Search

The sugarcane crushing stage is one of the most important technologies being developed at the moment. In this paper, the control of the multiple-stage evaporation system was addressed, as it is a crucial stage in the first- and second-generation ethanol production from sugarcane. A neural network model was proposed based on a dynamic phenomenological model developed in EMSO (Environment for Modeling, Simulation and Optimization). The phenomenological model was used to build a neural network prediction model for an MPC (Model Predictive Control) scheme using a DMC (Dynamic Matrix Control) algorithm. Simulations were carried out to evaluate the performance for tracking the set-point. Also, disturbance rejection tests were performed, considering different step disturbances. The analysis demonstrated that the MPC scheme performed well in the tests and showed superiority when compared to classical PID controllers.

With the recent development of machine learning, reinforcement learning is an interesting alternative to PID controllers. In this context, a discrete predictive Q-learning approach is applied in the control of a sugarcane biorefinery multiple-effect evaporation system. The algorithm is built using Scilab and learns to control the multiple-effect evaporator outlet concentration by manipulating its feed steam flow rate. Based on multiple episodes, the state-actions that consist of discrete changes in steam flow rate are chosen with a greedy algorithm. In order to increase the training efficiency and overcome the large dead time of the system, a neural network is applied to predict the outlet concentration of each control action after reaching the steady-state. The control policy was built and tested through simulations on a phenomenological model. The controller performance was evaluated in set-point tracking and disturbance rejection tests and compared with PID responses. The research showed that the Q-learning controller exhibited better performance than the PID controller.

The divided wall column (DWC) can achieve sharp separations of three or more components in a single shell, substituting conventional sequences of two or more binary distillation columns, with lower expenses. Despite these advantages, DWC models are not available in commercial chemical process simulators. To simulate DWC, users must employ instances of conventional column model and couple them in different configurations. In this paper, a DWC model was developed in EMSO (Environment for Modeling, Simulation and Optimization). DWC model was then used for simulating the separation of an equimolar mixture of three hydrocarbons. Results show that, depending on the number of trays, DWC presented energy savings compared to two ordinary distillation columns. Better separation was obtained when the number of divided trays was close to half the number of total trays. However, the liquid and vapor flow rates split into the divided section play a key role in the separation.

Optimal operation of integrated heat exchangers is a challenging task in the field of process control due to system nonlinearities, disturbances and adequate model identification. This paper describes the design of an advanced neural network predictive control (NNPC) applied to a heat exchanger network. A case study with two hot and one cold streams, through three counter-current heat exchangers is used to test the proposed strategy. A lumped dynamic model is built based on the concept of multi-cells topology (mixed tanks), where the hot and cold cells are connected by a wall element throughout the heat exchanger length. Each cell is assumed perfectly mixed and all physical properties are constant. A distributed behavior is achieved by increasing the number of cells. The main assumptions of the lumped model are constant temperature in each cell, heat exchanger volume and area equally distributed between cells and negligible heat loss to the environment. The predictive controller relies on a neural-based model of the plant that is used to identify the system and to predict future performance over a predefined horizon. Results were compared to a traditional controller, and the control performance was improved when compared to the Ziegler-Nichols tuning method.

Renewable diesel, or green diesel, is the biofuel with the fastest growing use in the world. Its use as a replacement for fossil diesel is important to reduce the impact of carbon emissions. Green diesel is formed by a mixture of hydrocarbons with a chemical composition similar to that of fossil fuel, making it a drop-in biofuel. Commercial processing employs hydrotreatment of vegetable oil, which, in addition to producing green diesel, can also be used to produce biofuel for aviation, bionaphtha and biopropane. The present paper proposes a model for the hydrotreatment of soybean oil for the production of green diesel. The process was simulated in Aspen Plus® v10; the decarboxylation, decarbonylation and hydrodeoxygenation reactions were carried out in a stoichiometric reactor in the presence of an NiMo/Al2O3 catalyst. In addition to the cracking reactions and fractionation, the fuels produced green diesel, aviation biofuel, and light gases. Brazilian soybean oil was used as the raw material to produce the green diesel. Sensitivity analyses were performed to adjust the best fractionation conditions, considering the composition of the product and the distillation column energy. The best results were chosen when the distillation column was configured with 22 stages, feed at stage 12 and reflux ratio equal to 0.8. The hydrogen/soybean oil ratio was 0.031. The energy used in the process was 1181 kW/h. More than 80% of the soybean oil was converted to renewable fuels, with approximately 65% to green diesel. In addition, a prediction of the simulated physical chemical properties was made and compared with the current Brazilian legislation for green diesel obtaining values within the established specifications. The ASTM D86 distillation curve was compared with simulated results from green diesel obtained from experimental data and fossil diesel. The experimental and simulated distillation curves agreed, showing only a small difference between curves up to 25% of recovered volume. This can be explained by the lighter components present in the experimental green diesel. The proposed simulation supported the potential of Brazilian soybean oil for the production of quality renewable fuels.

Rigorous black-box simulations are useful to describe complex systems. However, it cannot be directly integrated into mathematical programming models in some algebraic modeling environments because of the lack of symbolic formulation. In the present paper, a framework is proposed to embed the Aspen HYSYS process simulator in GAMS using kriging surrogate models to replace the simulator-dependent, black-box objective, and constraints functions. The approach is applied to the energy-efficient C3MR natural gas liquefaction process simulation optimization using multi-start nonlinear programming and the local solver CONOPT in GAMS. Results were compared with two other meta-heuristic approaches, Particle Swarm Optimization (PSO) and Genetic Algorithm (GA), and with the literature. In a small simulation evaluation budget of 20 times the number of decision variables, the proposed optimization approach resulted in 0.2538 kW of compression work per kg of natural gas and surpassed those of the PSO and GA and the previous literature from 2.45 to 15.3 %.

Water Distribution Networks (WDN) are systems of water distribution used in industrial processes and urban centers. The optimal WDN design can be very effective in saving energy, specifically in pumping service, to carry water to nodes of demand, at appropriated velocities and pressures. Indirectly, it can contribute in reducing liquid pollution and accidents caused by pressure overestimation in nodes. The design of WDN can be treated as an optimization problem with a Mixed Integer Nonlinear Programming (MINLP) formulation. The objective function, to be minimized is the WDN cost, given by the product of the pipe diameters and their lengths. The problem constraints are the mass balances in each node, the energy balances in the WDN loops and pressure and velocities limits. A set of commercial diameters is available, with proper costs and rugosity coefficients. The majority of paper published in this research field use external hydraulic simulators and meta-heuristic methods to solve the optimization problem. In the current paper a mathematical model using a deterministic Mathematical Programming approach is proposed and all variables are simultaneously optimized, avoiding the use of external software for pressure and velocities calculations. Two case studies were used to test the model applicability and coded in GAMS, using the global optimization solver BARON. Results showed that for both cases global optima was achieved, proving that it is possible to solve the problem, independently of external hydraulic simulator.

The synthesis of work and heat exchange networks (WHEN) has drawn much attention from process synthesis researchers in the present decade, arising as a promising field of study. The current literature has shown that simultaneously integrating heat and work may lead to substantial savings in utilities usage in highly energy-demanding processes. The problem, however, comprises highly nonlinear constraints and efficient solutions are difficult to obtain. Optimisation-based approaches have presented interesting solutions. However, in some cases, they may not be practical. When coupling several compressors and turbines via one single-shaft for work exchange, it may become technically difficult to maintain the same rotation speed. Moreover, discharge temperatures in compressors/turbines in some literature solutions may be considered impractically high/low. In order to address the first issue, the number of units that can be coupled via single-shaft was limited, and the use of multiple axes rotating at different speeds was considered. Such a new constraint is handled with a new stage coupled to a previous meta-heuristic method for solving the WHEN synthesis model. The method is able to identify optimal couplings respecting the constraints proposed for more mechanically practical designs. Regarding the discharge temperature issues, penalty functions were employed and the optimization approach was able to find solutions within practical operating range for compressors/turbines temperatures.

The heat exchanger network (HEN) synthesis problem can be characterized as non-convex, non-linear and highly combinatory. Mainly for large-scale HEN, the complexity increases even more and mathematical formulation may require very sophisticated methods. In this work, the simultaneous synthesis problem is solved by a bi-level hybrid method: in the upper level a Genetic Algorithm (GA) is used to optimized discrete variables and in the lower level, the continuous variables are optimized by Differential Evolution (DE). Two examples from the literature were used to test the applicability of the proposed solution and results found in terms of the total annual cost (TAC) were lower than those published in the literature.

Heat integration by means of Heat Exchanger Network (HEN) synthesis is fundamental in Process Engineering. There are aspects and design options that so far have been scarcely explored in this area. That is the case of heat and work integration. Recent literature shows that the simultaneous optimization of HEN with pressure recovery units can yield substantial gains. Superstructure based mathematical models for HEN synthesis are usually extended for dealing with heat and work integration formulations. Those models are complex to solve, containing nonlinearities and non-convexities that need to be overcome with simplifying assumptions (e.g., isothermal mixing after stream splitting or limiting the structural possibilities) and efficient solution approaches. An important trend in the heat integration literature is the use of meta-heuristic methods. These have arisen as an interesting alternative to the deterministic approaches. Meta-heuristics are attractive since they do not require advanced derivative-based concepts and can treat objective functions as “black-boxes”. Those strategies have been used for solving several industrial problems, including HEN synthesis. Promising solutions have been achieved in several benchmark case studies, often with formulations less simplified than those used when deterministic solvers are employed. This work aimed to fill a literature gap by applying a meta-heuristic approach to an enhanced stage-wise superstructure for HEN synthesis extended to handle streams with pressure recovery. The aforementioned superstructure-based model comprises more options than usually found in the literature, such as utilities allocation prior to units that perform heat exchange between two process streams. For achieving its goal, the model entailed several new variables related to pressure recovery, which had to be included to the meta-heuristic scheme. In order to test the methodology, it was applied to a case study taken from the literature. The strategy here used was able to outperform the solution previously reported in the literature, demonstrating that meta-heuristics can be efficiently applied to a heat/work integration model derived from a superstructure more complex than that employed in previous investigations.

Synthesis of heat exchanger network (HEN) is a subject of great industrial interest. In an industrial process, it is possible to take advantage of the existing energy in its own process streams and, consequently, to reduce the utilities consumption. Operability considerations are important for the HEN design due to the uncertainties in streams temperatures and flow rates, which are very common in real operations. Therefore, the HEN design, in addition to economic interests, must consider flexibility, controllability and resiliency aspects. This paper performs the analysis of controllability and resiliency (C&R). For the configuration of the HEN control system, it is used the Relative Gain Array and Disturbance Cost index. This analysis of operability is applied in a literature example. A modification in the literature HEN is then suggested to improve network resiliency.

Heat Exchanger Networks (HEN) are able to reduce the operating costs in an industrial plant. Cost-optimal HEN generally leads also to a substantial reduction in pollutant gases emissions. However, HEN synthesis approaches are based generally on the optimization of total annual costs (TAC) only. In that manner, this work aims to address environmental impacts (EI) more directly. The EI were converted into an objective function by using a Life Cycle Assessment (LCA) methodology. With two objective functions (EI and TAC), a multi-objective optimization (MOO) may be performed. A meta-heuristic approach is adapted to handle MOO in the formulation proposed. As expected, a conflictive behaviour was noticed. The approach proposed was able to yield Pareto fronts efficiently, providing good intermediate solutions that could be used by decision-makers in search of a trade-off HEN configuration which is able to present low costs and is also environmentally friendly. The results reported in the case study also demonstrate the method reliability with intermediate solutions that are much more environmentally friendly than the cost optimal configurations, presenting costs only marginally higher.

Abstract preview not available - see full-text PDF article.