurate results Additionally, maintaining steady operations during the optimizer's execution is crucial, although some advanced optimizers may not necessitate this stability.
Multivariable controllers often incorporate embedded linear programming (LP) to address optimization challenges These embedded LPs utilize linear relationships among controlled variables (CVs), manipulated variables (MVs), and the economic implications—either costs or benefits—associated with each variable.
Unit level optimizers are essential for determining operating targets in multivariable controllers, leveraging a wide array of variables and incorporating thermodynamic, equilibrium, or kinetic relationships These optimizers necessitate functions for data validation, reconciliation, and parameter estimation, and they can access external economic data By employing sophisticated nonlinear models, unit level optimizers effectively identify the optimal operating conditions for the entire unit.
Linking multiple unit level optimizers enables the system to identify a global optimum for the entire plant, though this process is highly complex Achieving accurate results necessitates the use of rigorous models Additionally, data reconciliation with overall material balances and yield accounting systems is essential for effective plant level optimization For instance, if the product from one unit serves as the feed for another, even a minor error from a flawed model or inconsistent data can propagate through the optimizer, leading to suboptimal outcomes.
The effectiveness of these optimizers is frequently constrained by the precision of unit optimizers in large systems, where the computational demands and input data volume can be substantial However, notable progress has been achieved through the application of offline linear programming (LP) models across various plants.
Expert Systems
Rule-based systems are ideal for scenarios with predetermined events, offering detailed operational insights based on past occurrences Commonly utilized in an advisory role, these systems operate independently from direct control systems A notable emerging application of expert systems is in Abnormal Situation Management.
Fuzzy Logic Systems
Fuzzy logic is employed in scenarios where rules are imprecise, allowing for the transformation of vague guidelines, like "if the temperature gets too hot, then slowly increase the cooling water," into concrete actions As a growing technology in process industries, fuzzy controllers can be integrated into equipment controllers or offered as part of software packages Additionally, many process control systems include fuzzy control blocks within their algorithm frameworks.
Batch and Sequence Systems
Control strategies are implemented for processes with a defined sequence of finite steps, such as reformer regeneration, water treatment, and coke handling Detailed descriptions of these operations can be found in ISA S88.01, but they fall outside the scope of this RP.
Blending Systems
Component blending plays a crucial role in fuel production complexes, utilizing systems that consist of blend ratio regulatory controls, property estimators, and optimizers to ensure recipes align with final property specifications While this document details advanced control strategies employed in these systems, it does not cover the specific blending practices involved.
Oil Movement Systems
Oil movement systems use heuristic rule-based systems and are not within the scope of this RP.
Manufacturing Execution System
This application focuses on managing manufacturing operations, addressing key aspects such as manufacturing scheduling, raw material management, and resource planning It operates at the business network level, which falls outside the scope of this document.
Resource Requirements
Identifying enhanced control opportunities requires a multidisciplinary approach, where incremental economics, statistical data reconciliation, and operations planning are crucial The success of advanced control projects heavily relies on the engineers' expertise and experience, underscoring that while suitable technology is significant, it cannot compensate for insufficient skills among the personnel involved in implementation, maintenance, and application usage.
A feasibility study assessing the benefits of a single process advanced control application can be effectively conducted by a qualified individual with access to specialized support However, for larger-scale studies, particularly those involving an entire facility, it is advisable to adopt a small, mixed-discipline team approach This team should consist of members from site process/planning, operations, and control system engineering to ensure comprehensive analysis and effective implementation.
Effective benefit identification hinges on a blend of technical and interpersonal communication skills It necessitates a deep understanding of the economic factors influencing processes, such as incremental process and energy economics, as well as the timing of value changes Additionally, one must comprehend how these processes function and interact, recognize their inherent limitations and constraints, and secure buy-in from advanced control users and business support specialists, including budget planners, schedulers, and economists, to realize the identified benefits and solutions.
Economic Drivers
Effective business planning, both long- and short-term, is essential for assessing potential benefits and should take into account various economic drivers While not every driver will apply to a specific market or facility, they provide valuable insights into the economic forces at play Key economic drivers to consider include factors that influence market dynamics and financial performance.
— feedstocks and finished product pricing;
— volume and shipping methods for crude and other feedstock;
— demand and shipping method of finished products;
— utility usage, prices, and contractual obligations (e.g requirement to meet energy forecasts).
— seasonal variations in product demand, quality, and supply;
— the economy in which the plant operates (e.g open or closed market economic operation);
Identification of Potential Applications
Identifying opportunities for process improvement begins with discussions involving site personnel from various functions, alongside a preliminary assessment of performance data This assessment should include a review of actual plant performance against production targets, comparisons with best-in-class benchmarks, and an analysis of economic drivers that highlight value potential Additionally, it is essential to consider any anticipated changes to the process plant, recognize improvements achieved through existing advanced control systems, and explore further control enhancement opportunities using well-tuned steady-state models aligned with statistically reconciled data.
A number of improvement areas that may be considered are listed in Table 1
The preliminary assessment yields a list of potential applications along with their associated benefits, which should be reviewed by relevant personnel to prioritize these benefits This process will help identify applications with high potential that warrant further investigation.
Identification and Quantification of Benefits—Feasibility Study
A detailed examination of the top-ranked items reveals significant potential benefits, as outlined in Table 2 For existing plants, relevant data can be sourced from their operational observations, while for new plants, equivalent data can be gathered from similar operations or process models.
The initial phase of a feasibility study involves defining the goals of the proposed control application through discussions with site representatives across planning, technical, and operations teams It is crucial to scrutinize and question process constraints and production targets to ensure they are accurate Any identified constraints should be evaluated through testing and data collection, as they may be based on perceptions rather than actual limitations.
Identifying the potential impacts of proposed control improvements on other units or utilities is crucial, as an increase in the rate of an intermediate product stream could surpass the capacity of downstream equipment Nevertheless, it is essential to evaluate the advantages of selling any surplus intermediate products.
— Reduced quality giveaway on intermediate or final product
— Control closer to targets Stability
— Production closer to constraints for better control
— Higher production by better operation against process equipment constraints and limits such as:
— minimization of unwanted material in feedstock (i.e nC 4 in alkylation feed)
— Improved reactor performance by better control of:
— reactant and other key process variables
— Reduce total energy costs by better control of:
— Improved efficiency of heater operation through:
— swing fuel firing control to maximize usage of lower value or variable fuel
— injection steam ratio control where appropriate
— multiuser control strategy (i.e hot oil systems)
— pass outlet temperature balance control
(See API 556, Instrumentation, Control, and Protective Systems for Gas Fired Heaters for basic combustion control schemes, including excess O 2 and stack temperature)
— Enhanced understanding of the process for better management of all site functions
— Improved equipment reliability due to more stable operation at intended conditions
— Sharing of best practices from other applications
The goal of this step is to gather all essential data and information needed to measure the actual performance of the process against its operational targets and control objectives.
To ensure statistical significance and consistency, it is essential to collect various types of data, including operating targets such as product qualities and yields, and comparisons to LP vectors for raw material selection Actual values achieved, including reconciled feed and product rates, should also be documented Furthermore, it is important to identify process operating constraints and limits, including throughput and environmental restrictions, as well as reasons for outages or restricted operations Measurement availability, covering analyzers and flow data with a focus on accuracy and reliability, must be considered Additionally, discrete events like coke drum sequencing and control system performance indicators should be monitored Future plans for process modifications and potential impacts, along with changes to production operating targets and operational flexibility requirements, should be outlined, alongside economic driver data and long-term planning considerations.
Table 2—Benefit Feasibility Study Steps
2) Identify representative period of operation
3) Collect and validate process data
4) Analyze data for performance against targets
6) Analyze control infrastructure performance and requirements
8) Categorize benefits on a confidence basis
9) Review and agree benefits analysis with site, including business support personnel
10) Set up postapplication benefit audit basis
In the absence of necessary data, it is essential to make informed assumptions regarding missing information and its effect on the reliability of economic forecasts These assumptions must be communicated and agreed upon with site personnel, including business support staff, before advancing the project beyond the initial phase.
To effectively identify benefits, it is crucial to select one or more representative operational periods that reflect "normal" process conditions, free from shutdowns, deficiencies, or unusual production demands Variations such as weather impacts and day/night operations should be accounted for A minimum duration of one month is advisable, especially when using infrequent laboratory data for analysis The analyzed data must be statistically significant and consistent, aligning with the overall material balance of the site during the selected period.
To accurately assess potential benefits, it is essential to analyze data from various periods throughout a unit's operational run This analysis should consider factors such as seasonal variations in plans and targets, ambient conditions, and the differences between the start and end of the run.
The representative periods will be utilized to prorate potential benefits on an annual basis, based on a mutually agreed number of stream days per year, which can differ by process and site In the absence of specific data, a default of 350 stream days per year is commonly applied.
Failure to establish realistic representative periods can result in inaccurate benefit predictions For instance, a specific process unit demonstrated actual audited benefits of $900,000 during summer months, contrasted with only $100,000 in winter The initial feasibility study relied solely on summer data, projecting an annual benefit of $1.8 million, which led to an overestimation of $800,000 per year.
Before utilizing collected data for benefits computations, it is crucial to validate it through several key steps First, eliminate any clearly erroneous process data, particularly from periods of identified "non-normal" operations Next, conduct a basic statistical analysis, calculating the mean and standard deviation of the actual values for targets and constraints Finally, compare the data against site-wide reconciliations and consider creating inference equations to predict flows and qualities, as understanding the trade-offs between different streams and qualities is essential.
Statistics can vary based on the established targets and often indicate a nonlinear response of the process Therefore, it is more insightful to compute the statistical average and standard deviation of the differences between target and actual values, as well as to calculate a time-weighted average for the actual values.
Statistical analysis is a crucial method for evaluating the benefits of potential control improvements Practitioners commonly aim to reduce the standard deviation of controlled variables, such as targets and setpoints, to enhance operational efficiency and approach constraints more closely While a 50% reduction in standard deviation is often assumed, it is essential to assess the current value before making this assumption, as actual audits of control applications may reveal that this reduction can be overly conservative.
Improving the stability of controlled variables often requires more frequent adjustments of manipulated variables (MVs), leading to a significant reduction in variability This reduction allows operations to function closer to process constraints, resulting in tangible benefits such as increased process rates, enhanced product quality, and minimized giveaway, ultimately yielding economic advantages.
Another valuable analysis is to compare operations variability among operating crews Often significant benefits can be quantified by comparing “best” operation versus average operation within the site.
Choosing the right prediction method requires skill, as limitations in information or current capabilities may lead to the selection of an inappropriate approach Additionally, it is crucial to account for the time-weighted average of control targets, especially when those targets are subject to change.
General
This section focuses on the distinct elements of project management and execution specific to advanced control projects, rather than covering general project execution topics like budget, schedule, and contracting It highlights the unique challenges and considerations that arise in the context of advanced control initiatives.
An advanced control project typically involves the implementation of multiple applications, along with infrastructure elements like field instrumentation, measurements, and enhancements to regulatory control This encompasses both the necessary hardware and software associated with these applications.
See API 554, Part 1 for additional guidance on project planning and execution.
Master Plan
Before initiating advanced control projects, a facility must establish a comprehensive automation master plan that integrates advanced process control, backed by management support and aligned with strategic business objectives This plan should outline long-term goals and phase the work accordingly, while identifying the benefits of each phase It is essential to define key performance indicators (KPIs) for monitoring post-implementation success and specify the control and computing equipment, including software platforms and their permissible applications Additionally, the plan must assess existing and future control networks, establish implementation standards, and address critical timing for potential upgrades during unit turnarounds A schedule for regular plan reviews, resource requirements for ongoing maintenance, and a smart instrumentation strategy should also be included, ensuring data consistency across the site while considering the accuracy of instrumentation in flow and quality assessments.
Project Execution Plan
Before starting an advanced control project, it's essential to create a project execution plan that outlines how the project's objectives will be achieved and sustained This plan should be based on the insights from the feasibility report and master plan, ensuring a comprehensive approach to project implementation.
— use of dynamic simulators to validate the design and train personnel,
— design plant testing and implementation plan,
— ongoing application support and maintenance.
Implementation Issues
Advanced control projects face unique implementation challenges not typically found in other types of projects It is essential for the project engineer to understand the facility master plan's requirements and ensure that the project is executed in accordance with these guidelines.
Implementing advanced control applications presents several challenges, including the need for upgrades to existing control systems, which can range from complete overhauls of outdated instrumentation to incremental improvements Projects often require additional or upgraded process measurements, some of which may be inaccessible during normal operations, necessitating careful planning for their delayed availability, particularly for online analyzers that must be operational before use In certain cases, inferred values can substitute for direct measurements, though this may require further instrumentation New construction projects typically delay advanced control system implementation until after construction is complete and operational data is available, and they may also necessitate modifications to existing systems, potentially requiring redesigns to accommodate changes in process equipment or operating conditions Additionally, many advanced control projects depend on unit shutdowns to install necessary measurements, which can significantly influence project timelines Compliance with process safety management regulations mandates conducting hazard analyses and implementing management of change procedures to ensure proper review of system design developments The design of an advanced control system may involve various techniques, including process modeling and empirical testing, and significant process modifications may delay some of this work Lastly, integrating multiple software packages from different suppliers requires a comprehensive integration plan, balancing the benefits of a single software suite against the complexities of maintaining diverse applications, with thorough testing in a virtual environment being essential.
Personnel Commitments
Successful execution of an advanced control project hinges on the commitment of a skilled workforce The availability of knowledgeable and experienced personnel is crucial for the project's success, as attempting to implement such initiatives without adequately trained staff often leads to failure.
Advanced control projects necessitate a significantly higher allocation of total costs towards engineering compared to other design and construction projects To ensure ongoing benefits, these advanced control applications also demand regular maintenance Therefore, process facility management must acknowledge the necessary personnel commitments and weigh them against the anticipated benefits when making staffing decisions.
Personnel involved in advanced control projects must possess a strong understanding of relevant practices and techniques This expertise encompasses process control engineering, as well as familiarity with control system hardware and software, alongside general process engineering knowledge.
Appropriate training should be planned and budgeted to ensure necessary skill set capabilities See 5.6.4 for additional guidance with regard to training.
Effective project management skills are crucial for the successful execution of advanced control projects, which require a tailored management methodology These projects often face unique challenges, such as critical path schedules influenced by unit shutdowns, extended phases with minimal visible progress due to operational constraints, and prolonged implementation, testing, and validation periods.
The involvement of specialists in process engineering, computer implementation, and analytical methods is essential for the success of advanced control projects IT professionals, equipped with strong communication and data management skills, play a vital role in fostering virtual and collaborative environments Whether part of the owner's team or contracted from manufacturers or vendors, these experts may need to dedicate considerable time to the project.
When an owner lacks adequately skilled staff, engaging manufacturer, vendor, or consultant personnel can enhance existing capabilities and maximize application benefits It is crucial for the owner to verify that these external personnel possess the necessary skills and training for effective support.
5.5.6 Plant Operations and Maintenance Support
Successful advanced control projects rely heavily on the active support and involvement of plant operations and maintenance personnel Their insights are crucial for establishing control objectives and strategies, as well as for the effective implementation and ongoing maintenance of control schemes and operator interfaces It is advisable to include a skilled operator and a knowledgeable unit engineer on the project team to ensure comprehensive understanding of the process and systems involved.
Effective maintenance input is essential for the advanced control system to manage both routine and unscheduled maintenance of sensors, transmitters, control elements, and equipment Additionally, it is crucial for implementing engineering-oriented changes, including loop tuning and configuration adjustments Addressing these aspects is vital for the comprehensive design of an advanced control system.
MOC processes should be followed for all facility modifications necessary to support the project.
Schedule
Advanced control projects exhibit distinct scheduling characteristics, leading to phases of intense activity followed by significant intervals of minimal or no observable activity.
Before initiating an advanced control project, it's essential to outline all major tasks and their anticipated durations, while also identifying any schedule constraints related to resource availability and operations Typical tasks involved in such a project are detailed in Table 3, though the specific tasks, their durations, and lead times can vary significantly based on the project's scope, plant operations, and the experience level of the engineering, operations, and maintenance teams.
The availability of skilled personnel for advanced control projects is limited, making it crucial to define the required skills and the number of personnel during the scheduling process This availability can significantly impact the overall project timeline.
Implementation of advanced control projects is often dependent on ongoing operations Shutdowns or turnarounds may be required to add or modify process measurements or final control elements
Process testing is essential for developing process models or relationships in advanced control systems This testing can take several weeks and may be affected by unstable operations or conditions that differ from those anticipated Additionally, it is common for changes in operating conditions or objectives to necessitate further testing after the initial phase is completed.
Developing tailored training programs for operations, engineering, and maintenance personnel is essential, as the content and timing of the training will vary based on individual responsibilities and prior knowledge An example of a comprehensive training program can be found in Table 4.
The training program must encompass initial training, refresher courses for current staff, onboarding for new personnel, and operations training without the advanced control application It should also include a system for tracking individual training histories Additionally, certification may be required for operators to ensure compliance with local regulations regarding the operation of the process and its control systems.
Training manuals for operators and maintenance technicians must be comprehensive, detailing the application and its implementation, all operator interfaces and reports, and a complete list of inputs and outputs along with their maintenance requirements, particularly for specialty sensors and online analyzers The manual should also provide clear instructions for putting the application on and off control, responding to alarms and alerts, and managing operations in degraded control modes, such as partial shedding and using substituted values Additionally, it must outline the application maintenance requirements and procedures to ensure effective operation.
Initial classroom training introduces operators to the general application, while on-console training and support from advanced control application engineers are essential during initial testing and commissioning This comprehensive training may span several weeks, and formal assessments should be conducted for console operators to ensure their proficiency.
Table 3—Typical Advanced Control Project Tasks
Facility master automation plan 12 to 60 months prior to project initiation
Feasibility study and project identification 6 to 18 months prior to project initiation
Project implementation plan At initiation of project
Functional specification Start of project scope definition
Regulatory control and measurement design 6 to 12 months prior to commissioning
Design specification With regulatory control and measurement design
Advanced control software purchase 6 to 12 months prior to commissioning
Hardware platform purchase 6 to 12 months prior to commissioning
Implementation and testing on a dynamic simulator In conjunction with design development
Model identification testing 3 to 6 months prior to commissioning
Model identification and implementation 2 to 4 months prior to commissioning
Engineer training Initial at software purchase General training 4 to 6 months prior to commissioning Operator and engineering graphics implementation 2 to 4 months prior to commissioning
Advanced control hardware and software installation 2 to 4 months prior to commissioning
Regulatory control and measurement installations must be completed within 2 to 4 months prior to commissioning, aligning with operational and design scope Additionally, advanced control applications should be installed, tested, and simulated within the same timeframe to ensure optimal performance at launch.
Operator and maintenance training 1 month prior to commissioning and during commissioning
Commissioning 1 to 2 months—subject to operations schedule—reference point for other tasks Initial advanced control system operation Commissioning and 2 to 4 months after commissioning
Full time advanced control system operation 2 to 4 months after commissioning
Close-out documentation 1 to 4 months after commissioning
Postcommissioning economics audit 3 to 6 months after commissioning
Advanced control system adjustments and modifications 2 to 6 months after commissioning
It is important to note that the tasks outlined may vary based on the specific control technologies employed, and additional or modified tasks might be required Training should encompass both classroom and on-console instruction, with a comprehensive training plan that includes on-process training for shift operators and post-commissioning sessions to inform personnel of any changes made during the commissioning phase.
If advanced controls fail, unit operators must be prepared to take safe control of the system Modern process control systems enhance stability, significantly reducing operator exposure to upset conditions when these controls are functional To ensure operators can effectively manage situations with degraded control, additional training is essential.
Process simulators serve as an efficient training tool, leveraging advancements in computer technology to create high-fidelity dynamic simulations These simulators effectively mimic the behavior of processes, control systems, and operator interfaces, making them a cost-effective solution for training Regular refresher training for operators using these dynamic simulators is increasingly recognized as a practical method for sustaining operator skills.
Testing and commissioning advanced control systems involve extended processes due to their complexity These systems require thorough testing and monitoring of numerous variables and various operating conditions to ensure optimal performance.
Table 4—Advanced Control System Training Program
Organization Personnel Content Location Timing Method
Console operator/ liaison with advanced process control project
Detailed commissioning, operation with and without controls
Vendor facility or plant training facility
At functional design with continuation during implementation and at commissioning
Classroom, detailed simulators, on process
Other operators Operation with and without controls On site
Prior to commissioning and continuation during commissioning
Classroom, on console, and simulator
Overview of functions and objectives Office Prior to commissioning Presentation
Objectives, process, control technologies, and detailed application
Prior to project and during functional design Classroom, simulator
Objectives, application details, commissioning, operation with and without controls
Vendor facility or plant training facility Prior to commissioning Presentation and classroom
Project engineer Objectives and scope Office At start of project Presentation
Control technologies and detailed application maintenance
Plant training facilities During design Classroom, simulator, actual system
Instrument and process control system technicians
Maintenance procedures with and without controls On site
Prior to commissioning and continuation during commissioning
Testing an advanced control application involves offline or simulated operations to ensure its performance meets expectations Key objectives include verifying input and output connections, assessing the application’s response to bad values, and confirming the functionality of the watchdog timer for communication and program failures Additionally, it is essential to validate that intermediate calculations yield expected results, ensure that predictions and control outputs align with expectations, and develop preliminary tuning constants Finally, the testing process must also engage all interfaces for operators, engineers, and maintenance personnel.
Commissioning consists of placing the application on process and observing and adjusting its performance for a long enough period to demonstrate acceptable operation
Initial commissioning typically occurs on a limited basis, usually during day shifts or when key operating personnel are present This phase requires significant specialist support to monitor control system performance and identify necessary modifications For larger or more complex applications that cannot be easily turned on and off, it is essential for the control application and commissioning support to be available 24/7.
Application Documentation
Every advanced control project must generate a comprehensive documentation package for handover to the plant's operations, maintenance, and engineering teams Each application requires its own documentation package, which generally includes essential items outlined below.
In addition to essential documents, various project-oriented records may be archived for historical reference, including feasibility studies, scope definitions, estimates, and project execution records However, this description does not encompass files that are specifically required for turnover.
A functional specification for the advanced control application should be prepared and maintained during implementation This specification will be an update of the specification developed for the feasibility study.
The design specification outlines the detailed requirements necessary for implementing the application as described in the functional specification It includes a comprehensive description of the application's objectives and functions, defines control technology and communication methods, and delineates the scope of regulatory control and measurement enhancements Additionally, it addresses security concepts, expected results, and key performance indicators (KPIs), while also detailing business processes and workflows, input/output (I/O) requirements, and model specifics.
— expected MVs, CVs, and DVs;
The article outlines essential components for system management, including the estimated response time for each controller, application alarms and alerts, and user interfaces for operators, engineers, and maintenance personnel It emphasizes the requirements for activating and deactivating the application, as well as managing control degradation behaviors Additionally, it details the database structure and contents, maintenance and training requirements, and the importance of history collection for effective system oversight.
Section 7 describes a number of issues that should be considered while developing this specification.
The following types of documents represent the actual installation in the field and shall be kept current.
The applications configuration database will be documented in detail, including its contents and structure Regular backups of the database will be conducted to ensure data integrity Additionally, procedures for verifying the database's consistency with the officially approved set will be implemented.
The essential documentation for graphics includes a hard copy of each graphic in color, along with a detailed list of code or configurations that govern the behaviors of all active graphic elements This documentation should also extend to all configured reports Additionally, it is crucial to create electronic backups of all graphics and report files at regular intervals to ensure data integrity and accessibility.
Maintain current listings of all application programs in both electronic and hard copy formats It is recommended to include descriptive information detailing the purpose, function, inputs, and outputs of each program Additionally, incorporate program flow diagrams where relevant, with a preference for those that illustrate the operator's actions, such as business processes, over standard program flow charts.
Copies of training procedures, records, and manuals shall be maintained.
Copies of maintenance procedures, records, and manuals shall be maintained.