Argiris J. Dentsoras

MODELING OF SOLAR DOMESTIC WATER HEATING SYSTEMS USING ARTIFICIAL NEURAL NETWORKS

Artificial Neural Networks (ANN) are widely accepted as a technology offering an alternative way to tackle complex and ill-defined problems. They can be trained to predict results from examples, are fault tolerant, are able to deal with non-linear problems, and once trained can perform prediction at high speed. ANNs have been used in diverse applications and they have shown to be particularly useful in system modeling and for system identification. The objective of this work was to train an ANN to learn to predict the useful energy extracted and the temperature rise in the stored water of solar domestic water heating (SDHW) systems with the minimum of input data. An ANN has been trained based on 30 known cases of systems, 22 varying from collector areas between 1.81 m and 4.38 m . Open and closed systems have been considered both with horizontal and vertical storage tanks. In addition to the above, an attempt was made to consider a large variety of weather conditions. In this way the network was trained to accept and handle a number of unusual cases. The data presented as input were the collector area, storage tank heat loss coefficient (U-value), tank type, storage volume, type of system, and ten readings from real experiments of total daily solar radiation, mean ambient air temperature, and the water temperature in the storage tank at the beginning of a day. The network output is the useful energy extracted from the system and the temperature rise in the stored water. The 2 statistical R -value obtained for the training data set was equal to 0.9722 and 0.9751 for the two output parameters respectively. Unknown data were subsequently used to investigate the accuracy of prediction. These include systems considered for the training of the network at different weather conditions and completely unknown systems. Predictions within 7.1% and 9.7% were obtained respectively. These results indicate that the proposed method can successfully be used for the estimation of the useful energy extracted from the system and the temperature rise in the stored water. The advantages of this approach compared to the conventional algorithmic methods are the speed, the simplicity, and the capacity of the network to learn from examples. This is done by embedding experiential knowledge in the network. Additionally, actual weather data have been used for the training of the network, which leads to more realistic results as compared to other modeling programs, which rely on TMY data that are not necessarily similar to the actual environment in which a system operates.

Artificial neural networks used for the performance prediction of a thermosiphon solar water heater

Artificial Neural Networks (ANN) are widely accepted as a technology offering an alternative way to tackle complex and ill-defined problems. They can learn from examples, are fault tolerant, are able to deal with non-linear problems, and once trained can perform prediction at high speed. ANNs have been used in diverse applications and they have shown to be particularly effective in system modelling as well as for system identification. The objective of this work is to train an artificial neural network (ANN) to learn to predict the performance of a thermosiphon solar domestic water heating system. This performance is measured in terms of the useful energy extracted and of the stored water temperature rise. An ANN has been trained using performance data for four types of systems, all employing the same collector panel under varying weather conditions. In this way the network was trained to accept and handle a number of unusual cases. The data presented as input were, the storage tank heat loss coefficient (U-value), the type of system (open or closed), the storage volume, and a total of fifty-four readings from real experiments of total daily solar radiation, total daily diffuse radiation, ambient air temperature, and the water temperature in storage tank at the beginning of the day. The network output is the useful energy extracted from the system and the water temperature rise. The statistical coefficient of multiple determination (R2-value) obtained for the training data set was equal to 0.9914 and 0.9808 for the two output parameters respectively. Both values are satisfactory because the closer R2-value is to unity the better is the mapping. Unknown data for all four systems were subsequently used to investigate the accuracy of prediction. These include performance data for the systems considered for the training of the network at different weather conditions. Predictions with maximum deviations of 1 MJ and 2.2°C were obtained respectively. Random data were also used both with the performance equations obtained from the experimental measurements and with the artificial neural network to predict the above two parameters. The predicted values thus obtained were very comparable. These results indicate that the proposed method can successfully be used for the estimation of the performance of the particular thermosiphon system at any of the different types of configuration used here. The greatest advantage of the present model is the capacity of the network to learn from examples and thus gradually improve its performance. This is done by embedding experimental knowledge in the network.

Soft computing in engineering design - A review

The present paper surveys the application of soft computing (SC) techniques in engineering design. Within this context, fuzzy logic (FL), genetic algorithms (GA) and artificial neural networks (ANN), as well as their fusion are reviewed in order to examine the capability of soft computing methods and techniques to effectively address various hard-to-solve design tasks and issues. Both these tasks and issues are studied in the first part of the paper accompanied by references to some results extracted from a survey performed for in some industrial enterprises. The second part of the paper makes an extensive review of the literature regarding the application of soft computing (SC) techniques in engineering design. Although this review cannot be collectively exhaustive, it may be considered as a valuable guide for researchers who are interested in the domain of engineering design and wish to explore the opportunities offered by fuzzy logic, artificial neural networks and genetic algorithms for further improvement of both the design outcome and the design process itself. An arithmetic method is used in order to evaluate the review results, to locate the research areas where SC has already given considerable results and to reveal new research opportunities.

Integration of fuzzy logic, genetic algorithms and neural networks in collaborative parametric design

Abstract The solution to a design problem is extracted through the exploitation of the design knowledge in the context of a space of solution alternatives. The design process incorporates a series of decision making and knowledge management issues, which should be often addressed through collaboration among diverse stakeholders. The alternative solutions must usually be shaped under different formalisms and evaluated against commonly accepted objective criteria. The current paper presents an approach that integrates soft-computing techniques in order to facilitate the computer-aided collaboration among designers. CopDeSC (Collaborative parametric Design with Soft-Computing) is the name of the system developed in order to implement this approach whose key features are: (a) the collaborative structuring of design parameter hierarchies, (b) the modeling of the design objectives through fuzzy preferences stated by the designers on certain design parameters, (c) the deployment of genetic algorithms for locating the optimum solution and (d) the utilization of records of elite solutions that are submitted in a neuro-fuzzy approximation in order to produce a simplified problem formulation, suitable for addressing redesign tasks in significantly less computational time. The efficiency of CopDeSC is evaluated in an example case of the parametric design of oscillating conveyor that has been conducted by a group of designers.

Case-DeSC: A system for case-based design with soft computing techniques

Abstract The present paper introduces a case-based design with soft computing (Case-DeSC) system that uses soft computing techniques for addressing parametric design problems. Design case representation relies on digraphs of design parameters, supported by fuzzy preferences on specific parameters’ values and weighting factors, which capture the parameters’ relative importance. The final design solution is either extracted via a genetic algorithm that searches for the solution with the maximum aggregated preference, or it is retrieved by a competitive neural network. This neural network utilizes the medium of the maximum or the centroid of the assigned fuzzy preferences as similarity measures and it is trained by utilizing the available cases in the case base. Several functionalities are incorporated to the proposed system (case selection through aggregation of fuzzy preferences, case adaptation through genetic optimization with retrieved solutions used as initial population, multi-layered neural networks trained with retrieved cases used for adaptation tasks etc.). The system is evaluated through an example case of parametric design of an oscillating conveyor.

A knowledge-based system for the conceptual design of grippers for handling fabrics

The paper presents a knowledge-based system (KBS) for the conceptual design of grippers for handling fabrics. Its main purpose is the integration of the domain knowledge in a single system for the systematic design of this type of grippers. The knowledge presented, in terms of gripper, material and handling process, are classified. The reasoning strategy is based upon a combination of a depth-first search method and a heuristic method. The heuristic search method finds a final solution from a given set of feasible solutions and can synthesize new solutions to accomplish the required specifications. Details of the main features of the system are given, including its ability to take critical design decisions according to four criteria, weighted by the designer. The knowledge-based system was implemented in the Kappa P. C. 2.3.2 environment. Two examples are given to illustrate some critical aspects concerning the KBS development, to explain the operation of the proposed searching heuristic method, and to show its effectiveness in producing design concepts for grippers.

Use of expert systems for the selection and the design of solar domestic hot water systems

The aim of this article is the study of the application of expert systems to a mechanical engineering research domain with practical and commercial interest, such as design and manufacturing of Solar Domestic Hot Water (SDHW) Systems. The issues studied were the selection and the design of SDHW systems. The application of an expert system was explored. Frame and class formalism was used for knowledge representation together with forward and backward chaining techniques for drawing conclusions and utilizing the accumulated information present. The appropriate computer program was developed to yield the selection of SDHW systems using the software tool LEONARDO 3.0 (1989), an integrated environment for the development of expert systems. The developed program was tested with data according to the Greek standard ELOT corresponding to the ISO/DIS 9459-2 and it performed successfully for 21 SDHW systems available on the Greek market. Apart from the possibility of selection of a SDHW system, the program also supports the facility for updating its knowledge base with new data so that it can be adapted to changes appearing on the market. The program proved to be functional and user friendly to a high degree.

Information generation during design: Information importance and design effort

The present paper studies the process of information generation during design and focuses on the relationship between the information importance and the required effort for its generation. Multiple associative relationships among design entities (handled as design descriptors) are used to represent the design knowledge. The characteristics of the dependent and the primary descriptors are examined and their distinct roles in the design process are discussed. Term definitions concerning the information importance and the design effort are also introduced. The descriptors are used to form a matrix. A number of operations on this matrix results in its transformation, with the final matrix reflecting the quantitative relationship between the information importance and the design effort. From the aforementioned matrix, a unique sorted list for the primary design descriptors is produced. Following this list during descriptor instantiation ensures the production of design information of maximum importance with the least effort in the early design stages. The design of a belt conveyor is used as a basis for a better understanding of the theoretical analysis and for a demonstration of the use of the suggested descriptor list.

Integration of genetic optimisation and neuro-fuzzy approximation in parametric engineering design

The present article introduces a systematic approach that applies soft computing techniques, such as genetic algorithms and neuro-fuzzy approximation, in parametric engineering design, such as genetic algorithms and neuro-fuzzy approximation. A generic design problem representation is utilised and genetic optimisation is used in order to extract the optimal design solution based on customised optimisation criteria. Apart from the extraction of the optimal solution, the genetic optimisation creates records of elite solutions that can be reused as meta-knowledge to enhance the design results in the context of different frameworks. The efficiency of the proposed neuro-fuzzy approximated models and the meta-knowledge supported architectures is evaluated against the conventional and analytical models, on the basis of an example case of parametric design of oscillating conveyors.

A Fuzzy Rule-Based Approach for the Collaborative Formation of Design Structure Matrices

Engineering design requires extensive decomposition and integration activities relying on a multidisciplinary basis. A design structure matrix (DSM) can be used as a representation and analysis tool in order to manage the design process under diverse perspectives. The design outcome is always subject to the abstract nature, the subjectivity and the low availability of the required design knowledge. This paper addresses the DSM as a communicating design tool among multiple designers. A fuzzy-logical inference mechanism permits the collaboration among designers on the qualitative definition of the interrelations among the design problem’s entities or tasks and the resulting DSM may be then utilized for various tasks (partitioning, clustering, tearing etc.) depending on the problem under consideration. A DSM is deployed for the case of the parametric design of an oscillating conveyor where two (2) designers are collaboratively involved.