• Energy Research

A multi-agent system consists of several computational entities capable of autonomous actions, called agents, which communicate with each other, and have the ability to coordinate their actions and to cooperate. Multi-agent systems received a great interest and attention over time, as they can be seen as a key enabling technology for complex applications, where distributed and processing of data, autonomy, and high degree of interactions in dynamic environments are required at the same time. Therefore, in view of current and future developments of the digitalization of industrial production cycles promoted by Industry 4.0, multi-agent systems are foreseen to play an increasing role for industrial production management and optimization. Because of barriers represented by large presence of legacy systems, in the steel sector agent-based technology is not widely applied yet, and multi-agent systems applications are very few. On the other hand, steel manufacturing industries are complex and dynamic systems whose production processes held a strategic role in the global economy. During last decades, the steel sector has undergone relevant transformations, especially through the massive digitalization and the innovation introduced by Industry 4.0. A further evolution is foreseen in the incoming years to improve the sustainability of the production cycle by improving energy and resource efficiency. Therefore, steel industries must face several challenges on the path toward the factory of the future. In such context multi-agent systems, through their intrinsic properties, such as autonomy, social abilities, reactivity, proactivity, and mobility, can overcome existing drawbacks and barriers, by increasing flexibility, improving resources efficiency, handling production operations, reacting to unpredicted events, optimizing production processes, and supporting legacy systems. In this paper, some applications of multi-agent systems in steel sector are presented to show the advantages and opportunities of agent-based technology.

AbstractThis article presents the application of a recent neural network topology known as the deep echo state network to the prediction and modeling of strongly nonlinear systems typical of the process industry. The article analyzes the results by introducing a comparison with one of the most common and efficient topologies, the long short-term memories, in order to highlight the strengths and weaknesses of a reservoir computing approach compared to one currently considered as a standard of recurrent neural network. As benchmark application, two specific processes common in the integrated steelworks are selected, with the purpose of forecasting the future energy exchanges and transformations. The procedures of training, validation and test are based on data analysis, outlier detection and reconciliation and variable selection starting from real field industrial data. The analysis of results shows the effectiveness of deep echo state networks and their strong forecasting capabilities with respect to standard recurrent methodologies both in terms of training procedures and accuracy.

To achieve the greenhouse gas reduction targets formulated in the European Green Deal, energy- and resource-intensive industries such as the steel industry will have to adapt or convert their production. In the long term, new technologies are promising. However, carbon capture storage and utilization solutions could be considered as short-term retrofitting solutions for existing steelworks. In this context, this paper presents a first experimental demonstration of an approach to the utilization of process off-gases generated in a steelworks by producing methane and methanol in hydrogen-intensified syntheses. Specifically, the integration of two methane synthesis reactors and one methanol synthesis reactor into a steel plant is experimentally simulated. An innovative monitoring and control tool, namely, a dispatch controller, simulates the process off-gas production using a digital twin of the steel plant and optimizes its distribution to existing and new consumers. The operating states/modes of the three reactors resulting from the optimization problem to be solved by the dispatch controller are distributed in real time via an online OPC UA connection to the corresponding experimental plants or their operators and applied there in a decentralized manner. The live coupling test showed that operating values for the different systems can be distributed in parallel from the dispatch controller to the test rigs via the established communication structure without loss. The calculation of a suitable control strategy is performed with a time resolution of one minute, taking into account the three reactors and the relevant steelworks components. Two of each of the methane/methanol synthesis reactors were operated error-free at one time for 10 and 7 h, respectively, with datasets provided by the dispatch controller. All three reactor systems were able to react quickly and stably to dynamic changes in the load or feed gas composition. Consistently high conversions and yields were achieved with low by-product formation.

Abstract The paper proposes two different approaches for steam turbine modelling for on-line monitoring applications, a hybrid-thermodynamic method and a neural network approach. Both models can predict power and other features that cannot be easily measured such as outlet Steam Quality, Pressure and Temperature at drums outlet. Training and validation of both models was carried out by exploiting a dataset created by means the GE sizing design tool. The models were tested by means of real field data of a High Pressure Turbine.

The valorization of integrated steelworks process off-gases as feedstock for synthesizing methane and methanol is in line with European Green Deal challenges. However, this target can be generally achieved only through process off-gases enrichment with hydrogen and use of cutting-edge syntheses reactors coupled to advanced control systems. These aspects are addressed in the RFCS project i3upgrade and the central role of hydrogen was evident from the first stages of the project. First stationary scenario analyses showed that the required hydrogen amount is significant and existing renewable hydrogen production technologies are not ready to satisfy the demand in an economic perspective. The poor availability of low-cost green hydrogen as one of the main barriers for producing methane and methanol from process off-gases is further highlighted in the application of an ad-hoc developed dispatch controller for managing hydrogen intensified syntheses in integrated steelworks. The dispatch controller considers both economic and environmental impacts in the cost function and, although significant environmental benefits are obtainable by exploiting process off-gases in the syntheses, the current hydrogen costs highly affect the dispatch controller decisions. This underlines the need for big scale green hydrogen production processes and dedicated green markets for hydrogen-intensive industries, which would ensure easy access to this fundamental gas paving the way for a C-lean and more sustainable steel production.

Abstract The paper proposes a methodology for modeling of energy transformation equipment which are commonly found in integrated steelworks, mainly focusing on steam production in the Basic Oxygen Furnace and auxiliary boilers, the electric power production in off-gas expansion turbines and some relevant steam and electricity consumers. The modeling approach is based on standard neural networks and Echo State Networks (ESN) for forecasting the variables of interest. All the models are intended as processes predictors to be used in a hierarchical control strategy based on multi-period and multi-objective optimization techniques and model predictive control. The overall target is the optimization of the re-use of off-gas produced in integrated steelworks by minimizing costs and maximizing revenues. Training and validation of models have been carried out by exploiting real historical data provided by steelmaking companies and have been successful tested.

This work investigates the cost-efficient integration of renewable hydrogen into steelworks for the production of methane and methanol as an efficient way to decarbonize the steel industry. Three case studies that utilize a mixture of steelworks off-gases (blast furnace gas, coke oven gas, and basic oxygen furnace gas), which differ on the amount of used off-gases as well as on the end product (methane and/or methanol), are analyzed and evaluated in terms of their economic performance. The most influential cost factors are identified and sensitivity analyses are conducted for different operating and economic parameters. Renewable hydrogen produced by PEM electrolysis is the most expensive component in this scheme and responsible for over 80% of the total costs. Progress in the hydrogen economy (lower electrolyzer capital costs, improved electrolyzer efficiency, and lower electricity prices) is necessary to establish this technology in the future.

Abstract The efficient use of resources is a relevant research topic for integrated steelworks. Process off-gases, such as the ones produced during blast furnace operation, are valid substitutes of natural gas, as they are sources of a considerable amount of energy. Currently they are recovered but sometimes part of such gas is flared due to non-optimal management of such resource. In order to exploit the off-gases produced in an integrated steelworks, the interactions between gas producers and users in the whole gas network need to be considered. The paper describes a model exploited by a Decision Support Tool that is under development within a European project. Such model forecasts the blast furnace gas amount and its heating power by obtaining an error between 3 and 7 % in a time horizon of 2 hours. The forecasted values of blast furnace gas allow a continuous optimal planning of the blast furnace gas usage according to its availability and to the needs in the steelworks, by avoiding losses of a valuable secondary resource and related emissions.