Patent Application
20100231919
The invention relates to a method for the determination of loads in a mechanical structure and/or of damage or states in the mechanical structure which result from the loads in the mechanical structure. The invention relates, in particular, to a method for the determination of building loads and/or building damage resulting from building loads and to a device suitable for this purpose.
Monitoring the structural integrity of buildings during and after building loads (for example, earthquakes, storms, weight load due to snow on a flat roof) is an important task. In addition to the need to reduce personal damage, in this context the aspect relating to minimizing downtimes in the utilization of buildings must also be taken into account. Moreover, towns which are constantly growing in size, which have highly varying terrain properties, lead to the situation where the effects of building loads, for example of earthquakes, may turn out to be very different from one part of a town to another. The type and structure of individual buildings within a very narrowly delimited region also vary very sharply. Consequently, in an extreme situation, it is possible for one building, when subjected to building loads, to be seriously damaged, whereas a directly adjacent structure remains virtually intact.
In terms of disaster protection, an extremely difficult situation consequently arises, since, as regards the optimization and coordination of rescue measures, there are no simple criteria for prioritizing the aid measures in the event of a disaster in a relatively large municipal area.
The object on which the invention is based is to specify a method and a device for the determination of individual building loads and/or building damage, so that a prioritization of the aid measures in the event of a disaster in a relatively large municipal area is possible.
To achieve this object, the invention provides for the determination of building loads and/or of building damage resulting from building loads, according to Patent claims 1 and 2. Furthermore, the invention provides methods for the determination of loads in a mechanical structure and/or of damage or states in the mechanical structure which result from the loads in the mechanical structure, according to Patent claims 7 and 8. Moreover, the invention provides corresponding devices according to Patent claims 9 and 10. Advantageous refinements or developments of the idea of the invention are found in the subclaims.
The method according to the invention for the determination of building loads and/or building damage resulting from building loads is distinguished in that the rotations of a building part which are caused by building loads or building damage are measured via a fibre-optic rotation sensor which is rigidly connected mechanically to the building part, and the building loads/building damage are deduced from the measured rotations.
In order to make the determination of the building loads/building damage more precise, a multiplicity of rotation sensors may be employed instead of a single fibre-optic rotation sensor; thus, rotations of a plurality of building parts which are caused by building loads/building damage may be measured via corresponding fibre-optic rotation sensors which are rigidly connected mechanically to the building parts, and the building loads/building damage may be deduced from the measured rotations. In other words, each rotation sensor measures the rotations of that building part to which it is connected. The overall state of the building can then be deduced from the sum of the determined rotations of the individual building parts or the building loads/building damage can be ascertained individually for each part of the building. Consequently, it is possible, for example, to detect torsions within a building, for example between two successive storeys of the building.
Each rotation sensor may, if required, be designed individually as a uniaxial, biaxial or triaxial rotation sensor, that is to say each rotation sensor may be designed individually as a sensor which measures rotations about one, two or three axes of rotation. Thus, for example, particularly critical building parts may be equipped with triaxial rotation sensors, in order to allow precise measurement, whereas, in the case of uncritical building parts, for example, biaxial or uniaxial rotation sensors may be sufficient.
In one embodiment of the invention, the rotation sensors are fastened to side walls of the building, in such a way that the relative angles between the side walls and the floors or ceilings of the building which are supported by the side walls can be deduced from the measured rotations. Such relative angles are a proven measure for the assessment of building loads/building damage, in particular of earthquake damage.
In addition to the fibre-optic rotation sensors, acceleration sensors or other sensors may be provided on the building, which are rigidly connected mechanically to corresponding building parts and via which translations of the building parts which are caused by building loads/building damage are measured, building loads/building damage being deduced from the measured translations. The additional provision of such sensors allows a particularly precise determination of building loads/building damage, since all six degrees of freedom of movement (to be precise, three degrees of freedom of translation and three degrees of freedom of rotation) can thereby be measured.
In one embodiment of the method according to the invention, the change in orientation of the building with respect to the axis of rotation of the earth is deduced from the measured rotations and/or translations. The change in orientation may be judged to be a reliable measure of the building loads/building damage to be determined, in particular of earthquake damage.
In generalized terms, the invention provides a method for the determination of loads in a mechanical structure and/or of damage or states in the mechanical structure which result from the loads in the mechanical structure. The method is characterized in that the rotations of a part of the mechanical structure which are caused by loads/damage in the mechanical structure are measured via a fibre-optic rotation sensor which is rigidly connected mechanically to the structural part, and the loads/damage/states in the mechanical structure are deduced from the measured rotations.
The invention relates, furthermore, to a method for the determination of loads in a mechanical structure and/or of damage or states in the mechanical structure which result from the loads in the mechanical structure. The method is characterized in that the rotations of a plurality of parts of the mechanical structure which are caused by loads/damage in the mechanical structure are measured via corresponding fibre-optic rotation sensors which are rigidly connected mechanically to the structural parts, and the loads/damage/states in the mechanical structure are deduced from the measured rotations.
The term “mechanical structure” is to be understood as meaning, for example, a building or part of a building (roof), a bridge, a tunnel, the fuselage or wings of an aircraft, part of the ground (for example, the rock of a mountain), a conduit or a traffic route (road, rail tracks, railway lines), etc.
The invention provides, furthermore, a device for the determination of building loads and/or of building damage resulting from building loads. The device is distinguished by
The invention relates, furthermore, to a device for the determination of loads in a mechanical structure and/or of damage or states in the mechanical structure which result from the loads in the mechanical structure, which device has:
The statements (embodiments) afforded within the framework of the method according to the invention apply similarly to the devices according to the invention: thus, the devices according to the invention may also consist of a plurality of fibre-optic rotation sensors; the rotation sensors may be designed as uniaxial, biaxial or triaxial rotation sensors; etc.
The evaluation apparatuses of the individual rotation sensors may be network-linkable to further instrumental units, so that rotation data determined can be collected in a simple way, so that a rough damage survey can be set up even shortly after building loads/building damage. In this case, it is possible both to provide each rotation sensor with a specific evaluation apparatus and to provide a single central evaluation apparatus which is connected to all the rotation sensors and other sensors. The network-linking of the evaluation apparatuses to further instrumental units or the network-linking of the rotation sensors to a central evaluation apparatus may take place, for example, via cable lines or via radio connections.
According to the invention, therefore, a sensor based on absolute rotation measurement is employed, which detects the deflection of the building during an earthquake dynamically in three spatial directions by means of a specialized fibre-optic gyroscope. In this case, the maximum deflection and the cumulative deflection of individual storeys or of the entire building are determined in real time and are compared with a predefined building-specific tolerance table which has previously been anchored in the sensor or in an evaluation apparatus connected to the sensor. To assist immediate measures, the result of this building evaluation is indicated, for example, in a simple multi-step colour code and made available at the same time from an external interface. This makes it possible to connect a plurality of such sensors into an overriding functional unit and to incorporate individual sensors or sensor complexes in a radio network.
The invention is explained in more detail below in exemplary embodiment, with reference to the figures in which:
Components or component groups corresponding to one another are identified in the figures by the same reference numerals. In the following description, it is assumed, for the sake of simplicity, that the building loads/building damage are earthquake damage.
As has already been indicated, the essential element of the concept is a (preferably) three-component inertial fibre-optic rotation sensor which is firmly connected to the structural substance. In this way, the sensor participates in the movements of the building (or building sections) and delivers the rate of rotation about three linearly independent spatial directions (for example, about the “longitudinal axis”, and the “transverse axis” of the building and about the horizontal plane of the latter) as a sensor signal. These rates of rotation are integrated in the processor part of the sensor, and the cumulative maximum deflection with respect to an inertial reference system is calculated. The maximum deflection angle between a supporting wall and the concrete ceiling of a building is one of the critical variables for assessing the remaining load-bearing capacity of this building part. Too great a deflection of a ceiling panel in relation to its support leads to an overstressing of the supporting structure and, subsequently, to crack formation which is detrimental to the load-bearing capacity. In this context, it is unimportant whether this deflection takes place abruptly or in a manner distributed over a relatively long period of time. What is decisive for the durability of the structure is the maximum deflection angle.
As compared with the use of conventional accelerometers, the method according to the invention has the advantage that it is based on the principle of absolute rotation measurement and covers a very broad dynamic range. Thus, not only high-frequency deflections are detected, but also inclination variations taking place very slowly in frequency ranges which are no longer accessible by inertia sensors.
Depending on the arrangement of a plurality of sensors in a building, therefore, not only is the inclination of a structure assumed to be rigid obtained, but also the differential inclination or torsion between individual storeys and building parts equipped with such sensors.
Accelerometers which function according to the concept of mass inertia may additionally be employed here. It thus becomes possible, for the first time, to acquire a complete sensor for all six degrees of freedom of movement (to be precise, three degrees of freedom of translation and three degrees of freedom of rotation).
The building sensor 3 according to the invention (which may also contain translation sensors in addition to a fibre-optic rotation sensor) detects rates of rotation or the integral deflection angle of a building or building part under the action of an earthquake. For this purpose, it is to be mounted rigidly on fixed side walls 1 which represent the building behaviour, as shown in
In principle, damage to the building may also take place in the other horizontal spatial direction, as may also occur in a rotation about the vertical axis (torsional vibrations). A complete sensor triad covers all the directions of movement. Depending on the structure of the building, however, there may also be directions which are less at risk, for example on account of higher rigidity, so that individual sensor components may be dispensed with under certain circumstances.
Since rotation sensors detect rotations absolutely on the basis of the Sagnac effect, the orientation of a building in relation to the axis of rotation of the earth can be evaluated automatically before, during or after an earthquake in real time as a measurement criterion. This makes it possible to determine the change in orientation of a building, without having to rely on local references which could, of course, have changed due to the action of an earthquake.
By dynamic radio network-linking of any of these sensors, for example in entire town parts, an area-covering rough damage survey can then be set up, close to real time (smart sensor application). This could be retrieved and utilized, for example, by disaster protection services, for the preparation of priority lists (step S4).
Further aspects of the invention are to be explained in the following description:
Although fibre-optic gyroscopes do not yet have the sensor sensitivity of large ring lasers, nevertheless, because of their small dimensions and low power consumption, they are suitable both for field use in the region near a seismic event (aftershock zones, volcanoes, etc.) and for the instantaneous detection of building states after earthquake actions, but also for the monitoring of static deformations. Combining rotation sensors and seismometers or accelerometers affords the possibility of developing and utilizing a complete sensor for all six degrees of freedom of movement. An important aspect in this context is to take into account the particular conditions of rotation sensors for measurement free of local reference symbols, for example the possibility of absolute orientation of the sensor in real time, and also the self-calibration of the sensor by means of the global earth rotation signal.
According to the invention, novel mobile and cost-effective sensors are provided, which, together with classic seismic sensors, make it possible to observe ground movements correctly, specifically in all six degrees of freedom of movement: for a complete description of ground movements, not only translations, but also rotations, must be observed with high resolution.
According to the invention, long-term extensive statements on variation and deformations can be made by means of broadened six-component recordings (novel measurement methods which can be used in a mobile, cost-effective and rapid way). The additional observation variable “rotational movements” can also be used, above all, in the inversion of seat parameters (dynamic rotation), and its accuracy can be improved. In this case, in particular, the otherwise inseparable coupling of inclination and translations of the measured signals on the sensor plane can be investigated and corrected by the provision of correspondingly high-resolution rotation measurements. Since the transverse acceleration and the rate of rotation during an earthquake are in phase with and proportional to one another, it is not possible to prevent the seismograph from including a coupling of earthquake-induced inclination fractions into the sensor signal of the other linearly independent spatial directions in a zone near the earthquake event. Only a sensor with all six independent degrees of freedom of movement, that is to say including the rotation, can afford an improvement here.
The concept according to the invention in the design of fibre-optic gyroscopes can cost-effectively increase the sensitivity of existing rotation sensors based on glass fibres by up to two orders of magnitude, as compared with the previous prior art, and, when supplemented by a broad-band seismometer, can for the first time make all six degrees of freedom of movement available with sufficient sensitivity in a portable implementation capable of being used in the field.
The invention concentrates on the utilization and further development of modern rotation sensors as a novel and innovative basic technology for use in early-warning systems. The diverse novel technological possibilities arising from this are to be applied, in particular, to the area of the monitoring of building structures and the area of seismology.
After an earthquake, particularly in larger metropolitan areas, a rapid estimation of the damage to important buildings is of central importance, particularly in order to optimize the rescue measures. If the co-seismic movements (time profile of the deformations, static displacements and static rotations) could be measured and could be evaluated in terms of critical variables, close to real time, in an independent sensor, then deformation-induced variations could be determined virtually instantaneously, above all for critical buildings and conduits or traffic routes. According to the invention, a cost-effective intelligent 6-C sensor is provided, which can be installed in the manner of a “black box” in buildings, bridges, tunnels or other structures and which records useful data material relating to the building behaviour during an earthquake or another deforming event and provides this data material for analysis. According to the invention, therefore, particular attention is also placed on utilizing the properties of the rotation sensors as absolute protractors. The devices according to the invention are distinguished by a low power consumption and are suitable in manufacturing terms for cost-effective mass production. The installed electronics, by the instantaneous integration and differentiation of the sensor signals within the sensor, can thus be expected to deliver a hazard or damage estimation, close to real time, for the respective building. The additional capability of the combination of such sensor units of independent design into autonomous local or regional networks is in this case possible.
The aim, in terms of structural mechanics, is, via suitable sensors, to determine dynamic excitation as a result of earthquakes or other subsoil movements, such as vibrations, both as regards amplitudes and as regards relevant wavelengths. By coupling geophones or acceleration sensors with rotation sensors as absolute protractors, both characteristic variables of dynamic excitation can be detected by means of a small number of measurement points. For linear structures, the curvatures caused by the action can be described from measured rotations. According to the invention, building on this, the evaluation of the excitation can be carried out via a comparison of the measurement results with the assumptions on which the dynamic design of the structure is based. Furthermore, in addition to the detection of dynamic excitation, the remaining static deformation caused by the action is of interest. For example, by lining up a plurality of sensors with one another, the remaining deformation line can be determined via the measured rotations. The possibility of measuring the (quasi-)static and the dynamic information by means of cost-effective sensors constitutes a considerable advantage, as compared with other measurement methods. The self-calibration of the sensor and its effect independent of position as regards weight open up a broad field of use also in terms of early warning against the consequences of antropogenic actions or critical action/resistance combinations, such as, for example, high snow loads on weakened structures, actions due to extreme wind loads, etc. and could thus allow the use of the sensors even for tasks which, with other measurement methods, can be carried out only in a very complicated way and not in real time. An important area of early-warning systems which goes beyond seismic excitation and is covered by the method according to the invention is the monitoring of subsoil movements, such as occur in tunnel building, during earth excavation, due to foundation loads or else in hill subsidence zones. Consequently, with the aid of sensors, monitoring covers dynamic actions, that is to say the causes of deformations or damage, together with the remaining deformations.
Deformation observations in earthquakes and on volcanoes
The problems in determining the static (and dynamic) deformation from seismometer observations have been known for decades, without a satisfactory stable solution having been found hitherto. The main reason for the problems is the additional effects in seismographs which occur due to rotations and changes in inclination and lead to the situation where speed or acceleration seismographs cannot be integrated. GPS instruments are only partially an improvement here, since (1) the vertical resolution is low and (2) the sensing rate is not sufficiently fine for the dynamic observations. With cost-effective 6-C seismometers capable of mobile use, an area-covering observation of dynamic and static deformations would be possible. This could, in the long term, also allow improvements in the real-time determination of seat parameters. Similar arguments apply to seismic observations of volcanoes in which sometimes pronounced deformations are to be expected. The unavoidable coupling of rotational and tilting movements influences the standard broad-band measurements by means of seismometers and makes modelling more difficult. The combined data analysis of rotations and translations makes it possible, in this area, to have an improved mapping of seat processes and of the state of the magma chambers.
In one embodiment of the invention, both the dynamic ground/building interaction and the structural dynamics of the building structure are to be taken into account in determining the loads/damage.
The aim of the invention is to reduce earthquake-induced hazards and risks. The observations of the entire movement components have been urged for decades by theoretical seismologists. The most recent developments in sensor technology now seem to allow the design of a suitable measuring instrument with the required accuracy. In this context, it should be noted that measurement technology has a broad potential for use in both the purely scientific sector and the engineering sector.
The ever-continuing rapid growth of large metropolitan areas in earthquake-endangered regions (for example, Istanbul, Tokyo, Los Angeles and New Mexico) leads to, in the case of a strong earthquake triggered beneath them, to tremendous damage and the loss of many human lives. On account of the completely heterogeneous building structures and variable subsoil properties in these mega-metropolises, it is not possible in a short time for the respectively responsible disaster management to obtain a survey of the damage situation. The rapid initiation of a coordinated rescue action is consequently seriously impeded, and the expected result is not the best possible. Autonomous monitoring systems, which co-log a damage event and evaluate, close to real time, according to predefined methods and make the result quickly available, could lead to a decisive improvement here. By dynamic weighting according to the function of the building and the automatic damage assessment carried out by the sensor, and also other criteria to be laid down (for example, optimization of the travel distance for rescue forces), an effective utilization of the critical first 6 hours after a damage event may be envisaged.
The building sensor concept according to the invention is innovative in two respects. On the one hand, it is intended to show the way to a monitoring system in which the building behaviour is logged in detail during the action of external forces (earthquake, extreme wind load, ground subsidences, etc.). In this case, a complete independence of local reference systems is achieved via the property of the rotation sensors used as absolute protractors. In this regard, the sensor corresponds to a “black box”, such as is employed in aircraft. In addition to these logging functions, key parameters (for example, maximum deflection) are investigated, during measurement, for the overshooting of building-specific predefined limit values, and, if possible, may allow a classification of the damage caused (for example, the degree of risk of collapse) and transfer this to a central location (local disaster protection cell) (use of mains power-independent radio technology). This would consequently perform an important step towards quasi-real-time damage prognosis.
The sensors developed for field use in the region of strong ground movement have several shortcomings in seismology. Inclinations of the seismometer induce both a sensor signal and the translations actually to be measured. In near-earthquake zones, the complete movement vector cannot be reconstructed on account of the absent degrees of freedom of rotation. Consequently, the measurements are both incomplete and, because of the intermixing of signal sources, faulty. This constitutes a serious impediment to the inversion of earthquake parameters. It is to be expected that the complete detection of all six degrees of freedom of movement and the clear separation of translation and rotation (inclination), along with sufficient sensor sensitivity, can make a marked contribution to improving the seismological models (independence from apparatus transfer functions).
Ring lasers are highly sensitive active optical interferometers and, hitherto, the only instruments which can quantitatively determine the rotation signals from teleseismic events. Even for use in seismology, where long-term stability is not a major factor, they still require a very high temperature constancy of better than 0.1 degrees per day. Thermally induced expansion or contraction of the optical resonator within the range of a few hundredths of a micrometer cause a drift in the optical operating frequency which then leads to sudden jumps in the longitudinal mode index (mode jump). Furthermore, despite all the simplifications which it has been possible to implement successfully in the course of the GEOsensor project, ring lasers remain a complex installation. This makes them unsuitable for three specific applications. (1) A short-term transfer of the sensor into the vicinity of the epicentre after a strong earthquake, in order to record aftershocks in the near field. (2) The detection of the complete movement (all six degrees of freedom) under the influence of strong ground movements. (3) As part of a monitoring building sensor system, they are too costly and much too sensitive. On the other hand, fibre-optic rotation sensors, although not having the sensitivity of ring lasers, are nevertheless robust and compact sensors with a markedly higher temperature tolerance. Furthermore, they are significantly more cost-effective.
The method according to the invention can utilize the self-calibratability of the system. Since rotational sensors measure rotations absolutely on the basis of the Sagnac effect, the earth rotation signal is also always superposed on the sensor signal. This signal may be considered, within the framework of the application proposed here, as a constant reference. It is available at any time point, that is to say before, during and after an earthquake, as a reference independent of the local surroundings and can be used in order to determine the instantaneous orientation of the sensor in relation to the axis of rotation of the earth. A comparison before and after an earthquake delivers a change in orientation of the sensor due to the seismic event or to the building deformation. Since the sensor is mounted firmly on a wall for a measurement, this change in orientation corresponds either to a building inclination or to a deformation or to a displacement between individual storeys. This signal is determined by the integration of the measured rate of rotation, and, even after a short integration time of approximately one minute, a resolution of less than 1 degree is achieved. Strictly, this method of determining the orientation of the sensor applies with high resolution in the north/south direction only. The resolution is lower in the east/west direction. The investigation of the optimal sensor implementation (reference system of the sensor) is therefore a specific operating point. After the investigation as regards the utilization of the orientation signal, the utilization of the rotation signal of the buildings is a major factor. For this purpose, (1) criteria are to be set up between measured angular deflection and stress of the building or various materials. (2) The effect of the displacement of individual storeys in relation to one another is to be quantified (interstorey rotation) and transmitted to the sensor system.
The output signal of an FOG (Fibre-Optical Gyro) is the phase difference between light waves running around a surface in opposite directions. It is proportional to the rotational speed of the sensor and to the surface circumscribed by the light beams. The main advantage of an FOG, as compared with a ring laser, is that the optical signal can be guided in a glass fibre. In an FOG, the effective surface can be increased, without the dimensions of the sensor having to be varied appreciably. On the other hand, fibre-optic gyroscopes differ from ring lasers in that they carry out phase measurement, whereas ring lasers determine a frequency difference by interferometry and thus achieve a higher resolution by virtue of the concept adopted. This is a great disadvantage in the implementation of an FOG. The signal-to-noise coefficient (resolution) of an FOG depends not only on the scale factor, but also on the optical power density. For example, for a light source with a wavelength of 1.55 μm, a luminous power of PO=100 μW, a length of the glass fibre of 1000 m and a diameter D of the coil of 0.4 m, a theoretical value for the sensitivity of an FOG of 4.125·10-8 rad/s is obtained.
The coil consisting of polarizing optical fibre constitutes an essential component of the optical path. The fibre length and the geometric diameter of the coil are incorporated linearly into the scale factor of the gyroscope. In principle, therefore, lengthening the coil fibre used makes it possible to achieve ever higher accuracies in the measurement of the rate of rotation. At the same time, however, attenuation losses of the light intensity during passage through the fibre and phase change occurring due to disturbing effects, such as, for example, the Shupe effect or the Kerr effect, limit the implementable length of the fibre coil. Coils with a length of approximately 1000 m and with a diameter of up to 300 mm are therefore planned for the prototypes.
It may be mentioned, in this regard, that earthquake-induced rotation observations of teleseismic events are compatible with collated broad-band translation measurements, and that additional information is contained in the amplitude ratio.
| Number | Date | Country | Kind |
|---|---|---|---|
| 10 2006 005 258.7 | Feb 2006 | DE | national |
| Filing Document | Filing Date | Country | Kind | 371c Date |
|---|---|---|---|---|
| PCT/EP2007/000833 | 1/31/2007 | WO | 00 | 5/24/2010 |
