The aim of the paper is to present the application of the evolutionary algorithms to selected optimization and identification problems of mechanical systems. The coupling of evolutionary algorithms with the finite element method and the boundary element method creates a new artificial intelligence technique that is very suitable in computer aided optimal design and defect detection. Several numerical examples for optimization and identification are presented.
Keywords: evolutionary algorithms, finite element method, boundary element method, optimization, identification.

The aim of the paper is to point out the most important factors that should be taken into account during the designing of expert systems for technical diagnostics and advanced condition monitoring. The first such factor is the proper design of unified databases. It was assumed that discussed system consists of a network of coexisting and related nodes containing active statements looking for an equilibrium state. Such network represents a diagnostic model. Diagnostic models describe the relations between observed symptoms and their causes, i.e. the technical states of the object. Direct specification of such models is difficult due to the complex nature of state-symptom relations. An interesting idea is connected with example based inverse diagnostic models. Suggested solutions simplify the development and reduce maintenance costs for the whole system. A very important benefit for industrial application is the opportunity to arrange an incremental development of the final diagnostic expert system.
Keywords: expert systems, blackboard, reasoning strategy, inverse models.

The methods of nuclear power plant safety assessment and accident management used in Sweden and selected computerized decision support systems are briefly described. The defense-in-depth strategy, which comprises three elements, namely prevention, protection and mitigation, is essential for keeping the fission product barriers intact. The safety assessment program, which focuses on prevention of incidents and accidents, has three main components: periodic safety reviews, probabilistic safety analysis and analysis of postulated disturbances and accident progression sequences. Even if prevention of accidents is the first priority, it is recognized that accidents involving severe core damage, including core melt, may nevertheless occur. Therefore, measures are required to achieve reasonable capability to manage such accidents. Emergency Operating Procedures and Severe Accident Management Guidelines are the vital components of the Accident Management System. Computerized decision support systems, to be used during normal, disturbed and accident states of a plant are expected to play increasingly important role in safety assessment and accident management, including support in rapid evaluation of possible radioactive releases in the event of a severe accident.

The paper focuses on using of artificial neural networks in model-based fault detection and isolation. Modelling of a system both at its normal operation conditions and in faulty states is considered and a comparative study of three different methods of system modelling that use a linear model, neural network nonlinear autoregressive with exogenous input model, and neural network Wiener model is presented. Application of these models is illustrated with an example of approximation of a dependence of the juice steam pressure in the stage two on the juice steam pressures in the stages one and three of a five stage sugar evaporator. Parameters of the linear model are estimated with the recursive pseudolinear regression method, whilst the backpropagation and truncated backpropagation through time algorithms are employed for training the neural network models. All the considered models are derived based on the experimental data recorded at the Lublin Sugar Factory.
Keywords: fault detection and isolation, neural network models, parametric models, evaporation stations.

The paper deals with selected problems of knowledge acquisition for intelligent information systems that may be applied for aiding technical diagnostics of machinery and equipment. Two main kinds of knowledge are discussed, i.e. declarative and procedural knowledge. Some methods of declarative knowledge acquisition from domain experts and from databases are presented, the latter being divided into machine learning methods and knowledge discovery ones. Examples of declarative knowledge acquisition and discovery from databases are shown. Moreover, an example of procedural knowledge acquisition from a domain expert is presented. The paper concludes with new issues of knowledge acquisition methodology.
Keywords: intelligent information systems, knowledge base, procedural knowledge, declarative knowledge, knowledge acquisition, knowledge discovery.

In this study, we develop an idea of knowledge elicitation realized over a collection of databases. The essence of such elicitation deals with a determination of common structure in databases. Depending upon a way in which databases are accessible abd can collaborate, we distinguish between a vertical and horizontal collaboration. In the first case, the databases deal with objects defined in the same attribute (feature) space. The horizontal collaboration takes place when dealing with the same objects but being defined in different attribute spaces and therefore forming separate databases. We develop a new clustering architecture supporting the mechanisms of collaboration. It is based on a standard FCM (Fuzzy C-Means) method. When it comes to the horizontal collaboration, the clustering algorithms interact by exchanging information about local partition matrices. In this sense, the required communication links are established at the level of information granules (more specifically, fuzzy sets forming the partition matrices) rather than patterns directly available in the databases. We discuss how this form of collaboration helps meet requirements of data confidentiality. In case of the horizontal collaboration, the method operates at the level of the prototypes formed for each individual database. Numeric examples are used to illustrate the method.
Keywords: fuzzy clustering, collaboration, data confidentiality and security, data interaction, cluster (partition) interaction, vertical (data-based) and horizontal (feature-based) collaboration.

The paper presents a concept of knowledge based software supporting a long period machine design analysis - intelligent personal assistant. The concept of intelligent personal assistant is based on the maze model and optimisation model of the design process. The functional structure of the whole system is shown.

The paper discusses some methods of commonsense reasoning applicable to analysis of truss structures. The proposed method, based on qualitative representation of trusses, allows to reach conclusions in the case of highly incomplete knowledge about the system. Two cases are considered. First, when only the general geometry of the structure is known, without the quantitative knowledge of stiffness coefficients of bars. Second, with additional assumption that all stiffness coefficients of bars are roughly equal.

The paper tries to show the role that can be played by genetic optimization strategies in solving huge global optimization problems in computational mechanics and other branches of high technology. Genetic algorithms are especially recommended as the first phase in two-phase stochastic optimization. The self-adaptability of genetic search is shown on the basis of the mathematical model introduced by M. Vose. Main goals of adaptation are used as leading criteria in the simple taxonomy of genetic strategies.
Keywords: genetic algorithms, stochastic search, two-phase strategies.

This paper presents possibility of identification of loads in mechanical systems based on response measurements during operation. Information of loads is very useful in diagnostic process, if usage of structure is under investigation. State of the art in the field of loads identification is presented. The problem of load identification is defined, and some methods are presented. The paper is focused on the problem of loads identification based on measurements of process parameters or movement parameters for vehicles or airplanes. Two methods are applied to show applicability of this approach in industrial practice; neural network based method and regression model based method. A case study of identification of load of helicopter structure during flight is presented.
Keywords: loads identification, neural networks application for loads identification, helicopter loads identification, regression model for load identification.