What does optimal plant control mean? Firstly, it is the need to keep maximum equipment utilization. Secondly, it is the current characteristics of the raw materials and adjusting the technological process for these peculiar properties, i.e., the correct selection of distillation cuts, taking into account the objective parameters of the feed. Thirdly, it is the environmental conditions, for example the cooling water temperature in summer/winter. And finally, the fourth component is market factor as the cost of a specific oil fraction or other product, as well as the ability to quickly adapt to market conditions.
Solving all these issues which are named as «optimal plant management», may become more difficult due to economic reasons. Of course, departments for continuous optimization and improvement or mathematical modeling require significant expenses. Not many companies can afford to have such departments. Small refineries are particularly sensitive in this regard. They face many other tasks, on the solution of which the success of the plant depends. This includes purchasing and selling raw materials; hiring employees who can operate the plant responsibly and with understanding; ensuring uninterrupted logistics of raw materials / products; solving problems and maintaining the operation of utilities and off-cites (recycling water supply unit, fire extinguishing systems, alarm systems, communications, flare system, boiler room, etc.); interacting with federal environmental and supervision services.
Against the background of all these daily issues, often it is not possible to recognize the need for enterprise transformation. If management sees a need for process improvement, it is not easy to implement this in a small refinery. There may be several reasons for this:
- control of the unit is made in accordance to the regulations developed many years ago and not always including all the changes made to the design;
- the lack of specialists at the enterprise who are familiar in practice with thermodynamic modeling and have sufficient competencies for a comprehensive analysis of the plant and the development of its mathematical model;
- The general skepticism, and the conviction of specialists or enterprise managers that their knowledge of the operation of the plant is optimal and it is not possible to optimize its operation.
For example, to develop a digital twin, first of all, it is needed to collect initial production data. The more documentation is involved in the process, the better product will be developed. The regulations of the enterprise with technological schemes, and certificates for equipment with drawings, and operator logs for several years of work, and information on the upgrades carried out are also important. If some of the data is missing, alternative ways of building a mathematical model of the enterprise are being developed - they may not be as accurate, but they are sufficient to achieve the overall goals of the project.
After that, the specialists begin to analyze the initial data. They check the operator logs for incorrect sensor values and thus eliminate erroneous readings from the simulation basis. In addition, the analysis of the schemes and equipment of the enterprise is carried out.
The next step is to build a mathematical model, which may include several blocks (thermodynamic model; statistical model; model describing the hydrodynamic regime, etc.). This stage allows RRT Global specialists to provide the customer with an exact copy of his plant, which must be verified - an accuracy check on newly received sensor readings. The quality parameters calculated by the model and obtained at the facility rarely differ by more than 2-3% - as a rule, this depends on the of the initial data quality.
The fourth link in the optimization chain is a comprehensive analysis of the customer's plants, taking into account additional information about the operation of each equipment, obtained using a mathematical model, as well as the development of optimization scenarios. The latter can reduce production costs and energy costs, increase the yield of the most expensive product and reduce the loss of raw materials or the target product. Based on the results of the work, the customer receives specific recommendations for achieving optimization goals, whether it is changing the operating mode of a particular apparatus, cleaning heat exchangers, replacing the internals of column equipment with an indication of the type of devices to be installed or installing new positions of the equipment. It is noteworthy that optimization solutions can be both high-cost and practically free. The customer can compare the amount of costs with the payback period and decide on the feasibility of implementing the scenario. Another important aspect of work at this stage is that specialists develop various scenarios for controlling the plant, differing, for example, in different production of a particular type of product (gasoline fraction, kerosene, diesel fuel, heating oil, etc.). These scenarios will subsequently form the basis of the RTO system, which will be able to adjust the operation of the equipment to produce the most expensive product, taking into account the physical capabilities of the installation, such as incoming raw materials, environmental conditions, etc.
After all operations, the RRT Global team offers to create a digital twin of the plant, which works independently of the control loop of the plant, will receive the data from its sensors and would calculate a mathematical model for the current state. This digital twin becomes a great production assistant. It allows the operator and plant management to receive online additional information about the processes taking place in the plant equipment. This applies to data on the actual load of the equipment, and on the efficiency of heat exchangers, and on the current flowrate and composition of all product streams. This means that it becomes possible to more effectively manage the plant itself.
But this is not the end of the optimization process. Specialists of RRT Global are also engaged in the development of the «what if - analysis» system. It is aimed at calculating various scenarios of the installation when changing specific parameters. For example, you can calculate how much the gasoline cut will increase and what its quality will be with an increase in the column bottom temperature. It also allows to make a more comprehensive analysis of the current state of the plant, or to choose a scenario to proceed with.
Thus, a small business may well increase the level of knowledge about the existing production process. The main thing is choosing the right approach. Even at low costs, it will help reach the same level as the leading companies.
Author: Andrey Karmanovsky, CEO