The immersive virtual reality plant
Capital-intensive industries face the challenge of replacing an ageing workforce with a computer-savvy, gaming generation over the next five years. Industries such as oil and gas, and refining and power companies must institutionalise their workforce knowledge in efficient and effective ways. Leveraging virtual reality models to improve time-to-competency in critical areas such as safety and environment protection systems, knowledge and performance training, and reliability provides a vehicle to rapidly train this new workforce in ways that align with their interests and skills.
Today, with continuing advances in hardware and software techniques, virtual reality (VR) is viewed as the best aid to improving multimedia training, process design, maintenance and safety, which are currently based around conventional 2D equipment views.
The real-time rendering of equipment views places demands on processor time, with the use of high-fidelity simulators becoming the standard in process understanding and training. For many past VR commercial projects, the results were unrealistically slow or oversimplified to the detriment of the solution effectiveness. Today, as technology evolves, these will no longer be an issue with the new process simulation era.
A revolutionary new training medium
Virtual reality is a rapidly growing technology that utilises the increased power of computers to simulate real or imaginary environments and situations with a high degree of realism and interactivity. It is an emerging technology with potential applications in areas such as product design and modelling; process simulation; planning, testing and verification; real-time shop-floor controls; and training and maintenance. One speculation on the development of virtual reality in the manufacturing industry is a computing architecture that creates virtual production environments within concurrent engineering contexts, achieving zero defect and non-risk production. In fact, virtual reality has already been applied to a wide range of problems associated with manufacturing, industrial maintenance, post-production training and customer services in areas such as visualisation of complex data; robot control and remote operation of equipment; communication, training and planning; and virtual prototyping and design. The success of those applications has relied heavily on a realistic virtual environment.
Currently, VR technology is used for training applications in a variety of process industries, and offers the potential to expose personnel to simulated hazardous situations in a safe, highly visual and interactive way. Customised simulations of chemical plants layouts, dynamic process operations and comprehensive virtual environments can be set up and allow users to move within the virtual plants, making operational decisions and investigating processes at a glance. The consequences of correct and incorrect decisions are sent immediately back to the trainees, giving them the opportunity to directly learn from their mistakes.
Users can interact with the virtual worlds using a variety of hardware devices such as joysticks and data gloves. Special optical and audio devices (such as head-mounted displays, three-dimensional graphics and surround sound) allow users an enhanced impression of being in the virtual world.
The overall aim in developing an immersive virtual reality plant (IVRP) is the development of novel training techniques that improve the operation efficiency and skills of chemical plant personnel. This includes the development of a large range of training scenarios for application in the chemical process industry. Using IVRP will increase process understanding, readiness, safety awareness and knowledge of safety procedures, improving production and reducing the plant accident rate.
Augmented reality is a combination of virtual and graphic three-dimensional images that are transmitted to a user. Augmented reality differs from virtual reality in overlaying images onto the real (virtual) life, engaging the user in an immersive, interactive and three-dimensional ‘augmented’ environment.
The external environment represents the real (virtual) world and can be linked to additional information systems. For example, in a medical application, it is feasible to overlay a sketch of human organs onto a real (virtual) patient via an optical system. In order to achieve accurate alignment of the real/virtual environments and multimedia data source, it is important to provide the most appropriate link among the display technology, the simulation environment and the information processing systems. All VR scenarios and applications must relate to the external data source either as simulated or as field data. In fact, the value of VR tools and related benefits for training purposes are strictly related to the ‘action/reaction’ feeling of the trainees, as well as the availability of additional information and data that provide a ‘supernatural’ sense of power and understanding of the process.
By superimposing virtual images, videos or text onto real (virtual) life, an experience can be heightened or even modified.
Figure 1 shows an example of a trend diagram, which can be activated or deactivated by the user with a simple touch on the hand device at any time during the plant walkthrough or a task procedure. Variables and trends can be selected and customised by type (such as temperature, pressure and flow) and equipment exactly displayed in the control rooms. The data source will be identical, simulation engine or real-time database. Alternatively, a different type of augmented reality is possible, where equipment units become transparent and the information related is shown as a video animation of the process behaviour.
Emergency, risk and safety
There have been a number of well-publicised accidents in which human error has played a prominent role.
In order to understand the contribution of human behaviour to the risk of accidents, it is essential to examine the errors people make and what leads to such errors. The reduction of human error probability can lead to a reduction in the probability of accidents in process industries. A useful classification framework identifies the human errors as slips/lapses, mistakes or violations. Slips or lapses typically occur through lack of attention or from stressful situations, with the result that individuals fail to achieve what they intended. Slip or lapse human errors include forgetting to turn off power, becoming inattentive to important safety checks or forgetting important steps in safety procedures, which may cause equipment damage or loss of life.
Training has always been considered an important factor in staying competitive in a global economy. Operators must remain up to date with the latest methods and technology. Training should also involve an introduction to basic hazards and plant safety procedures incorporating fire alarm systems and detailed work safety processes. Therefore, there is a need to create hazardous simulated conditions to provide a real-world (virtual) situation for training without causing harm to the trainee and the environment.
IVRP systems allow users to navigate in any direction within a computer-generated environment, creating crisis conditions, deciding what actions to take and to immediately see the impact of those actions. IVRP also allows trainees to walk throughout the plant, observe all the equipment that constitutes the process, and to have the possibility of starting, running and shutting down equipment and responding to error conditions.
In theory, every abnormal situation an operator can imagine but never test in reality can now be tested, helping the operator understand atypical plant behaviours. Expected and predictable malfunctions can be tested and forced until the accidental sequence results in a disaster. Learning from a virtual disaster can help avoid a real disaster. The main goal is that safety can now be tested and experimented with for training, and to help risk assessors better identify hazardous scenarios. Most importantly, IVRP can aid operators in making the right decisions in time.
Maintenance and reliability
Plant maintenance is a best practice that requires a team-based approach, where operators perform various equipment maintenance activities and maintenance crews work closely in the daily operation of equipment.
To understand requirements for maintenance training, the first logical step is to review the task the trainee is expected to perform and the outcome of that training. For example, in process operations, the nature of the maintenance task is heavily dependent on the organisation’s industry sector, range of equipment maintained and specific culture.
Irrespective of the subject matter, maintenance tasks generally can be broken down into the following subtasks:
- Replication: reproducing the reported fault;
- Identification: accurately diagnosing the source of the fault;
- Rectification: correcting the fault by taking the appropriate action based on the policies of the maintenance establishment;
- Confirmation: verifying that the identified fault has been cleared.
Each of the four stages described above requires a mixture of generic and specific physical and mental skills. Using VR facilities, maintenance operators can be trained to have a full understanding of the maintenance task and the science behind the equipment they are dealing with.
Inherent in VR models is the ability to understand the geometry, layout and limits of process units and their supporting utility infrastructure. Using a physical representation of an operator or maintenance personnel (avatar), IVRP can be used to optimise gauging and inspection rounds by imitating the operational and maintenance staff behaviour.
This is primarily a spatial system analysis where working spaces, escape routes, risky areas and transportation routes in the plant could be investigated from a logistical point of view. The resulting analysis can be used to optimise maintenance procedures, or make requests to maintenance management for improvements or modifications.
With the aim of preventing production losses, maintenance during operation has high priority in such facilities due to the commercial impact. Therefore, to verify that maintenance is properly conducted and performed on time, maintenance personnel are asked to participate in maintenance training using avatars or in ‘first person’ through VR models of the process machinery and layout. Maintenance issues on diagnostics, timing and procedures are highlighted and optimised by interactive links to the virtual equipment, as well as to a computer-integrated maintenance and documentation management system. Fault-diagnostic training and rectification training can therefore be based on a comprehensive multimedia tool that transfers knowledge and skills.
Furthermore, model areas can be colour coded to represent areas of the plant that require inspection, as well as establish safety and integrity risk boundaries.
Benefits and value proposition
IVRP can improve design procedures, saves staff time and money in maintenance and is a superior way to train operations personnel by:
- the provision of more realistic training environments for trainees
- the provision of opportunities for practice in training sessions
- training staff to react quickly and correctly in high-stress situations
- improving skills for rarely performed, but safety-critical tasks (such as emergency shutdown)
- optimising the transfer of skills from the training environment to the work environment
- reliable and valid evaluation of operational procedures and performance
- team training for the control room, field, shift, operational and safety managers.
Typical return on investment on an IVRP is expected to be more than 50%. The technology’s main economic benefits fall within the following main categories:
- On-the-job training: Saving 30 to 40% on time and costs for new large personnel requirements, familiarisation in case of mobility from plant to plant and scheduled replacement of the ageing workforce.
- Start-up efficiency: Saving 15 to 25% of the time to be ‘back-on-run’ and optimal production for planned or unplanned shutdown, while frequently having all plant crew refreshed on all critical procedures.
- Maintenance: Saving 1 to 3% on maintenance budgets by improving maintenance operator training and by using IVRP equipment pre-analysis for predictive maintenance.
Trainers and trainees can be located throughout the world, and using web facilities, remote workers can work together in the same virtual area on a chemical plant or a building site, resulting in a reduction in travel costs. Interactive programs with multiple sites networked for group learning and communication could also be developed.
By Maurizio Rovaglio and Tobias Scheele, Invensys Operations Management
Digital technology is helping mining companies across the world realise the true potential of...
Commissioning a new system for a food manufacturing plant requires software assistance to plan...
Online condition monitoring involves a number of challenges aside from data collection: data...