An Expert System for the Seismic Protection of Architectural Building Heritage

An expert system for the seismic protection of architectural building heritage.

Silvia Bruno(1), Angela Di Stefano(1), Carlo Gavarini(1), Antonio Padula(1).

(1) Department of Structural and Geotechnical Engineering, University of Rome, “La Sapienza”, via Eudossiana, 18, 00184 Rome, Italy.

Abstract

The definition of a form for the seismic protection of Architectural Building Heritage, a work carried out for the Italian Research Council, has led to the implementation of a knowledge system, that presently consents to partially analyse the walls of historical buildings. Among other things the system uses geometrical reasoning techniques that permit to survey the building by means of articulation and composition of elementary items, “built”, case by case, with a specific geometric, material and constructive entity. Such items can represent an elementary section of the building, for instance a pillar, an arch, a brick etc., and starting from their position and characteristics the walls and all the used masonry and finishing types can be individuated in detail without the necessity of applying to approximate formulations or to formal and typological evaluations that, with their rigid terminologies, can oversimplify the complex of historical buildings and their importance in the context.

1 Introduction

Seismic protection of historical building heritage is an extremely complex operation hardly fit for procedures of simple translation of operative modalities That is why the ample and articulated engineering discipline, that can be synthesised as aseismic, must be supported by introduction of elements of restoration. The operative procedures require therefore close examination of the historical building in order to evaluate all the parameters attesting its historical and/or artistic values. Only after we have established which characteristics of the building testify its being a unique manufacture, considering its properties from all the points of view (geometric, constructive, material, conservative etc.), we can go on and analyse what should be done to assure its protection against earthquakes.

In most cases information relevant to a building with historical and/or artistic value is organised though operations of framing that permit to survey the properties of the building according to a series of opportunely predisposed parameters. Most of the forms so realised have different investors and users, and consequently, also their finalities differ. In some cases the card-indexing serves for expeditious surveys made after a seismic event (that of Pompei is a didactically important example [1] ); in other cases it has been used to know if the inhabitants of an inner city accept a rehabilitation operation (as in the case of Vicenza [2] and of Milan [3] ), still in other cases it has helped in assessments of vulnerability, (as at Pigna di S. Remo [8] ). Unfortunately these experiences are often only episodic, limited in presupposes and aims, but the possibility to evaluate their effects with time could render the study of these analysis routines interesting. Moreover, many experiences of this kind use informatics only to improve the management of data collection.

2 Data structure and expert system

The work aims first of all at creating both the presupposes for expeditious, large-scale censuses as well as the conditions for the detailed evaluation of the leading characteristics of each single building. It is undoubtedly useful to carry out computer-aided data collection for acquisition of knowledge about historical buildings; it can serve also for the local government units that often permit to carry out interventions on their building heritage ignoring the necessity for this kind of preliminary data acquisition. If the results of this kind of interventions at the end are anyhow acceptable, the merit is in the diligence of the operators and not in the planning of rehabilitation operations to which the definition of modalities of an intervention should always be based.

The first stage of the work has consisted in the definition of the data necessary for the evaluation of the present situation of the building, for the singling out of the structural behaviour models and for the directives of possible interventions. Our main goal is to construct an “easy” technology for all the future users, and for this reason the patterns for data acquisition must have a clear approach and each inference from acquired data must be based on evaluations that the same system possesses, as a guide for the evaluations. In the same time the necessity to evaluate buildings different from each others, more or less complex, and within contexts that can be both easy or difficult to specify, produces the possibility to deepen the survey according to the necessities of the single case in question. As a matter of fact the architectonic building heritage is notoriously various and consistent: if the variety suggests the use of approaches that could be called of “case by case”, the consistence instead suggests more global strategies.

We have then passed to the design of a knowledge system, that, as a prototype, should consent to evaluate the complex of buildings of which the inner cities are constituted.

The data to acquire have been divided in four main sections, relative to principal parts of which a building is composed: foundations, walls, floors and elements of vertical communication.

Actually the part relevant to walls has been implemented.

Each wall, distinguished as internal or external, is denominated and measured, then the characteristics of materials, openings, finishing and static decay are described. The principal aim of the section relevant to data regarding the materials is to single out the different materials present in the object of survey, defining their nature, typology and provenience, as well as the constructive characteristics, visible disposition, thickness, finishing; then the dimensions of the different elements and the characteristics of mortar (when present) are given (values per cent of fractions of fine, medium and coarse grain materials and their quality are indicated) (Fig.1).

Figure 1

For each fragment of masonry so defined the grade of degradation is evaluated, distinguishing among physical, chemical and biological degradation and giving their value in percentage. Then the fragment, with a form representing it, is positioned on the general graph of the wall. At the end of the survey of materials, the part is represented as a sum of numerous sections of masonry and the geometrical disposition of these elements will serve to the same system in evaluating the structural behaviour of the building item, through geometric reasoning and rule bases that will be inserted successively. The section of data regarding the openings serves to position on the wall the various openings present, and their most recurrent forms, and to define the different frames and the relative typologies, materials and their degradation, always estimated as already explained.

The section regarding finishing presents itself as a series of Chinese boxes. A principal element can be defined together with all the elements of which it is composed, as for instance a cornice with fillet, rod, drip, etc. The composition and the state of conservation of each element is surveyed.

The section relevant to static decay specifies first of all which kind of decay is in question and, for each case, the area and the entity of degradation (Fig. 2).

Figure 2

The survey operation has a graphic interface that, on the scale drawing of the wall, visualises all the elements present. Such elements, numerated automatically, can be thought to be many essential cells, elementary items with a geometric, material and constructive entity of their own, apt to evidence all the solutions of continuity that the wall presents.

Their number, form and dimensions are proportional to the subject that has been surveyed and therefore they are the guarantee of the exact evaluations, not only typological or formal (Fig. 3).

Figure 3

Through geometric reasoning these elementary items “built” case by case, can be managed by the system for inferences, for instance to single out the macroelements with which the most suitable models of structural behaviour are formed, or to define contextually the elements of particular value (requiring conservation and therefore contributing to render the potential intervention more complex).

At the present stage of the work the surveyed data serve to give information about the historical building at the simplest level. In the overall design this stage has been defined as the phase of identification and of census; to this we must add all the data necessary to collocate the events chronologically in order to have a general view of the evolution of the phenomena that have interested the building and to be able to reconstruct its synchronic and diachronic development.

As well known, a common survey method establishes defined knowledge levels, and then deepens them progressively; in some ways this kind of methodology serves also to ordinate the different arguments of study involved. In the implementation of our system, instead, the data are dealt with less rigidity. The system, in fact, according to a more dynamic vision, accepts input of data from whatever level. In fact, often (apart from exceptions) the same possibility to acquire data is a sign of the importance of the building.

In a second phase of the work the system has been tested evaluating a series of real cases. This phase has first of all defined the intervals and the domain areas of the given parameters and made possible a first verification of the characteristics of the knowledge system, from facility of compilation to correspondence of the model. Moreover, it has been possible to single out a series of pilot-cases, that are going to serve as an aid in the formulation of evaluations.

On the other hand it isn’t necessary that the user himself is a specialist; within the Expert System the surveyor can find the simplest approaches for the acquisition of knowledge relevant to the item in question, without the necessity to proceed through extremely formalised concepts that are often unfit for the choices relevant to investigation methodologies.

As it is difficult to pursue deep knowledge within schematic and immutable readings, the present methodology aims at flexibility, at direct connections with the realty of the construction surveyed. We should aim at the quality of our operations, indiscriminately, both in knowledge acquisition, or in interventions and also in the successive utilisation of the building unit.

The collected data will constitute the knowledge dominion relative to the given item and only to it. From the compilation of more items we can gather information relative to a series of items, starting from which we are able to outline, within a territory, the dislocation of constructive typologies, the entity of phenomena of decay etc. in order to suggest the most suitable intervention strategies.

3 Conclusions

The goal we are aiming at with this work is not to have an easy support for the simple informatics application of the data, but also to have at disposal an “intelligent” aid for the acquisition of knowledge about the item itself and for the evaluation of intervention criteria that require, in order to assure the respect of material, architectonic and artistic qualities of the manufacture, assessments regarding different and complex doctrines of study.

The methodological moment of recognition of a work of art or of the historical value of a population of buildings taken as a whole should be the premise for every intervention; it can be realised during the survey operations only if we offer various keys for the interpretation, transversal to different research spheres, of historical, artistic and functional type. The research methodologies characteristic to the history of architecture could be managed, in the continuation of this work, by a Fuzzy system and the evaluations of different observers and whizes should be the examples of expert behaviour that feed the knowledge-base of the system. The elements constituting the evaluations form the canons of interpretations in architecture. Many of them (confrontation, typology, deduction-induction, iconology, …) are often used unconsciously, taken for granted, overlooked or not at all used. It is anyhow necessary to make these elements of evaluation operative within the system and make them interact with the user, transmitting him the possible survey methodologies as a guide for the formulation of evaluations relevant to the item surveyed.

In definitive the tool we are constructing should consent a complete multidisciplinary approach, with possibilities of closer examination at different levels to the problematics of data collection and to their intelligent and profitable management, resolving in particular a sort of conflict that up to now has existed between the too “engineeristic” approaches, such as vulnerability forms, and the too “architectonic” approaches, such as catalogues; that is why we aim at consenting an approach similar to that for which Giuffré had proposed the name "Code of Practice" supplying many examples [04,5,6,7]. This kind of approach has then been used also by others, both in Italy and at international level. [08/13].

References

1. Braga F.: Un metodo per una campagna di rilevamento dello stato di conservazione del patrimonio monumentale, I C R 1983

1. (A cura di) Conti G., Paolillo P. L., Vianello D.: Riuso: dai piani alle realizzazioni, Peg Editrice,1983

1. Molinari C., Treu M. C. : Rinnovo edilizio a Milano,“Recuperare”, a. 2, 1983, n. 5, p. 185-195.

1. Giuffré A. (coordinatore), Zampilli M., Ceradini V., Jacovoni F., Pugliano A.: Centri storici in zona sismica. Analisi tipologica della danneggiabilità e tecniche di intervento conservativo: Castelvetere sul Calore, Rapporto n. 8 della serie "Studi e ricerche sulla sicurezza sismica dei monumenti", Dipartimento di Ingegneria Strutturale e Geotecnica, Università di Roma "La Sapienza", Facoltà di Architettura, nov. 1988; pubblicato anche quale appendice al capitolo IV nel volume II di "Ingegneria antisismica", 3a edizione, editoriale ESA, Milano 1991

1. Giuffré A. (coordinatore): Sicurezza e Conservazione dei Centri Storici. Il caso Ortigia, Editori Laterza, Roma-Bari, 1993

1. Giovanetti F. (coord.), Giuffré A. ed altri: Manuale del Recupero di Città di Castello, Edizioni DEI Tipografia del Genio Civile, Roma

1. Giuffré A., Carocci C.: Codice di pratica per la sicurezza e la conservazione dei Sassi di Matera, Edizioni La Bautta, Matera 1997

1. Chirico A., Corsanego A. ed altri: Recupero Vulnerabilità e Rischio Sismico nei Centri Storici della Liguria, Il caso del Centro Storico "La Pigna" di Sanremo, Regione Liguria e CNR/GNDT, maggio 1992

1. Cremonini I., Foschi M.: Il recupero edilizio in zona sismica, Regione Emilia Romagna, Comuni di Santa Sofia e di Sarsina, maggio 1992

1. The Martin Centre of Cambridge University (coord.), ISMES Bergamo, Dipartimento di Scienza delle Costruzioni della Università di Napoli, DISSAR della Università di Pescara, LRC of the National Technical University of Athens, Istituto Superior Tecnico of Lisbon: The TOSQA Project, Earthquake Protection for Historic Town Centres, Final Report for the CEE, D.G. XII, June 1996

1. European Prestandard, Draft prENV 1998-1-4: "EUROCODE 8: "Design provisions for earthquake resistance of structures, Part 1.4 Repair and strengthening", CEN, European Committee for Standardization, 29\06\1995, Appendix D "Vulnerability methods"

1. Idem: Appendix F "Particular considerations for historical buildings and monuments"

1. Giuffré A, Gavarini C.: Strengthening and repair of masonry buildings in historical urban areas - Draft document within the activity of PT 1.4/EC8, Atti del Convegno Inter-nazionale sulla Protezione Sismica degli Edifici, Ancona, giugno 1991

1. Ballardini R., Gavarini C.: Normativa Tecnica - Regole del Restauro: Sicurezza e Livello di Rischio accettabili, V Convegno Nazionale ASS.I.R.C.CO., Orvieto maggio 1997

1. Gavarini C.:Linee Guida per il restauro di Noto, Convegno "Edifici Storici: Tutela e Normativa di Sicurezza", Centro Congressi Torino Incontra, Torino, 22-23 maggio 1997

1. Braga F., Gavarini C.: Quale futuro per la Normativa Sismica Italiana, "L'Ingegneria Sismica in Italia": 8o Convegno Nazionale, Taormina, 21-24 Set. 1997

1. Bruno S., Gavarini C., Padula A., Vittori F.: Costruzione di un Sistema Esperto sulla vulnerabilità sismica tramite l'impiego della Shell KSBT, "L'Ingegneria Sismica in Italia": 8o Convegno Nazionale, Taormina, 21-24 Set. 1997

1. Gavarini C.: Il problema della sicurezza sismica nel recupero dei centri storici, "Seminario di divulgazione urbanistica per il rilancio dei Centri Storici", Carpineto Romano, 29 novembre 1997

1. Gavarini C.: Dalla vulnerabilità degli oggetti alla vulnerabilità dei sistemi, Seminario Internazionale "Evoluzione nella sperimentazione per le costruzioni", CIAS, Corinto (Grecia), 26 aprile/3 maggio 1998