Synthesis of Environmentally-Benign Energy Self-Sufficient Processes under Uncertainty
Annamaria Vujanovića, Lidija Čučekb, Zorka Novak Pintaričb, Bojan Pahora, Zdravko Kravanjab,*
aPerutnina Ptuj d.d., Potrčeva cesta 10, 2250 Ptuj, Slovenia
bFaculty of Chemistry and Chemical Engineering, University of Maribor, Smetanova ulica 17, 2000 Maribor, Slovenia
Published in Journal of Cleaner Production (2014), doi: 10.1016/j.jclepro.2014.04.015
Summary
This contribution presents a multi-objective Mixed-Integer Linear Programming (MILP) synthesis of a dynamic supply-network under uncertainty applied to food industry. The previously-developed multi-objective model for achieving energy self-sufficiency by integrating renewables into companies’ supply-networks has now been extended to account for the dynamic consideration of variable supply and demand over the year, for uncertainties related to products’ demand and sun radiation, and for multi-objective optimisation, in order to obtain the most sustainable company’s supply-network. The sustainable synthesis of a company’s network is performed regarding the integration of renewables such as biomass and other wastes, and solar energy. The obtained solutions are those reflecting maximal profit, reflecting constantly-changing dynamic market conditions, accounting for several uncertain parameters, and protecting the environment.
Background
Nowadays more and more companies are developing technologies for using clean and renewable energy resources when improving their energy efficiencies and cost, whilst significantly decreasing environmental burdens. Since the main intention of companies is to maximise profit on the one hand, whilst decreasing environmental burdens on the other hand, this leads to multi-objective optimisation (MOO) problems.
Planning of resources supply is of major importance when utilising alternative energy sources. Renewable energy sources are limited locally and the quantities available differ over time. In addition, industrial plants encounter uncertainties during every operational step, which means that plants need to be flexible in order to adjust their operations to the variations of uncertain parameters. Certain fluctuations that need to be taken into account arise during the production of any product. Fluctuations are usually related to the process (fluctuations of flows, temperatures, compositions…) or those fluctuations due to external factors (demand for products, prices of raw materials, products, and utilities). Therefore it is important to consider dynamics of supply and demand during supply chain syntheses.
Aims
The emphasis of this contribution relates to achieving the environmentally-benign energy self-sufficiency integration of renewables into companies’ supply networks by considering the dynamic behaviour of the supply-network, and different uncertainties regarding products’ supply and demand. Performing an extended synthesis in order to account for dynamic behaviour, whilst also considering different uncertainties within the supply-network, thus enables us to bring the synthesis closer to the realistic situation of the market. Schematic representation of dynamic supply-network is shown in Figure 1.
Figure 1: Schematic representation of supply-network (modified from Kiraly et al., 2013a)
The main task of this work is to maximise the profit, whilst minimising the environmental burdens by utilising renewable sources. In order to accomplish this task, the following goals within an integrated methodology are proposed:
· Dynamic synthesis considering supply and demand variations over twelve time periods of the year. Synthesis is performed by a multi-period MILP model.
· Flexible dynamic synthesis upgrading the dynamic synthesis with accounting for uncertainties relating to product supply and demand and the number of sunlight hours. For performing the flexible dynamic synthesis, Monte Carlo method is being used with triangular distribution for input data.
· Multi-objective dynamic synthesis extending the dynamic synthesis with accounting for several environmental footprints relating to energy consumption and production activities. Those footprints are carbon (CF), nitrogen (NF), water (WF), and energy footprints (EF), and are defined as additional objectives besides economic objective. In order to perform the multi-objective dynamic synthesis, two-step approach and ε-constraint method are being used.
Methods
The previously developed model for the synthesis of companies’ renewable biomass and energy supply-networks (Kiraly et al., 2013a) has been first upgraded to a dynamic synthesis model formulated as a multi-period one in order to take into account market, seasonal, and other changes (e.g. van den Heever and Grossmann, 1999). The dynamic synthesis model was then further extended into a flexible dynamic one, and to a multi-objective dynamic synthesis model.
The mathematical model is formulated in a MILP form. It consists of mass and energy balances, production and conversion constraints, logical capacity constraints, cost functions, economic objective function, and environmental footprints evaluation. Environmental footprints are evaluated within the framework of direct (burdening) and indirect (unburdening) effects, combined forming total effects on the environment (Kravanja, 2012). Footprints are evaluated only from the activities contributing significantly to energy consumption and/or production (Vujanović et al., 2014): i) Biogas production; ii) Combined heat and power generation from CHP unit; iii) Photovoltaic-based electricity production; and iv) Energy consumption within transportation. For performing the dynamic synthesis under uncertainty, Monte Carlo method is being used.
As the model consists of data-independent equality and inequality constraints, it can be applied straightforwardly to different cases including the integration of renewables into a company’s supply-network, only by providing the necessary data.
Results
Several syntheses of industrial supply-networks were applied to an existing large-scale meat production company: i) dynamic synthesis, ii) flexible dynamic synthesis and iii) multi-objective dynamic synthesis. Several renewable energy sources are considered in order to maximise the self-sufficiency of the company’s energy supply, such as animal and organic wastes for biogas production, and solar energy for producing electricity through photovoltaic (PV) panels. Figure 2 presents the company’s monthly electricity self-sufficiency, and Figure 3 presents the results from multi-objective dynamic synthesis. It shows relative profit vs. relative total CF, WF and EF.
Figure 2: Electricity self-sufficiency (%) over the months (after Kiraly et al., 2013b)
Figure 3: Relative profit vs. relative total footprints by performing MOO (after Kiraly et al., 2013b)
Developed general mathematical model enables obtaining the efficient dynamic multi-objective solutions and solutions under uncertain conditions. In regards to the obtained results it was shown that energy self-sufficiency was especially outweighed in summer months due to photovoltaic electricity production, but it was higher as consumption also in winter months. Also, it can be seen that CF and NF have significant unburdening effect; however, on the other hand, WF has significant burdening effect, particularly due to biogas production. The obtained solutions are environmentally more benign except from WF’s viewpoint, implementing renewable energy production technologies.
Acknowledgments
The authors are grateful for the financial support from theEuropean Social Fund, theSlovenian Technology Agency – TIA (PhD research fellowship contract No P-MR-10/89), theSlovenian Research Agency – ARRS (programs P2-0032 and P2-0377), and from the EC FP7 project ENER/FP7/296003/EFENIS ‘Efficient Energy Integrated Solutions for Manufacturing Industries – EFENIS’.
References
Kiraly, A., Pahor, B., Kravanja, Z., 2013a, Achieving energy self-sufficiency by integrating renewables into companies' supply networks, Energy 55, 46-57.
Kiraly, A., Pahor, B., Čuček, L., Kravanja, Z., 2013b. Dynamic Multi-Objective Synthesis of Companies’ Renewable Biomass and Energy Supply-Network. Chemical Engineering Transactions 35, 73-78
Kravanja, Z., 2012. Process systems engineering as an integral part of global systems engineering by virtue of its energy — environmental nexus. Current Opinion in Chemical Engineering 1, 231-237.
van den Heever, S.A., Grossmann, I.E., 1999. Disjunctive multiperiod optimization methods for design and planning of chemical process systems. Computers & Chemical Engineering 23, 1075-1095.
Vujanović, A., Čuček, L., Pahor, B., Kravanja, Z., 2014. Multi-Objective Synthesis of a Company's Supply-Network by Accounting for Several Environmental Footprints. Process Safety and Environmental Protection, 92(5), 456-466.
Suggested citation: Vujanović, A., Čuček, L., Novak Pintarič, Z., Pahor, B., & Kravanja, Z. (2014). Synthesis of environmentally-benign energy self-sufficient processes under uncertainty. Journal of Cleaner Production, doi: 10.1016/j.jclepro.2014.1004.1015.