Petr Novák

Applications of ontologies for assembling simulation models of industrial systems

Simulation models are important parts of industrial system analysis and control system design. Despite the wide range of possible usage, their formalization, integration and design have not been satisfactorily solved. This paper contributes to the design phase of simulation models for large-scale industrial systems, consisting of a large number of heterogeneous components. Nowadays, it is the simulation expert who assembles the simulation model for a particular industrial plant manually, which is time-consuming and error-prone. We propose to use a semantic engine performing SPARQL queries, which assembles the structure of a simulation model semi-automatically, using formalized knowledge about real industrial plant and available simulation blocks represented in appropriate ontologies. As each real plant device can be represented by more than one simulation blocks, the selection of suitable simulation candidates is based on matching interfaces of neighboring blocks. Evaluation on a real-life industrial use-case shows improvements in reducing development time and avoiding design errors. Major results of this paper are the proposed structures of the ontologies and the SPARQL query realizing the selection of the appropriate simulation blocks.

Integration framework for simulations and SCADA systems

Abstract Stricter requirements on the quality of industrial plant operation together with environmental limits and decreasing energy consumption bring more complex automation systems. The intelligent control techniques, which are based on approaches from diverse disciplines including statistics, artificial intelligence or signal processing, have been widely used during the last years and their benefits have been proved. They cannot be developed and tested without simulation models and access to online and historical data. This article proposes a platform for the integration of simulations and industrial SCADA systems supporting complex data access and simulation code re-use. The idea of the presented framework is to connect simulations, data sources, optimizers, other calculations and SCADA systems into one integrated environment seamlessly. A technical level of the framework provides integration of stakeholders and a semantic level captures engineering knowledge in inter-mapped ontologies and configures the technical level, which is often called model-driven configuration. The semantic level utilizes a formal model implemented as set of ontologies. The major contribution of the article are the layered model of the integration architecture and formulation of the integration requirements in the industrial automation domain. The proposed solution has been implemented and tested on a software prototype level. It is demonstrated on two use-cases covering both design and integration of simulation models from the industrial perspective. The proposed architecture is intended to be as general as possible, however it has been tested on signal-oriented simulators only. It is the main limitation of this contribution and it should be addressed in upcoming work.

Integrating heterogeneous engineering knowledge and tools for efficient industrial simulation model support

The enhanced knowledge management improves the simulation design and integration.The article proposes a data model supporting simulation model engineering.An infrastructure called the Engineering Service Bus improves simulation processes.The approach is evaluated on a hydraulic system use-case showing its efficiency. Design of simulation models and their integration into industrial automation systems require knowledge from several heterogeneous data sources and tools. Due to the heterogeneity of engineering data, the integration of the tools and data are time-consuming and error-prone tasks nowadays. The key goal of this article is to provide an effective and efficient integration of heterogeneous data sources and tools as a knowledge basis to support dynamic simulation for industrial plants. The integrated knowledge is utilized both (i) in the design phase of simulation models for defining structure and interfaces of the models and (ii) in the runtime phase of industrial systems for model-driven configuration of the integrated environment. Reaching such goals with a manual approach or point-to-point integration is not beneficial as it may be possible for a few tools and data sources, but quickly gets very complex. A growing number of elements increases the risk of errors and the effort needed for integration. The proposed solution is based on a specification of a common data model to represent engineering knowledge and a service-oriented tool integration with the Engineering Service Bus. Engineering knowledge is integrated in a knowledge base implemented with ontologies in Web Ontology Language-Description Logic (OWL-DL). The proposed approach is demonstrated and evaluated on an educational hydraulic system. Major results of the article are: (i) a data model to represent engineering knowledge for dynamic industrial systems, (ii) the integration platform that, based on this model, integrates the tools for system design and runtime, and (iii) basic design-time and runtime processes for the integrated industrial simulations.

Framework for Simulation Integration

Abstract This paper presents an idea of integration of simulations into the control system architecture. The benefits of Advanced Process Control (APC) has been clearly shown in the last years. Advanced control methods rely on quality models and simulations. These models cannot be developed and operated without access to online and historical data. Integration of simulators, data sources, optimizers and other tools is discussed in this paper. The whole problem is divided into two main phases and the first one is presented here. This phase is design oriented but it can be used for solving specific tasks. We illustrate the idea on the real world example.

Semantic design and integration of simulation models in the industrial automation area

Simulations are software tools approximating and predicting the behavior of real industrial plants. Unlike real plants, the utilization of simulations cannot cause damages and it saves time and costs during series of experiments. A shortcoming of current simulation models is the complicated runtime integration into legacy industrial systems and platforms, as well as ad-hoc design phase, introducing manual and error-prone work. This paper contributes to improve the efficiency of simulation model design and integration. It utilizes a semantic knowledge base, implemented by ontologies and their mappings. The integration uses the Automation Service Bus and the paper explains how to configure the runtime integration level semantically. The main contributions are the concept of semantic configuration of the service bus and the workflows of simulation design and integration.

Design and verification of simulation models of passive houses

Passive houses are featured with very low energy consumption for heating and operation, which is reached not only by excellent insulation, but also by proper control algorithms. Optimization of passive houses must be done both by civil and control engineers; furthermore, these disciplines must collaborate. For the optimization, simulation models are very useful test-beds. This paper presents the Building Simulation Library (Bldsimlib) and its mathematical-physical background. The universal library blocks simulate temperature, carbon dioxide concentration, relative humidity, and air pressure. The paper compares preliminary really measured data from an experimental passive house and simulation results. Finally, an ontology-based methodology for semi-automated design of simulation models is proposed and evaluated.

Performance Modeling Extension of Directory Facilitator for Enhancing Communication in FIPA-Compliant Multiagent Systems

The growing adoption of highly distributed and intelligent industrial systems frequently denoted as Industry 4.0 or Industrial Internet of Things brings great opportunities as well as new research questions because connection of single devices into a cooperating community creates much larger and complex systems with new challenges regarding verification and security. This paper proposes a method to solve problems related to setup of timeouts used in communication protocols. The static setup is replaced with an enhancement of directory facilitator that in cooperation with a specialized sniffer agent continuously monitors the system load and eventually postpones starts of new tasks. This precaution eliminates saturation of computational resources and preserves the maximal reachable system performance. The proposed method is demonstrated through a simulation of a chilled water system using a test-bed of six minicomputers Raspberry Pi that host agents implemented in Jade.

Engineering of Coupled Simulation Models for Mechatronic Systems

Simulation models play a crucial role in testing and fine-tuning of control systems for complex industrial systems. They are important parts of the sustainable, service-oriented manufacturing value chain by facilitating early and efficient defect detection and risk mitigation. However, the design phase of simulation models is time-consuming and error-prone, thus it should be improved to become more efficient. The design of simulation models for mechatronic systems has to cope with two basic challenges: (1) the heterogeneous nature of mechatronic systems, which are described with various overlapping engineering plans, and (2) the separation of monolithic simulations into distributed simulation modules, to better conquer the computational complexity of simulation models. This paper addresses both challenges: (a) we propose an application of semantic integration and linked data for sharing and capturing knowledge for simulation model design between various engineering plans; (b) we explain how to structurally connect simulation modules, which are dynamically coupled. The proposed method utilizes the extended bond-graph theory. The simulation modules work independently of each other but with interaction, similar to a multi-agent system. Since the computational execution of coupled simulations is a crucial obstacle especially for mechatronic systems, we show how performance analysis can significantly improve the definition of simulation workflows.

Component-Based Design of Simulation Models Utilizing Bond-Graph Theory

Abstract Simulation models are becoming efficient tools for real plant testing and control system fine-tuning. They are necessary for advanced process control. Since a design phase of simulation models is time-consuming and error-prone, this paper proposes a method for the systematic and semi-automated design of simulation models. It is intended mainly for signal-oriented simulators such as MATLAB-Simulink. The method presumes that large-scale industrial systems consist of subsystems, whose behavior is known. The task for the proposed approach is to create a simulation model from atomic simulation blocks approximating system components. The proposed solution utilizes the bond graph theory in a non-traditional way. Compared to the classical use of bond graphs, the proposed method does not create simulations from scratch, but it re-uses existing simulation components. It extends bond graphs with a concept of component implementations and annotations of their interfaces. Thus, models are assembled from atomic component implementations, whose selection is done by the proposed method for each node of the system topology.

Extraction of Automation System Engineering Knowledge for Mapping Plant and Simulation Interfaces

Design and integration of industrial automation systems require cooperation of different engineering disciplines and various engineering tools during both design-time and runtime. The scope of this chapter is to bridge the gap between the design-time description and the runtime integration in a semi-automated way. Furthermore, this chapter focuses on the problem of the integration of legacy systems with limited access to original engineering data. It is assumed that an interface between a real industrial system and software automation tools is OPC Unified Architecture (OPC UA) and we propose to derive the plant knowledge from the OPC UA tag list. Since it is not possible to solve this task in general, we require the tag list to adopt the naming convention defined by the international standard IEC 81346, which is widespread in large-scale systems. This tag list is parsed and the engineering knowledge is derived and stored in the Engineering Knowledge Base. Consequently, the plant knowledge is mapped to knowledge related to software automation tools, such as simulations. The proposed methodology is evaluated on a real-life example dealing with a laboratory tank model.