ES2664131B1 - PROCEDURE FOR CHECKING THE COLLAPSE OF STRUCTURES OF INDUSTRIAL ESTABLISHMENTS IN CASE OF FIRE THROUGH DEBILITATION VIA DECREASE OF THE PROFILE SECTION - Google Patents

Description

D E S C R I P C I O N

PROCEDURE FOR CHECKING THE COLLAPSE OF STRUCTURES OF

INDUSTRIAL ESTABLISHMENTS IN THE EVENT OF A FIRE THROUGH

WEAKENING VIA DECREASE OF THE PROFILE SECTION

SECTOR OF THE TECHNIQUE

The present invention falls within the field of construction, more specifically in the field of fire protection measures in establishments for industrial use.

BACKGROUND OF THE INVENTION

Currently, fire protection measures in establishments for industrial use are regulated, in Spain, by Royal Decree 2267/2004, of December 3, which approves the Regulations on Fire Safety in Industrial Establishments.

In the aforementioned Regulation, industrial buildings are classified according to their Configuration and Location in relation to their surroundings, creating three types of establishments located in a building:

• Configuration type A: Ships attached to others with shared structure.

• Configuration type B: Ships attached to others without shared structure.

• Type C configuration: Isolated building with a distance greater than 3 m from the rest of the buildings.

The fire protection measures for a ship type A are more demanding than for a type B, and in turn, the measures of type B are more demanding than those of type C. This causes that for a ship type A, and In particular, if it is intended to comply with the requirements of the regulations, whether old or new, the implementation of fire protection measures requires a large economic investment.

As it is well known, in most industrial estates, the type of construction that predominates is type A, because normally the execution of these buildings is done in groups, making a common structure for a set of buildings.

Well, the procedure object of the present invention has as its purpose, in essence, to achieve that the structural behavior of an industrial building type A is adjusted to that corresponding to an industrial building type B, thus reducing the needs, and legal requirements, with regard to fire protection measures to be implemented in these industrial establishments, as well as opening the door to place them in locations not allowed for buildings type A.

. EXPLANATION OF THE INVENTION

The object of the present invention is a method of controlling the collapse in case of fire of structures of industrial establishments comprising the following stages:

1. Characterization of the industrial establishment whose collapse in the event of fire is to be controlled.

2. Design and calculation of the impairment to be practiced in the structure of said industrial establishment.

1. Characterization stage

In a preferred embodiment of the method object of the invention, the characterization stage of the industrial establishment comprises obtaining information about its configuration and its location with respect to its surroundings, in particular information on its geometry, of the profiles of which the structure consists of the establishment (section and materials), and of its geographical location, including the corresponding wind and climate zones. In an even more preferred embodiment, said characterization step of the industrial establishment further comprises the determination of the intrinsic risk level of the industrial establishment as a function of the weighted or corrected fire load density. In any of the cases, this information can be obtained either experimentally, by observing and using the tools and methods that correspond, either from sources of documentation in which this information is available, or by combining documentary sources with experimental determinations. .

In an even more preferred embodiment, the characterization stage of the industrial establishment comprises obtaining information on the actions or loads that act on the structure of said establishment, especially those acting on vertical walls and roofs. In an even more preferred embodiment, said actions comprise:

• Permanent loads: Own weight of all elements (structure, facades, roofs, anchors, doors, etc.).

• Snow overload: Action on the roof that, in a horizontal terrain, is determined by the altitude and by the winter climate zone.

• Action or wind load: Action on the roofs and walls.

• Overload of use: It takes into consideration the weight of the people on the cover, for example when they go up to it to carry out maintenance tasks.

• Seismic action or load: Its value depends directly on the mass of the building.

This information can be obtained either in an automated way, through the use of some calculation program to obtain actions in structures (for example, a program that automatically determines the values of the actions to be applied, defining the user, previously, the set of charges that act on the surface); in a non-automated way, by calculating the loads, with the help of the Technical Building Code (CTE), and its decomposition on the belts and pillars in the form of linear loads. In a preferred embodiment, determining the action of the wind on a vertical or covered face comprises the determination of:

• Dynamic pressure, dependent on the geographical location of the establishment;

• exposure coefficient, depending on the height of the building and the degree of roughness of the environment, whose value is always greater than unity, varying strongly with height; Y

• pressure coefficient, which determines the direction of the wind (a positive value indicates that the wind exerts a pressure on the surface, a negative sign indicates suction), which, in turn and in a more preferred embodiment, is decomposed into:

o Coefficient of external pressure, dependent on the shape of the roof or wall being studied; Y

or internal pressure coefficient, depending on the slenderness of the building and the location of the holes in it. This coefficient subtracts the previous one, and considers the action of the wind inside the building, understanding that there are open gaps in the envelope of the same.

In an even more preferred embodiment, the determination of the action of the wind on a vertical or covered face is made taking into consideration the following three wind directions:

■ Wind + X: Corresponding to the direction parallel to the frames, will determine the most important wind load on the roof and on lateral pillars. In case of non-symmetrical structures, it is necessary to consider the wind in -X.

■ Wind + Y: Wind perpendicular to the front façade, which will serve to check the front gable and the bracing.

■ Wind -Y: Same as the previous one, but on the rear facade.

In this even more preferred embodiment in which the three possible directions of the wind are taken into account, the calculation of the wind (+ X) in vertical walls assuming closed gaps, load acting on the four sides of the building, comprises the determination of:

• Dynamic pressure,

• exposure coefficient, dependent on the average height of the enclosure and the degree of roughness; Y

• pressure coefficient, dependent on the wall in question, and it is possible to distinguish between three different zones or regions in the case of lateral façades, in which case it is possible to adopt a single pressure coefficient per wall, calculated as the weighted average of the coefficients of each region, that is: facing façade, opposite façade, and lateral facades.

In this even more preferred embodiment, the calculation of the wind ( + X) in vertical walls assuming open gaps, assumed in which the internal pressure acts, it is necessary to determine, in addition to the dynamic pressure and the coefficient of exposure, the pressure coefficient interior, dependent on the slenderness of the ship, and the percentage of holes that are suctioned (or under pressure) when the wind takes the X direction. This internal pressure coefficient can be calculated, as in the case of the pressure coefficient in the case of wind ( + X) in vertical walls assuming closed gaps, separately for (1) facing facade, (2) opposite façade, and (3) lateral facades.

In said still more preferred embodiment, the calculation of the wind ( + X) in covers with closed gaps it is necessary to determine in addition to the dynamic pressure, the coefficient of exposure, which in this case is dependent on the average height of the cover; as well as two pressure coefficients per skirt, one negative and one positive, which can be combined to generate up to a total of 4 wind load hypotheses on decks, the decks being divided into 5 regions: 3 on the windward slope, and 2 on the lee side skirt. These pressure coefficients are dependent on the length of the nave, the heights of pillars, the length of light, and the percentage of slope, are calculated, according to the above, both in suction and pressure, and both windward as to leeward, resulting in four pressure coefficients that in turn determine four values of wind load, according to the scenario: windward / suction, leeward / suction, windward / pressure, and leeward / pressure.

In said still more preferred embodiment, the calculation of wind ( + X) in roofs with open gaps comprises the determination, in addition to the dynamic pressure, of the coefficient of exposure, and of the two pressure coefficients per skirt, one negative and one positive according to the previous case, of two internal pressure coefficients, one per skirt (to windward and leeward), so that four global pressure coefficients are obtained and, therefore, after multiplying them by the exposure coefficient and the dynamic pressure, four values of the wind loads (suction / windward, pressure / windward, suction / leeward) , and pressure / leeward).

According to the above, up to 8 wind modes can be taken into account for the + X direction (4 with open holes, and 4 with closed holes):

• Mode 1: Closed gaps, pressure on both flaps.

• Mode 2: Closed gaps, pressure-suction.

• Mode 3: Closed gaps, suction in both flaps.

• Mode 4: Closed gaps, suction-pressure.

• Mode 5: Open gaps, pressure on both flaps.

• Mode 6: Open gaps, pressure-suction.

• Mode 7: Open gaps, suction in both flaps.

• Mode 8: Open recesses, suction-pressure.

However, it is possible to reduce the number of wind modes to be considered for the X direction, for which it is necessary to analyze the aforementioned wind modes, paying particular attention to the most unfavorable, and more particularly to the so-called "mode" modes. 3 "(closed gaps, suction in both flaps) and" mode 5 "(open gaps, pressure in both flaps).

In this even more preferred embodiment in which the three possible directions of the wind are taken into consideration, the calculation of the wind ( + Y / -Y) in vertical faces, wind that will determine the dimensioning of the gable pillars and the elements of bracings of the ship, is determined similarly to the case of wind + X but "turning the building around", including the determination of, in addition to the values of dynamic pressure and exposure coefficient, which are the same as in the case of wind (+ X) in vertical walls, three coefficients of exterior pressure, depending on the length of light and height: One for the front facade, another for the rear facade, and another for the facades parallel to the wind.

In this even more preferred embodiment in which the three possible directions of the wind are taken into account, the calculation of the wind ( + Y / -Y) in roofs, wind that produces a suction effect in them and whose load is equal in both skirts, includes, in addition to the dynamic pressure, the coefficient of exposure, which in this case is dependent on the average height of the roof, the determination of an external pressure coefficient.

According to the above, up to a total of 4 wind modes can be taken into account for the direction + Y / -Y (2 with open holes, and 2 with closed holes):

• Mode 9: Wind + Y with closed gaps.

• Mode 10: Wind + Y with open holes.

• Mode 11: Wind -and with closed gaps.

• Mode 12: Wind - With open gaps.

However, it is possible to reduce the number of wind modes to be considered for the X direction, for which it is necessary to analyze the aforementioned wind modes, paying particular attention to those more unfavorable, and more particularly to the so-called "10 mode". (wind + Y with gaps open).

In said even more preferred embodiment, and for the particular case of non-symmetrical structures, the determination of the action of the wind on a vertical face or cover is made taking into consideration also the wind in -X.

In a preferred embodiment, the calculation of the seismic load or action is carried out by the simplified method of the equivalent static forces. In an even more preferred embodiment, said calculation comprises the determination of:

• Acceleration calculation,

• periods and vibration modes,

• the response coefficient, which depends on the level of ductility;

• the masses involved, and

• the equivalent static forces.

In an even more preferred embodiment, the determination of the calculation acceleration comprises the determination of:

• The basic acceleration,

• the dimensionless risk coefficient, which measures the probability that the building exceeds its calculation life;

• the terrain coefficient, and

• the amplification coefficient of the terrain.

In an even more preferred embodiment, the determination of the vibration periods and modes comprises the determination of:

• The characteristic periods of the response spectrum,

• the fundamental period of the building, which depends on the type of structure and the number of floors of the building, and on which the number of vibration modes depends; Y

• the value of the response spectrum.

In a preferred embodiment, the masses to be considered in the calculation are the own weight, the permanent loads, the snow overload, and the overload of use.

2. Design and calculation stage

The design and calculation stage of the impairment to be practiced in the structure includes both the calculation of the thermal load and the analysis of the behavior of both the weakly weakened structure and the theoretically weakened structure, depending on which the impairment is finally decided upon. implement in a practical way, what constitutes the last stage of the procedure object of the invention.

A thermal load or thermal stress is a force, solicitation or indirect action that appears in a resistant structure as a result of an impeded or conditioned dilation. That is to say, when applying heat to a resistant element, it undergoes temperature changes and deforms as a result of them, that deformation alters the distribution of tensions in the body. The result of the new distribution of stresses are loads and stresses not exerted directly by any external agent but which have an effect that can affect mechanical stability.

As indicated in section 3.4 of the Basic Document of Structural Safety - Construction Measures (DB SE-AE) of the CTE, the buildings and their elements are subject to deformations and geometric changes due to variations in the outside ambient temperature . The magnitude of the same depends on the climatic conditions of the place, the orientation and of the exhibition of the building, the characteristics of the constructive materials and of the finishes or coatings, and of the regime of heating and internal ventilation, as well as of the thermal insulation.

These variations of the temperature in the building lead to deformations of all the constructive elements, in particular, the structural ones, which, in cases where they are impeded, produce stresses in the affected elements.

On the other hand, the provision of expansion joints can help to reduce the effects of temperature variations. In normal buildings with concrete or steel structural elements, thermal actions may not be considered when expansion joints are arranged so that there are no continuous elements of more than 40 m in length.

In a preferred embodiment, the calculation of the thermal load in case of fire comprises the determination, for a given time, of the air temperature, depending on the location of the structure, and the temperature of the construction material of the structure. In a still more preferred embodiment, the determination of the air temperature comprises the determination of a maximum temperature, a minimum temperature, and a reference temperature.

In a preferred embodiment, the analysis of the behavior of the structure in case of fire is made according to Annex 8 ("Actions for thermal analysis) of the EAE-11 and therefore comprises:

• The selection of fire scenarios,

• The determination of the corresponding calculation fire action; the calculation fire action, or abbreviated "calculation fire", is defined by the curve of temperature increase of the gases of the fire room as a function of time, which is adopted to characterize the action of the fire. When selecting the calculation fire, it is possible to choose either an appropriate mathematical model of a real fire, or the normalized time-temperature curve that represents the thermal program of the test ovens.

• The calculation of the evolution of the temperature inside the structural elements as a consequence of their exposure to the calculation fire adopted. If a real fire model is chosen, the calculation must cover the entire duration of the fire, with the cooling phase included. If you opt for the normalized fire, in which there is no cooling phase, the time of exposure to mandatory fire should be set following the specifications of the regulations in force.

• The calculation of the mechanical behavior of the structure exposed to fire over a specific time interval.

In this way, to identify the accidental calculation situation, the appropriate calculation fire scenarios and the associated calculation fires are determined, based on a fire risk assessment. For each calculation fire scenario, a calculation fire is considered in a fire sector. The calculation fire applies only to one building fire sector at a time, except when specified otherwise in the fire scenario. For those structures for which national authorities specify fire resistance requirements, it can be assumed that the appropriate calculation fire is standard fire, except where otherwise specified.

On the other hand, depending on the calculation fire adopted, the following procedures are used:

• With a normalized time-temperature curve, the thermal analysis of the structural elements is applied for a specified period of time, without considering the cooling phase. This curve is defined by

9g = 20 345 log10 (8í 1) [° C] EC. one

Where:

Qg Temperature of the gas in the fire sector [° C]

T elapsed time [min]

The coefficient of heat transfer by convection is: ac = 25 W / m2 K.

• With a real or natural fire model, the thermal analysis of the structural elements is done for the entire duration of the fire, including the cooling phase. The models of real or natural fire are models that, with greater or lesser complexity, incorporate diverse physical parameters present in the development of a real fire. Simplified and advanced fire models are among the models of natural fire.

Simplified fire models are based on specific physical parameters with a limited field of application. When simplified fire models are used, the coefficient of heat transfer by convection will be ac = 35 W / m2 K, and gas temperatures will be adopted based on physical parameters, considering at least the fire load density and the ventilation conditions. When it is unlikely that the sudden widespread inflammation (flashover) is reached should take into account the relevant thermal actions to a localized fire in a non - uniform temperature versus time distribution is assumed, as opposed to the fires of sector.

Advanced fire models must take into account the properties of gas and the exchange of mass and energy. In particular, they should take into account the following gas properties, mass exchange, and energy exchange.

As regards the thermal analysis of an element, the position of the fire with respect to that element must be taken into account. For exterior elements, exposure to fire is considered through the openings in the facades and roofs. For the delimiting walls of a fire sector, it is considered, where appropriate, the exposure to a fire inside said sector and, alternatively, to an external fire in other fire sectors.

Also, to obtain the evolution of the temperature in the structure, it is necessary to distinguish between elements without protection and elements with protection, as it is considered by the EAE-11. The latter are usually made, for the case of steel, by application of intumescent paints, rock wool / plaster based mortars or installation of fibrosilicate plates Calcic, with a thickness that is provided by the manufacturer in his tests depending on the massiveness of the element to be protected.

With regard to mechanical analysis, the duration considered for such analysis must be the same as for thermal analysis. The verification of fire resistance can be done in one of the following ways:

• The calculation value of the fire resistance is greater than the required fire resistance time.

• The calculation value of the resistance of the element in fire situation at a time t, is greater than the calculation value of the relevant effects of the actions in fire situation at time t.

• The calculation value of the material temperature is lower than the calculation value of the critical temperature of the material.

When a structure is subject to fire, there is a considerable increase in temperature in it. This increase in temperature causes the mechanical properties of the structural elements to vary, such as the elastic limit and the modulus of elasticity, for which reason it is necessary to adopt the following correction coefficients of the mechanical characteristics of the structural steel, depending on the temperature reached for the same (0 ^a ):

KK ^ye

Ratio between the effective elastic limit for temperature (ea) and the elastic limit at 20 ° C.

Ky, e - fy, e / f and EC. two

^{KKE, e} Quotient between the modulus of elasticity in the linear phase of the stress-strain diagram, for the temperature (e ^a ) and the modulus of elasticity at 20 ° C.

K ^{E, e} - E ^{a, e} / E ^a EC. 3

The application of these coefficients is valid if the models of simplified calculation of the temperatures of the steel collected in the Instruction, or other procedures admitted by the same, are applied, but in this second case it must be verified that the speed of increase of temperature is maintained between the limits 2 <d ^ea / dt <50 ° C / minute.

On the other hand, the following parameter:

^{kkp, e} Quotient between the proportionality limit for temperature (ea) and the elastic limit at 20 ° C.

k ^{p, e} - f ^{p, e} / f ^and EC. 4

Together with the previous ones, it intervenes in the formulation of a tension (a) -formation (e) uniaxial diagram (figure 1) that can be adopted if advanced calculation methods are used.

In a preferred embodiment, the behavior of the structure in case of fire is analyzed by calculation of planar structures, more preferably by analysis of flat reticular structures, more preferably still by the rigidity method, considering that (1) the load is due to a linear variation of the temperature at the edge of the bar, and therefore is defined by its average value and its gradient along the edge, and (2) that the temperatures are uniform along the entire length of the bar, although it is possible to analyze said bar as if it were divided into several "virtual segments" so that the temperature will vary along the profile that forms the lintel.

In a preferred embodiment of the invention, the design and calculation stage of the weakening to be practiced in the structure of said industrial establishment is carried out taking into account the weakening method by reducing the profile section.

As indicated in the RSCIEI Application Guide, the protection systems of metallic structures are essentially based on the covering of the profiles with insulating materials.

Among the most used systems are the following:

• Fire resistant panels or panels, which are composed of calcium silicates or other materials. They are installed covering the entire perimeter of the metal profile and its thickness depends on the form factor, the thermal conductivity coefficient of the coating and the layout in the profile work, being able to reach fire resistance up to R 240.

• Intumescent paints, which are products that in contact with heat undergo a transformation due to chemical reactions, which prevents the transmission of heat to the element to be protected. The most usual thing is that resistance to fire up to R 60 (currently there are already paintings that reach an R90 if the mass is not very unfavorable). Keep in mind that this product is in full evolution. • Mortars, which are protective systems by covering the profile with mortar projection. Like the plates, the protection thickness will depend on the shape factor, the thermal conductivity coefficient of the coating and the layout of the profile, and fire resistance up to R 240 can be achieved.

The method of weakening by reducing the profile section consists of modifying (weakening) the structure at a certain point, in the least requested areas as far as stresses are concerned, modifying their dimensions, geometry, mechanical values, etc., in order to make it be precisely that point where the structure collapses, so that the collapse is "controlled".

To do this, first you have to check the structure, with a certain weakened area, in a conventional situation, that is, without a fire. Once it has been verified that the structure is within the limits of adequate resistance, we proceed to study the evolution of the behavior of the structure in case of fire, depending on the temperature that is reached at each moment, considering the alterations of the characteristics of the material due to the effect of fire.

As you can imagine, there are many ways to weaken a profile based on making cuts in it to decrease the section of it. Depending on where we make the cut, it will influence to a greater or lesser extent in its mechanical characteristics, and fundamentally in its moment of inertia. In reality, what must be done is to develop an iterative process in which one starts from uncertainty and the results are refined step by step, until the most optimal solution is reached. To a large extent, the number of iterations will depend on the previous experience of the engineer.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1. Tension diagram (a) - uniaxial deformation (e) for steel.

Figure 2. Pressure coefficients for wind (+ X) in vertical walls assuming closed gaps.

Figure 3. Pressure coefficients for wind ( + X) in roofs assuming closed gaps. Figure 4. Wind load values for wind (+ X), closed holes, pressure in both skirts.

Figure 5. Wind load values for wind (+ X), closed holes, pressure - suction. Figure 6. Wind load values for wind (+ X), closed holes, suction in both skirt.

Figure 7. Wind load values for wind (+ X), closed holes, suction - pressure. Figure 8. Values of wind loads for wind (+ X), open holes, pressure in both skirts.

Figure 9. Wind load values for wind (+ X), open holes, pressure - suction. Figure 10. Wind load values for wind (+ X), open holes, suction in both skirts.

Figure 11. Wind load values for wind (+ X), open holes, suction - pressure. Figure 12. Wind load values for wind (+ Y), closed holes.

Figure 13. Wind load values for wind (+ Y), open holes.

Figure 14. Wind load values for wind (-Y), closed holes.

Figure 15. Wind load values for wind (-Y), open holes.

Figure 16. Illustrative diagram of a ship attached to other ships on both sides and with a shared structure.

Figure 17. Illustrative diagram of a ship attached to another ship only on one side, being in contact with the outside on the free side.

Figure 18. Option 1 of weakening the profile.

Figure 19. Option 2 of profile weakening.

Figure 20. Option 3 of weakening the profile.

Figure 21. Case 1-0-0-3-0, mode 5 wind loads (but without lateral wind) and thermal load under conventional conditions.

Figure 22. Case 1-0-0-3-0, linear loads in the gantry.

PREFERRED EMBODIMENT OF THE INVENTION

Although the process object of the present invention must be particularized for each specific case, it is illustrated below by means of its application to an industrial building whose typology is quite common in the industrial areas of the city of Malaga, of its municipality, and very possibly of a large part of the national territory. Also note that the present embodiment of the invention aims to control the collapse in the event of fire of structures of industrial establishments in accordance with the legislation prevailing in Spain, especially with Royal Decree 2267/2004, of December 3, by the one that approves the Regulation of Fire Safety in the Industrial Establishments, Regulation that includes the typology of ships type A, B and C in relation to the establishment located in building, as well as the protection measures legally required for each one of said types A, B and C.

It is also noteworthy that, for the surface of the industrial establishment chosen, the option to qualify it as type A or type B is particularly relevant. For example, it is not the same to have to occupy a space of about 30 m2 in a warehouse of 350 m2 than in one of 1000 m2. The same goes for the investment, it is assumed that the initial investment for a larger vessel must also be higher than for a smaller vessel, so that the percentage of the total budget allocated to fire protection measures will be lower.

1. Characterization of the industrial establishment whose collapse in case of fire you want to control

1.1. Geographical, climatic and geometric characteristics

According to the above, having determined that the predominant ship in the industrial estates analyzed has a portico light comprised between 10 and 14 m and a height comprised between 5 and 8 m, depending on the polygon in question, it has been established as an industrial establishment type for the present example of realization a ship with the geographic and climatic characteristics, as well as geometrical, referenced in tables 1 and 2, respectively.

Table 1

Table 2

1.2. Actions in the building

Once the geometry of the ship is known, it is necessary to determine the actions that act on the structure.

1.2.1. Permanent loads

The permanent charges to consider will be:

• Galvanized steel sheet 1 mm thick: 0.09 kN / m2 «9 kg / m2

• Sheet anchors: we can consider 0.03 kN / m2 «3 kg / m2.

• Weight of the doors: we will consider 0.15 kN / m2 «15 kg / m2.

• Own weight of the structure: we approximate it to 0.33 kN / m2.

In this way, permanent loads will be 0.6 kN / m2.

1.2.2. Overload of snow

The snow overload is calculated from the data that appears in Annex E of the CTE-DB-SE. To specify by m2 of cover it is necessary to take into account the inclination of the same.

Sn = 0.2 kN / m 2 • cos 9.46 ° = 0.197 kN / m 2 EC. 5 Due to the shape of the cover, it is not necessary to take into account snow accumulation.

1.2.3. Wind action

In principle, the structure for three directions of the wind will be calculated: Wind + X (since the structure is symmetrical, it is not considered necessary to pose the wind in -X), wind + Y and wind -Y.

Wind (+ X) in vertical walls: assuming closed gaps.

This load acts on the four sides of the building. The steps to follow to calculate the vertical load will be:

• Dynamic pressure: As indicated in table 3.4 of section 3.3.3. from DB SE-AE, ANNEX D, ZONE A, we obtain a value of 0.42 kN / m2. It can also be obtained by looking at sheet AE.01, for zone A we have 0.42 kN / m2 (logically, this value will be the same for all building wind loads).

• Exposure coefficient: again on sheet AE.01, we look at the exposure coefficient for the average height of the enclosure (DB-SE-AE 3.3.3 Pto.1). The height of the enclosure under study is 7.00 meters, as we can see in the plans. For a degree of roughness IV (industrial environment), we obtain an exposure coefficient equal to 1.336.

• Pressure coefficient: it will take a different value depending on the wall being treated, and in the case of the lateral facades, the CTE distinguishes between three different zones. For simplicity, we will use a single coefficient of pressure per facing, calculated as the weighted average of the coefficients of each region. We can extract the average values of the external pressure coefficient of sheet AE.02:

o Frontage faced: for a height h = 8.00 meters, and 12 meters of light from the nave, it is obtained in the "Side façades" table of sheet AE.02, cp = 0.756

o Opposite facade: taking the same table as in the previous case, cp = -0.411.

o Lateral façades: in this case, the height / light ratio is equal to 8.00 / 12 = 0.67, and the slope of the roof is 15%. By consulting the table "Front / Back" of sheet AE.02, we obtain cp = -0.854.

Operating with the values obtained, the values referred to in table 3 are obtained and represented in figure 2.

Table 3

Wind ( + X) in vertical faces: assuming open gaps

Assuming open the gaps of the front facade, so that the internal pressure acts, the value of the internal pressure coefficient can be obtained in sheet AE.01 of the query. The slenderness of our building is 8/12 = 0.67 <1, and 100% of the holes are suctioned when the wind takes the + X direction . Thus, entering the corresponding table, the internal pressure coefficient is -0.5 (internal suction).

The pressure coefficients will be:

• Frontage faced: in this case they add: 0.756 - (-0.5). cp = 1,256

• Opposite facade: in this case subtract: -0.5 - (-0.411). cp = -0.089.

• Lateral façades: in this case subtract: -0,854 0.5. cp = -0.354.

Operating with the new coefficients, the values referred to in table 4 are obtained.

Table 4

Wind ( + X) on roofs with closed gaps

For the calculation of the wind load in roofs we will take the same value of the dynamic pressure: 0.42 kN / m2. Regarding the coefficient of exposure referred to the average height of the roof: 7.50 meters; interpolating between the values of table 3.4 of DB-SE-AE, we obtain ce = 1,588.

The DB-SE-AE defines in each skirt two pressure coefficients, one negative and one positive, which can be combined together to generate up to a total of 4 wind load hypotheses in decks.

For the calculation of the pressure coefficient, the CTE divides the roofs into 5 regions: 3 on the windward skirt, and 2 on the leeward skirt. Calculating the loads for each of the regions can be unnecessarily complicated, so we will use the tables in sheets AE.03a and AE.03b, which contain average values of the pressure coefficient for a 50 meter long ship, for heights of pillars of 5 and 10 meters.

Interpolating for a height h = 7.5m from the coefficients calculated for h = 5m and h = 10m, the pressure coefficients referred to in table 5 and represented in figure 3 are obtained, and from these pressure coefficients the wind loads referred to in table 6.

Table 6

Wind (+ X) in roofs with open holes

As with vertical wall loads, the internal pressure coefficients should be taken into account to consider the case of open gaps. Thus, considering the internal pressure coefficient equal to -0.5, the new global pressure coefficients are referred to in table 7, and multiplying them by the exposure coefficient and the dynamic pressure, obtain the values of the loads of wind referred to in table 8.

H = 7.5m

B arlo come to S o ta ve n to

Table 7

V iento (+ X) on decks

kN / m 2

Barlovento Sotavento

Suction -0.114 -0.121

Pressure 0.381 0.490

Table 8

Based on the calculations made in the previous cases, there are a total of 8 wind modes for the + X direction (4 with open holes, and 4 with closed holes):

• Closed gaps. Pressure on both flaps (mode 1, figure 4).

• Closed gaps. Pressure - Suction (mode 2, figure 5).

• Closed gaps. Suction in both flaps (mode 3, figure 6).

• Closed gaps. Suction - Pressure (mode 4, figure 7).

• Open gaps. Pressure on both flaps (mode 5, figure 8).

• Open gaps. Pressure - Suction (mode 6, figure 9).

• Open gaps. Suction in both flaps (mode 7, figure 10).

• Open gaps. Suction - Pressure (mode 8, figure 11).

This number of wind modes would generate a large number of very similar hypotheses and of little use. The ideal is to stay with two modes of wind only, since of the eight possible cases, most will be negligible. To choose the wind modes to be used in the calculation, we consider which are the most unfavorable:

• Greater load on the lintel: MODE 3.

• Largest arrow in the lintel: MODE 5, because the actions on the cover go in the same direction as the gravitational loads, and its effect is added.

• Greater load on a pillar: MODE 5, also 6-7-8, but mode 5 is also the one that causes the greatest arrow on the lintel.

• Greater horizontal displacement: MODE 2, because the sum of the actions has the greatest possible horizontal component.

Thus, in principle the most unfavorable wind modes are 2, 3 and 5. However, the importance of the criterion of horizontal displacement in industrial buildings is moderate (this would not happen, for example, in buildings with several floors), since the CTE only requires a limit of that displacement for almost permanent combinations. So in principle mode 2 will be neglected, and mode 3 and mode 5 will be studied.

Wind ( + Y / -Y) in vertical walls

The dynamic pressure is still 0.42 kN / m2, and the exposure coefficient is 1.336.

To obtain the coefficient of exterior pressure in the front and back facades, we will enter the table of Facades perpendicular to the wind TABLE AE.02 with a light of 30 meters and a height of 8.00 meters, which interpolating, gives coefficients of 0.705 on the front and -0.327 on the back.

The H / L ratio is now 8.00 / 30 = 0.267. Entering with this value in the table of Facades parallel to the wind we obtain, by interpolation, a value of the external pressure coefficient of -0.674

Thus, the wind loads + Y / -Y in the facades will take the values referred to in table 9.

Table 9

However, the internal pressure coefficient will vary depending on whether the wind blows in + Y or in -Y, since the ratio of suction holes / pressure gaps changes:

• Wind Y: The percentage of suction holes is 0, and the internal pressure coefficient is 0.705.

• Wind - Y: The percentage of suction holes is 100, and the internal pressure coefficient is -0.327.

Operating with the previous values, similar to how it was done with the wind + X, the values referred to in table 10 are obtained.

Table 10

Wind ( + Y / -Y) in Covers

The frontal wind produces a suction effect on the roofs, which is included in table D.6 b) of the DB-SE-AE. In sheet AE.04 of the handbook, the average values of the external pressure coefficient are tabulated for this case, which in industrial buildings will always be around -0.6.

If we observe in the aforementioned sheet the values we have for our case range from -0,576 of the ship with 5 m of height and 20% of slope to the 0,618 of the ship with 10 meters of height and 15 % dependent. In a simplified way, we will take cp = -0.6.

Thus, the wind load + Y / Y in the covers (same in both skirts) takes a value of:

q = 0.42 kN / m2 -1,336 - (- 0,6) = -0,337 kN / m2 (suction) EC. 6

Now, as in the previous case, of the wind And in facades, in the case of open gaps the values of the loads will change in the + Y / - directions , depending on the value of the internal pressure coefficient, the values are obtained referred to in table 11

Table 11

As in the case of wind + X, it is possible to distinguish a total of 4 modes of wind for the + Y direction (4 with open holes, and 4 with closed holes):

• + Y with closed gaps (mode 9, figure 12).

• + Y with open gaps (mode 10, figure 13).

• -And with closed gaps (mode 11, figure 14).

• -And with open holes (mode 12, figure 15).

And now we have to consider what modes to take:

• Greater load on the lintel: MODE 10 (it is logical that the greatest suction on the roof occurs when the wind enters through the holes).

• Greater load on the posterior gable portico: MODE 10

• Greater load on the front gable end: MODE 9

• Greater load on main pillars: MODE 10.

It is clear that the most unfavorable mode is No. 10.

1.2.4. Overload of Use

The overload of use in covers of this type is given by the table 3.1 of the DB-SE-AE. The overload of use takes a value of 0.4 kN / m2. But there is a note under said table that indicates that "this overload of use is not considered concomitant with the rest of variable actions".

So, it does not make sense to consider this load, if we are already taking into account an overload of snow of practically the same value, which is concomitant with the rest of variable actions.

1.2.5. Seismic action

The basic acceleration of Malaga is 0.11g and therefore, the NCSE-02 standard is applicable since it is greater than 0.04g and it is a building of normal importance , that is, its collapse due to an earthquake can cause casualties or damages To thirds. Therefore, it is mandatory to consider the seismic load in the calculation of the structure, although as we will see, in industrial buildings it is not usually a load of importance.

We will apply the simplified method of equivalent static forces to determine the seismic action applicable to the ship.

Acceleration of Calculation

To calculate the calculation acceleration we need to know:

• Basic acceleration (ab), which for Malaga is 0.11g.

• Adimensional risk coefficient (p), which measures the probability that the building exceeds its calculation life. For constructions of normal importance it is equal to 1.0.

• Soil coefficient (C), it will be considered a firm clay, which according to NCSE-02, Section 2.4, corresponds to a coefficient of 1.6. In fact all the strata of the terrain located 30 meters below the level of support should be taken into account, but in the absence of more data, we will stay with this one.

• Ground amplification coefficient (S), according to NCSE-022.2 is:

1.6 0.11

EC. 8 S = 3.33 • (1 •

EC. 8

1.25 9.8) = 1271

In this way, the seismic acceleration will be worth:

ac = S - p - ab = 1.271 • 1 • 0.11 = 0.140 EC. 9

Period and vibration modes

For the building response model, it is also necessary to know the characteristic periods of the response spectrum, TA and TB, which are valid (NCSE-022.3):

1-1.6

0.16 EC. 10

0.16 EC. 10

10

K • C 1-1.6

0.64 EC. 11

0.64 EC. eleven

2.5 = 2.5

The fundamental period of the building for the case of rigid laminated steel frames takes the value of TF = 0.11 n, with "n" being the number of floors of the building. Therefore, our fundamental period is TF = 0.11 s.

The value of the response spectrum ( a) is a function of the period a = a (T), and for values less than T ^a is equal to 1 + 1.5 T / TA, which for our case is:

0.11

a (0.11) 1 1.5 --- = 2.03

0.16 EC. 12

A single mode of vibration shall be considered, since TF is less than 0.75 s (NCSE-023.7.2.1).

Response coefficient B

This coefficient depends on the behavior of the building in the event of an earthquake, and is given by table 3.1 of the NCSE-02 according to the level of ductility adopted by the designer.

The level of ductility depends on the way in which the structure resists horizontal actions. In the direction X / -X, our ship resists the horizontal actions by means of porticos of rigid knots, corresponding therefore a very high level of ductility (^ = 4).

In the perpendicular direction it resists them by means of crosses of San Andrés, corresponding therefore a high level of ductility (^ = 3). Simplifying, and on the security side, we will have the most unfavorable value (^ = 3).

According to table 3.1 of the NCSE-02, the response coefficient will be B = 0.36.

Masses that are involved in the calculation.

The equivalent static forces depend on the masses of each floor of the building. In the case of our ship, the only plant it has is the deck, and its masses to be considered in the calculation are:

• Own weight: considering the own weight of the belts and the lintels, we can approximate it to about 0.30 kN / m2.

• Permanent loads: the only permanent load we have is the weight of the covering material: 0.09 kN / m2.

• Snow overload: not being applied more than 30 days, it is not necessary to take it into account.

• Overload of use: the use for maintenance is not included in the masses to be taken into account in the calculation (NCSE-023.2).

Therefore, the mass to be considered will be 0.39 kN / m2, which refers to the tax area of each portico (each load will be applied to one of the upper gantry nodes):

5 -12 EC. 13

60.83m2

cos 9.46

The mass of the enclosing walls is not considered acting on the supports for constructive considerations, since it has been appreciated that there are usually no rigid links between those and the supports, so that interaction will not occur.

Static Equivalent Forces

A horizontal load equal to:

pi = ^{~ '} pi ^' Pi ^' ai ^' P EC. fifteen _"Y

0.14 g

Fi = ----- - • 23.72 • 1 • 2.50 • 0.36 = 2.99kN EC 16

9

Where:

• ai takes the value 2.5 since the calculation period (T) is less than TB.

• nor is the distribution factor, which depends on the height of the plant considered in relation to the total of the building. For buildings of one floor, take a value equal to the unit.

The resulting load per gantry is negligible when compared to the wind load + X. In addition, the seismic load is considered accidental, and therefore has much less weight in the calculation of the structure, to be much lower safety coefficients that are applied in combinations of this type.

So in our calculation, the seismic load will not be taken into account.

2. Design and calculation stage

2.1. Actions by Thermal Load

In our case we can find different situations. In reality, the thermal load will depend on the location of the ship with respect to the total set of ships that share the same structure.

In this way, even if we find a ship or a set of them, that has of expansion joints at a distance less than or equal to 40 m, actually although the structure is allowed to move, this displacement is not completely free, but each lintel-pillar joint would behave as if a spring were available with a determined elastic constant.

In our case, in order to get as close as possible to reality, we should study the set of ships that share the structure. For this, it has been considered that as the length of the structure continues to not take into account the thermal load is 40m, it would be equivalent to 3 ships of 12m maximum light (3x12m = 36m), which on the other hand is a very common case in the construction of a set of ships with a shared structure.

For this, the following parameters will be taken into account:

• The reference temperature will be 10 ° C, as indicated in section 3.4.2. of the DB SE-AE.

• For the city of Malaga a minimum temperature of -6 ° C will be taken (section 2, Schedule E of the DB SE-AE).

• The maximum temperature will be 48 ° C (section 1, Schedule E of the DB SE-AE).

To check the structure, an interactive calculation program of flat structures "CESPLA" was carried out and registered by Mr. Juan Tomás Celigüeta, in collaboration with the Higher School of Engineers of San Sebastián, University of Navarra. Version 5.01 will be used, it is freely distributed and it has been downloaded from the TECNUN website of the Higher School of Engineers of the University of Navarra: http://www1.ceit.es/asiqnaturas/Estructuras1/Proqramas.htm.

The CESPLA program (Calculation of flat structures) performs the analysis of flat reticular structures of any kind, such as lattices, frames or beams. The program uses the method of rigidity, for its simplicity of programming and generality and is based on the theoretical foundations explained in the book Course of Structural Analysis (Juan Tomás Celigüeta, Ed. EUNSA). In fact, this program is a complementary element for the reader of this book.

2.2. Study of steel structure in front of the fire.

To study the behavior of a steel structure in the event of a fire, the provisions of Chapter XII of Royal Decree 751/2011, of May 27, which approves the Structural Steel Instruction ( EAE-) will be taken into account. eleven)

This chapter establishes the criteria to be applied in the project of steel building structures to verify their bearing capacity under the action of a fire, considered as an "accidental situation", for the purposes of structural security.

The action of fire or thermal action is defined by the flow of heat that impinges on the surfaces of the structure elements exposed to fire.

Depending on the "calculation fire" adopted, the following procedures should be used:

• With the temperature-temperature standard curve defined by CTE, the thermal analysis of the structural elements is carried out for a specified period of time.

• With another fire model, the thermal analysis of the structural elements is carried out for the complete fire process.

The procedures for checking the safety of fire steel structures explicitly included in the aforementioned Instruction belong to the category of calculation models cataloged as "simplified", which are calculation methods based on appropriate hypotheses for their application to elements simple structural, or small subsets of them.

Calculation Fire

For each calculation fire scenario, a calculation fire is considered in a fire sector, according to section 3 of the UNE-EN 1991-1-2 Standard.

The ac coefficients applicable to the normalized time-temperature curves are indicated in A8.3 of the EAE, defined by the chosen calculation fire.

Temperature of the element surface [° C]. It is obtained as a result of the thermal analysis of the element according to Chapter XII relative to the structural calculation in fire situation.

O Form factor; if specific data are lacking, O = 1.0 should be adopted. To take into account the effects of position and shadow, a lower value can be adopted.

£ m: Emissivity of the surface of the element, m = 0.7 will be adopted.

£ _ Emissivity of fire; generally, e_ = 1.0 is adopted.

The gas temperatures of the fire sector 6g can be adopted in the form of nominal time-temperature curves conforming to A8.3 of the EAE, or according to the natural fire models indicated in A8.6 and A8.7. Between the nominal time-temperature curves, in addition to the UNE-EN 1363 standardized curve, the external fire curve can be used to characterize the less severe fires produced in outdoor areas adjacent to the building, or to measure the effects on external elements of the buildings. flames coming out windows

For the present study, only a simplified fire model will be used, so to obtain more information on this type of fire model, it is necessary to go to what is indicated in section 3.3.2. of the Standard UNE-EN 1991-1-2: 2004.

Calculation of temperatures in steel

For the calculation of the temperature in the steel, the information indicated in section 48 of Royal Decree 751/2011, of May 27, which approves the Structural Steel Instruction ( EAE) will be taken into account.

• Elements without protection

pa, ca Density and specific heat of the steel defined in section 45.1 of the EAE expressed in kg / m3 and J / (kg ° K) respectively. As indicated in the following section, a specific heat will be taken for simplified procedures, which will be independent of the temperature, taking a value of ca = 600 J / (kg ° K)

For steel density a value of pa = 7850 kg / m2 will be taken

• Elements with protective coating

^{Pa, ca} Density and specific heat of the steel defined in section 45.1 of the EAE expressed in kg / m3 and J / (kg ° K):

ca = 600 J / (kg ° K)

For the steel density, a value of: Pa = 7850 kg / m3

* List of total heat capacities of the coating and the steel element, in cases a) and d) of section 48.3 of the SEA

V Pp, D • Cpo • dp • Sp / p aCa EC. 19

^Ppdcpd Calculation values for density and specific heat of the coating according to 48.3, in kg / m3 and J / (kg ° K).

^{rp, ef, d = rpief, k / Yp} Calculation value of the effective thermal resistivity of the cladding, in m2 ° K / W, with Yp given in 48.3 of the EAE-11

^{rp, ef, k- (1 ^ / 3) dp / Apk} Characteristic value of the effective thermal resistivity of the coating, in cases a) and d) of section 48.3. of the EAE

^{rp, ef, k} Value determined according to 48.4, in cases b) and d) of section 48.3 of the EAE-11.

^{0 g, t} Temperature of the gaseous mass (° C) defined in 43.2. of the EAE-11.

Variation of the mechanical properties of steel in case of fire

For the resistant checks in fire situation, Ym.a - 1 will be adopted as a partial coefficient for the strength of the steel.

For its application in the resistant checking procedures defined in chapter 13 of the EAE-11, the following correction coefficients of the mechanical characteristics of structural steel shall be adopted, depending on the temperature reached by it (0 ^a ):

Ky, e Quotient between the effective elastic limit for temperature (0a) and the elastic limit at 20 ° C.

K _{y, e} - f _{y, e} / f _and EC. twenty

K ^{e, 0} Quotient between the modulus of elasticity in the linear phase of the stress-strain diagram, for the temperature (0 ^a ) and the modulus of elasticity at 20 ° C.

K ^{e, 0} - E ^{a, e} / E ^a EC. twenty-one

The values of these coefficients are taken from Table 45.1 of the EAE-11, in which it is allowed to interpolate linearly. The application of these coefficients is valid if the models of simplified calculation of the temperatures of the steel collected in the Instruction, or other procedures admitted by the same, are applied, but in this second case it must be verified that the speed of temperature increase is maintained between the limits 2 <d ^ea / dt <50 ° C / minute.

In simplified procedures, a linear relationship between dilation and temperature can be considered using the coefficient:

Similarly, in simplified procedures, specific heat can be considered independent of temperature, taking the value:

And you can consider the thermal conductivity independent of the temperature, taking the value:

Although the standard gives characteristic values for each protective material, it is possible to take into account real values such as those provided by the manufacturers themselves in their tests.

2.3. Method of weakening by reducing the profile section

In previous sections, we have calculated all the actions that can act in a building, in order to evaluate them, to be able to make the corresponding verifications and thus, be able to justify that complies with the provisions of the CTE.

Of all the hypotheses, it will be studied, with the help of the aforementioned CESPLA program, which was described in the thermal actions section, only the most unfavorable ones, since they are the ones that will determine the compliance, or not, of the structural security.

As what is sought is to justify that the failure of a structure in the event of fire does not affect the structures of the adjoining buildings, we find 2 well differentiated cases and it is understood that they should be studied separately, since this is what to depend on the choice of the hypotheses studied previously:

• Case 1: ship attached to other ships on both sides with a shared structure. • Case 2: ship attached to another ship only on one side, being in contact with the outside on the other.

The gravitational loads have a value of 0.6 kN / m2, as indicated above. Due to the assumption that the structure is existing, and of certain antiquity, it will be considered that the structure will be formed by A42b profiles, although the calculation process with current profiles is identical, since the variations of the mechanical properties of the same are done by coefficients, as seen in a previous section.

Given the great variety of cases that can be studied, then a criterion will be stipulated to be numbered, so that seeing the number that corresponds to each case, you can know perfectly what are the conditions of it. In this way, a case called "Case 1-2-60-3-3" would mean " Case of attached house on both sides, with the structure weakened according to option 2 and with an elapsed time of 60 minutes of fire, 3 cm of protection in the lintel and 3 cm of protection in the weakened zone. "

Since the study, which in this work is represented is different for each type of ship that can be found, and the objective of it is to serve as a guide to be able to study each singular case that can be presented, only the case 1, that is, ships attached to both sides; particularly the cases 1-0-0-3-0 (on the basis of which the less requested areas of the lintel are determined, adequate to perform the weakening), 1-3-90-3-1.5 (in which an increase is observed of the tension by Von Mises but still below the elastic limit), and 1-3-90-3-1.5 (in which the elastic limit is already exceeded).

In relation to the weakening options analyzed, there are three:

• Option 1 (figure 18): To weaken the profile, begin by trimming the lower wings of the same, a distance of 3 cm on each side, so that the lower edge of the IPE would measure 6 cm instead of 12 cm of the normalized profile, as shown in the following figure.

• Option 2 (figure 19): In order to be able to weaken the profile much more in the modified area, we choose to make a more drastic cut in the geometry of it, so that it is as follows.

• Option 3 (figure 20): In order to weaken the profile in the modified area a little less, we choose to make a smaller cut in the profile, so that it looks like this.

In accordance with the above, we now proceed to comment on the different cases studied.

Case 1-0-0-3-0: ship attached to both sides, structure without weakening, without fire, 3 cm of protection around the lintel

The mode to be studied in this case will be the most unfavorable of those studied in the calculation section of the actions in the chosen model structure, that is, mode 5, but without lateral wind thermal load under conventional conditions (figure 21).

From these data we calculate the linear loads in the gantry, which are represented in figure 22:

0.381kN 101.97kg 1m

• 5m • = 1.94 kg / cm

m2 kN 100cm EC. 22 0.490kN 101.97kg 1m

• 5m • = 2.50 kg / cm

m2 kN 100cm EC. 23 0.600kN 101.97kg 1m

• 5m • = 3.06 kg / cm

m2 kN 100cm EC. 24

With these data, the model is introduced in the calculation program of structures, obtaining the bending moments, shear and deformations referred to in tables 12 and 13.

Due to the fact that said program does not provide partial data of the complete bar of the porch with sufficient detail, the lintel has been divided into segments approximately 1m long, and the loads indicated in figure 22 have been introduced. thermal load corresponding to the maximum temperature in the province of Malaga, of 48 ° C (section 1, Annex E of the DB SE-AE).

As indicated above, the mathematical model of this program to calculate the effects of thermal loads, is to consider that this load is due to a linear variation of the temperature at the edge of the bar, and therefore is defined for his mean value and its gradient along the edge. These temperatures are assumed to be uniform along the entire length of the bar, as indicated in section 4.2 of the UNE-EN 1993-1-2 Eurocode 3: Steel structures project.

Due to the fact that both existing and new structures usually have constructive solutions called brackets, at the junction of the pillar with the lintel, and since the program does not allow the introduction of a bar of variable section with respect to the x axis, it is opted for the introduction of a bar in the "segment 2 and 13" of a larger profile, in our case an IPE-300, consideration that would be on the side of safety, since the geometric characteristics of the gusset, these have for the most part part of the length a larger section, in addition to that the option chosen causes an increase in the effort in bars 3 and 12.

Table 12

Table 13

As is already known, the shear values are not significant with respect to the beveller, and as can be seen in the corresponding data tables, these values are irrelevant. From the data obtained it is clear that bars 4 and 11 are the least requested areas of the lintel, so a priori this would be the ideal area to make some weakening of the profile, since, in the most usual case (in which the section is constant along the entire bar) is the area in which a reduction of the section compromises to a lesser extent the safety of the bar for any conditions of use compatible with those existing at the time of analysis.

In the same way, both the minimum value of bending moment and tension by the Mises criterion occurs in bar 3.

Case 1-3-80-3-1.5: ship attached to both sides. Weakened structure - option 3, 80 minutes of fire, 1.5 cm of protection in the weakened area and 3 cm of protection in the rest of the lintel

Table 14 shows the values of the temperature at 80 minutes with 1.5 cm of protection in the weakened area and 3 cm of protection in the rest of the lintel.

Of

Table 14

With these values and the help of CESPLA, we obtain the values referred to in tables 15 and 16.

Table 16

It is observed that the tension has risen by Von Mises, but it is still below the elastic limit. The displacement of the pillar-lintel knots continues to increase.

Case 1-3-90-3-1.5: ship attached to both sides. Weakened structure - option 3, 90 minutes of fire, 1.5 cm of protection in the weakened area and 3 cm of protection in the rest of the lintel

The values of the temperature at 90 minutes with 1.5 cm of protection in the weakened area and 3 cm of protection in the rest of the lintel are shown in table 17.

Table 17

With these values and the help of CESPLA, we obtain the values referred to in tables 18 and 19.

Table 18

2,51 2.80 Si1-3-90-3-1.5 • 1039.29 129.68 1063.28 • 1756.52 190.53 ^{1787, 25}

2.51 2.80 Yes

Table 19

It is verified that both the bar 4 and the bar 11 exceed the elastic limit of the weakened area, which, as indicated in the previous table of parameters of the material as a function of temperature, was 1656 kg / cm2.

In this way, it is observed that the following conditions are met:

• The only point where the elastic limit of the material is exceeded and therefore where the collapse of the structure would occur is in the weakened zone.

• The collapse occurs after overcoming the 60 minutes required by the Royal Decree 2267/2004, of December 3, which approves the Regulations on Fire Safety in Industrial Establishments.

• At the moment in which the structure collapses, the displacement of the joints of the pillar-lintel joints have a deformation lower than the maximum indicated in section 4.3.3.2. of the DB-SE of the CTE on horizontal displacements. In said section it is indicated that when the appearance of the work is considered, it is admitted that the global structure has sufficient lateral rigidity, if before any combination of actions almost permanent, the relative collapse is less than 1/250.

We must also remember, as indicated above, the need to reduce the thickness of the protection to ensure that the temperature in the weakened area increases more quickly, achieving that the elastic limit of this area falls before the rest of the structure.

Once the results are observed, it is considered that the weakening option 3, and protecting the structure with 1.5 cm thickness of rock wool mortar in the weakened area and 3 cm thick in the rest of the porch, the conditions are met to be able to affirm that the collapse of the structure of the ship in study in case of fire would not affect the adjacent ships nor commit the sectorization with respect to these.

Conclusions of the weakening method by reducing the profile section

Although currently, administration technicians, as a general rule, require that the materials used for passive protection, and that are called "CE marked", must comply at least with the thickness indicated in the tests carried out by the own manufacturers, in this case, and given the nature of research study that has the present work, it is understood that such compliance would not be necessary, since analytical calculations are made for the calculation of steel temperature in case of fire with standardized fire, using for them the data of thermal conductivity, specific heat and density contributed by the own manufacturer in his technical sheet of product. In this way it is intended that the present study can become a possible reference when designing an alternative study of passive protection in industrial establishments.

In spite of what is indicated in the previous paragraph, it is necessary to insist that the only zone that is going to have a thickness less than that indicated in the tests provided by the manufacturer, is in the weakened zone, this area being no more than 10 cm long, besides having a time interval of approximately 30 minutes, since the weakened area collapses in the minute close to 90, when what is required is 60 minutes of fire resistance.

Therefore, as indicated above, it is demonstrated that the behavior of the structure in case of fire does not impair the stability or other conditions of the adjacent buildings, because the deformations are less than the maximum allowed by the CTE, what can be affirmed that the sectorization with respect to the adjacent ships is not compromised by the collapse of the structure. This statement is extended both to the boundaries walls of the establishment, as to the possible short range of fires for sectorization by roof, since this strip is usually fixed by screws to the wall itself, and not to the roof structure, method , which on the other hand, is the one indicated in the tests of the main manufacturers of materials for passive protection.

Claims (8)

1. Procedure for controlling the collapse of structures of industrial establishments in the event of fire, characterized in that it comprises the following stages: 1. Characterization of the industrial establishment whose collapse in case of fire is to be controlled, which includes the determination of the actions or charges that act on the structure of said establishment; 2. Design and calculation of the impairment to be practiced in the structure of said industrial establishment; and 3. Implementation of the selected impairment. Method according to the preceding claim, characterized in that the characterization stage comprises the determination of the following actions or charges: • Permanent loads (own weight of all elements (structure, facades, roofs, anchors, doors, etc.); • Snow overload (action on the roof that, in a horizontal terrain, is determined by the altitude and by the winter climate zone); • Action or wind load (action on roofs and walls); • Overload of use (take into consideration the weight of people on the cover); Y • Seismic action or load, which depends directly on the building's mass. Method according to the preceding claim, characterized in that the characterization of the action or load of the wind takes into consideration the following three wind directions: • Wind + X, which corresponds to the direction parallel to the porticos; • Wind + Y, perpendicular to the front facade; Y • Wind -Y, just like the wind Y, but on the rear facade. Method according to the preceding claim, characterized in that it comprises the determination and evaluation of the wind for the + X direction considering either open or closed gaps, either pressure or suction in both flaps, and either pressure-suction or suction-pressure. Method according to the preceding claim, characterized in that the determination and evaluation of the wind for the directions Y and -Y considering either open or closed gaps. Method according to any of claim 2 characterized in that the characterization of the seismic load or action comprises the determination of: • Acceleration calculation, • The periods and modes of vibration, • The response coefficient, which depends on the level of ductility; • The masses involved, and • The equivalent static forces. Method according to any of the preceding claims, characterized in that the design and calculation stage comprises both the calculation of the thermal load and the analysis of the behavior of the structure, which analysis includes: 1. The selection of fire scenarios, 2. The determination of the calculation fire action ("calculation fire"); 3. The calculation of the evolution of the temperature inside the structural elements as a consequence of their exposure to the "calculation fire" adopted; Y 4. The calculation of the mechanical behavior of the structure exposed to said "calculation fire" over a specific time interval. 8. Method according to the preceding claim characterized in that the design and calculation stage takes into account, for the design of the weakening, the method of weakening by reducing the profile section, which consists in modifying (weakening) the structure in the less requested areas in terms of supported stresses, modifying their characteristics (for example, dimensions, geometry, mechanical values, etc.).