37 results about "Seismic engineering" patented technology

The utility model relates to a multifunctional bridge shock-absorbing and isolating support, which is composed of upper and lower parts. The upper part is composed of PTFE plate sliding bearings, which can realize the free expansion and displacement of the main beam structure under conventional loads, and thus have all the characteristics of conventional sliding bearings. The lower part is a combined shock-absorbing and isolating device with multi-level shock-absorbing and isolating functions, which is composed of a lead rubber pad in the middle part and an elastic-plastic energy-dissipating soft steel plate in the peripheral part. A lead core is arranged in the middle of the lead core rubber pad. Under earthquake action, lead rubber pads and elastic-plastic energy-dissipating steel plates are used to jointly dissipate energy to effectively limit the relative displacement between beams and piers. A rigid connection is made between the sliding support of the polytetrafluoroethylene plate and the combined shock-absorbing and isolating device to ensure the joint work of the upper and lower parts of the device. With the worldwide emphasis on the anti-seismic fortification of engineering structures and the continuous development of the economic level, this new type of bridge anti-seismic bearing will have a wide range of engineering application prospects.

The invention relates to a soil layer equivalent shear wave velocity and soil layer reflecting interface buried depth measuring method. The mathematical model of the relationship between soil layer shear wave velocity and buried depth and the fitting parameters are directly used to measure soil layer equivalent shear wave velocity and soil layer reflecting interface buried depth. Seismic engineering geological units are divided according to site seismic engineering geological conditions. In the same seismic engineering geological unit, soil layer shear wave velocity data are calculated to establish the mathematical model of the relationship between soil layer shear wave velocity and buried depth and the fitting parameters. The fitting parameters are substituted into a soil layer equivalent shear wave velocity calculation mode corresponding to the mathematical model of the relationship between soil layer shear wave velocity and buried depth to acquire the soil layer equivalent shear wave velocity of the seismic engineering geological unit. The fitting parameters are substituted into a soil layer reflecting interface buried depth calculation mode corresponding to the mathematical model of the relationship between soil layer shear wave velocity and buried depth to acquire the soil layer reflecting interface buried depth. According to the invention, the convenience and accuracy of soil layer equivalent shear wave velocity and soil layer reflecting interface buried depth measuring are improved.

The invention discloses an earthquake motion peak acceleration prediction method based on a second-order neuron deep neural network, belongs to the field of earthquake engineering, and aims to solve the technical problem of low precision of earthquake motion prediction. The earthquake motion peak acceleration prediction method comprises the following steps: 1, selecting earthquake magnitudes, projection distances, shear wave velocities, regions, cover layer thicknesses, fault types and periods as input parameters in a data set, and taking corresponding earthquake motion peak accelerations as target parameters; 2, establishing a deep neural network comprising three hidden layers, wherein neurons are second-order elements, a hyperbolic tangent function is adopted as an activation function, amean square error function and an Adam self-adaptive optimization function are adopted for back propagation, and an average absolute error function is adopted as an evaluation function; 3, training adeep neural network model; and 4, peak acceleration prediction. According to the method, a multi-input structure and a second-order neural network are adopted so that the precision of predicting theearthquake motion peak acceleration can be improved, and the applicability of the deep neural network model can be ensured.

The invention belongs to the technical field of structural engineering and seismic engineering and particularly relates to a dual-vibration table model test device capable of independently providing single-degree of freedom vertical torsional vibration load, which is composed of two sets of parallel vibration table systems capable of providing vertical torsional load. Each set of system is composed of parts such as a base, a fixing support, an actuator, a hinge system, a torsion transfer device and an acceleration sensor, and the two sets of parallel systems are connected and fixed through a connection auxiliary base. An actuator fixing support and a hinge support are welded on the base, one end of the actuator is provided with an acceleration control sensor, the other end is connected with the torsion transfer device, and thus, a vertical single-degree of freedom torsion function of a table surface can be realized. Two main body structures of a test object need to be firmly fixed on the table surfaces of the two vibration tables, the two actuators set output excitation forms, and through the collaborative effects of the whole device, the structural response, the mechanical properties and the damage form of the whole structure of the test object under effects of differential vertical torsional vibration can be measured.

The invention discloses an interpretable seismic oscillation parameter probability density distribution prediction method based on NGBoost and SHAP values, and relates to the technical field of seismic engineering, and the method comprises the following steps: 1, determining a research region, collecting seismic event waveforms and metadata information in the research region, carrying out the analysis and processing, and building a strong vibration database; 2, performing data cleaning on the strong vibration data, and selecting data for machine learning model training; 3, training a machine learning model for predicting the probability density distribution of the seismic oscillation parameters by using the selected strong vibration records and a natural gradient lifting algorithm; 4, calculating SHAP values of all features of all samples, analyzing importance of each feature according to the SHAP values, analyzing how to influence seismic oscillation parameter prediction, and explaining the machine learning model; and 5, predicting the seismic oscillation parameter probability density distribution of a newly occurring or hypothetical earthquake by using the trained machine learning model.

The invention provides high-expansion-energy steel slag concrete which comprises the following raw materials in parts by weight: 20-25 parts of cement, 12-15 parts of coarse steel slag, 40-45 parts ofbroken stone, 40-45 parts of steel slag sand, 10-14 parts of water and 0-5 parts of an additive. A preparation method comprises the following steps: selecting one group of samples from the three groups of samples as a reference mix proportion according to requirements; designing the mix proportion of the high-expansion-energy steel slag concrete by utilizing a strength design formula and an expansion rate design formula; comparing the design mix proportion with a reference mix proportion, and if the design mix proportion is far away from the reference mix proportion, adjusting the replacementamount of the steel slag sand in the high-expansion-energy steel slag concrete, the particle size range of the steel slag sand or the water cement ratio and other factors; and then preparing, pouringand curing the high-expansion-energy steel slag concrete. The high-expansion-energy steel slag concrete disclosed by the invention can be suitable for underground waterproof engineering, high-crack-resistance concrete engineering, constrained self-stress concrete engineering and anti-seismic engineering with high expansion rate and high strength requirements.

The invention provides a remote sensing based method for screening a three-dimensional seismic receiving combination central point. The method comprises the following steps: obtaining receiving goodness evaluation parameters according to an orthographic remote sensing image and a digital elevation model, determining the mandatory circumvent range according to the orthographic remote sensing image, wherein the receiving optimal evaluation parameters comprise high elevation, terrain slope and lithology classification; calculating theoretical receiving points according to seismic engineering acquisition parameters; determining whether the theoretical receiving points are enforceable receiving points or not according to the mandatory circumvent range; calculating the offset limiting parameters of the enforceable receiving points according to the seismic engineering acquisition parameters; determining whether a central point of a selectable receiving combination in the enforceable receiving points according to the offset limiting parameters; calculating the receiving goodness of the central point of the selectable receiving combination according to an optimized model corresponding to the receiving goodness evaluation parameters; calculating the combination expanding degree according to the combination area of the receiving combination and the most expanded area in proportion of combination elevation limit; screening the best receiving combination central point according to the receiving goodeness and the combination expanding degree.

The invention discloses a multi-working-condition one-dimensional wave equation solving method based on a neural network, belongs to the field of seismic engineering, and aims to solve the problem that time and labor are consumed when the neural network is used for solving one-dimensional wave equations under different working conditions. The one-dimensional wave equation solving method comprises the following steps: 1, selecting a one-dimensional wave equation as an equation to be solved; 2, determining a solution domain and the number of residues of an input variable; 3, establishing a full connection layer neural network comprising six hidden layers; 4, designing a specific loss function; and 5, performing pre-training and fine training of the neural network. The invention provides the one-dimensional wave equation solving method based on the neural network by taking the wave velocity as input, so that the model can learn the influence of different working conditions on an equation solution, and the interpretability of the solving method is also improved by adding a generic equation and a stress condition on the premise of keeping high solving precision.

The invention belongs to the technical field of earthquake engineering, and particularly relates to an earthquake casualty personnel gathering area assessment method based on unmanned aerial vehicle oblique photography. According to the method, an unmanned aerial vehicle with an inclined camera is used for performing information acquisition on the research area, earthquake prediction simulation is performed on the building in the research area in combination with the geographic position, the geological condition and the personnel distribution condition of the area, and casualties are estimated according to the damage state of the building and the basic information of the building. The casualty hot spots are obtained by performing hot spot analysis on the casualty distribution, so that the casualty gathering area in the whole area is obtained, and spatial guidance is provided for disaster prevention planning. According to the invention, building information is automatically identified and extracted, and technical support is provided for regional building information acquisition. For evaluation of regional casualties, the number of casualties caused by a disaster body and spatial distribution are considered, and a spatial correlation analysis method is introduced to analyze spatial distribution characteristics of the casualties.

The invention discloses an urban building group post-earthquake loss distribution calculation method based on a monitoring internet of things, and belongs to the technical field of earthquake engineering. Comprising the following steps: collecting single building information in a target city before an earthquake occurs, establishing a data information base of a city building group, and constructing a city building group numerical simulation model; a monitoring system of public buildings in a target city is incorporated into the Internet of Things, and the target city monitoring Internet of Things is built; after an earthquake occurs, earthquake loss calculation simulation is carried out on the urban building group according to actually measured earthquake waves; estimating the earthquake damage degree of the monitored building according to the building internal equipment and home decoration damage condition identified by the monitoring image; and estimating a corrected numerical simulation result based on the damage degree inferred by the monitoring image, and evaluating the damage state of other unmonitored buildings to obtain the post-earthquake loss distribution of the urban building group. According to the method, the earthquake damage distribution of the urban building group can be quickly estimated after an earthquake occurs, and important reference is provided for post-disaster rescue and emergency management.

The invention discloses ultrahigh-ductility concrete for seismic engineering and a preparation method of the ultrahigh-ductility concrete, and belongs to the technical field of building materials. Based on a micro-crack evolution effect theory, from the perspectives of improving the elasticity modulus of a reinforcing material and increasing the development scale of micro-cracks in concrete, the calcium carbonate whiskers with high strength and high elasticity modulus are introduced as an additional reinforcing material, and an inorganic gelling agent is used as an auxiliary material, the gaineffect of the calcium carbonate whiskers on the number of microcracks in the concrete is improved, so that the ultrahigh-ductility concrete material of which the ultimate tensile strain does not decrease or increase along with the increase of the strain rate is prepared, and the technical problem that the ultimate tensile strain of PEUHDC is reduced along with the increase of the strain rate is fundamentally solved.

The invention relates to a method for rapidly evaluating earthquake damage of a city building group and a bridge group. The invention relates to the technical field of earthquake engineering. According to the method, basic information of the building group and the bridge group is collected; according to the collected basic information of the building group and the bridge group, the earthquake resistance capacity of the building group and the bridge group is determined; the spectral acceleration is calculated according to acceleration time history data of seismic oscillation input by each building or bridge and basic information of the building or bridge; and the earthquake damage condition is evaluated according to the spectral acceleration calculated according to the input acceleration time history data of the earthquake oscillation and the earthquake resistance capability of the building or the bridge. The method is high in calculation efficiency, building group modeling and bridge group modeling in the city are not needed, the required basic information of the city building group and bridge group is less, and the method can be used for earthquake damage prediction and earthquake damage analysis of the city earthquake scene.

The invention belongs to the structural seismic engineering experimental technology field and particularly relates to a double-shake table model test device capable of independently providing a singleDOF horizontal torsional vibration load. The test device comprises a foundation and two sets of shake table systems, wherein the two sets of shake table systems comprise actuators, shake table surfaces, sliding rail sliding blocks, bayonet devices, torsion transmission devices and acceleration sensors, the actuators are fixed on the foundation through a fixing support, the shake table surfaces realize free rotation through the sliding rail sliding block system, one end of each of the actuators is connected with an acceleration sensor, the other end of each of the actuators is hinged with thetorsion transmission device, and the DOF torsion function of the two shake table surfaces is realized. The test device is advantaged in that two main structures of a test object are respectively fixedon the two shake table surfaces, setting excitation is respectively outputted by the two actuators, through the cooperation effect of the whole device, and structural response characteristics, node mechanical property and structural damage forms of the two main structures under different torsion vibration levels are measured.

The invention discloses an anti-seismic design method of a near-field region structure, and belongs to the field of seismic engineering. According to the method, the problem of poor safety of the near-field region structure due to the fact that the collapse reserve of the near-field region is smaller than that of the far-field region structure only by correcting the anti-seismic design spectrum toconsider the near-field pulse effect at present is solved. The invention relates to a near-field area structure anti-seismic design method for realizing collapse reserve equivalence by controlling the basic period of a structure. By adopting the method provided by the invention, the collapse reserve of the near-field region structure under the near-field pulse type earthquake action can be ensured, and the collapse reserve of the near-field area structure is equivalent to the collapse reserve of the far-field area structure under the action of far-field vibration, so that the safety of the near-field area structure is guaranteed, the urban earthquake safety of the near-field area in China is guaranteed, and the construction of earthquake-resistant tough cities is promoted. The method canbe applied to the field of seismic engineering.

The invention discloses a method for creating a finite element three-dimensional entity based on terrain remote sensing data, and belongs to the technical field of earthquake engineering. The method comprises the following specific steps: extracting latitude and longitude information-containing vector data from an integrated remote sensing data packet; converting the latitude and longitude information-containing vector data from a latitude and longitude coordinate format to a format which can be recognized by finite element software to obtain a finite element recognizable vector data format; reading finite element recognizable vector data, and transversely reading in the data point by point to generate scattered points; transversely scanning the scattered points in rows to generate a spline curve; longitudinally scanning the spline curve to generate a curved surface; generating a spare entity, penetrating the curved surface through the spare entity, combining the curved surface and the spare entity, and extracting a section, four surfaces below the section and connected with the section, and a bottom surface of the spare entity to generate a new entity by using the six closed curved surfaces, wherein the entity takes the section as an upper surface, namely the curved surfaces are embossed onto the three-dimensional entity; and gridding the obtained entity to achieve application of the remote sensing data in the finite element software.

The invention provides a building post-earthquake damage grade evaluation method based on deep learning, relates to the field of earthquake engineering, and is characterized in that damaged and damaged house images in an earthquake area are acquired, marked and a data set is established, and the buildings in the images are subjected to earthquake damage classification based on deep learning. According to the method, accurate damage classification can be quickly and efficiently performed on the building in the earthquake-bearing area, the calculation capability and the quick decision-making capability of the deep neural network can be utilized to replace on-site manual evaluation, time and energy consumption is greatly reduced, and under the condition that the accuracy is ensured, the damage condition of the building in a large-range earthquake-bearing area is efficiently judged.

The invention provides a three-component rotary seismograph, which comprises three fiber-optic gyroscopes and a circuit resolving module, wherein the sensitive axes of the three fiber-optic gyroscopesare mutually orthogonal; and the input end of the circuit resolving module is connected with the detection signal output ends of the fiber-optic gyroscopes, and the circuit resolving module is used for generating modulation signals required by the fiber-optic gyroscopes, demodulating the detection signals output by the detection signal output ends to obtain a detection angular speed and carryingout error compensation on the detection angular speed. The three identical fiber-optic gyroscopes are orthogonally assembled, and measurement errors caused by incomplete orthogonality of three sensitive axes are compensated. The invention provides the rotary seismograph which is high in precision, high in stability, small in angular velocity error and capable of measuring components in three orthogonal directions. Furthermore, when the rotary seismograph is applied to the monitoring of the seismic rotational motion, the indexes of the three rotational motions in the rotational seismology can be accurately and stably measured in real time, and the rotary seismograph has important guiding significance in the development of strong ground motion seismology, seismic engineering and seismic instruments.

The invention provides a six-component seismograph, which comprises three fiber-optic gyroscopes, three accelerometers and a six-component signal processing module, and is characterized in that the sensitive axes of the three fiber-optic gyroscopes are mutually orthogonal, and the sensitive axes of the three accelerometers are mutually orthogonal; the sensitive axis of each fiber-optic gyroscope is parallel to or coincides with the sensitive axis of one accelerometer one by one; and a six-component signal processing module is used for obtaining detection angular velocities and detection translational accelerations in three orthogonal directions and performing error compensation on the detection angular velocities and the detection translational accelerations. Orthogonal assembly of three identical fiber-optic gyroscopes and a three-axis accelerometer are assembled together, and measurement errors caused by incomplete orthogonality of three sensitive axes are compensated. The seismograph is high in precision, high in stability, small in error and capable of measuring angular velocities (rotation components) and translation accelerations (translation components) in three orthogonal directions, and has important guiding significance in development of strong ground kinematics seismology, earthquake engineering and seismographs.

The invention relates to a method for preparing high-toughness cement-based composite materials from waste plastics, belonging to the field of civil engineering and building materials. Disclosed is a high-toughness cement-based composite material prepared from waste plastics with high interfacial composite strength of plastics and cement and a toughening effect. The mass components of the cement-based composite material include: 100 parts of cement, 20-80 parts of fine sand, 30-50 parts of water, 1-20 parts of waste plastics, 0.3-0.8 parts of water reducing agent, and 0.095-0.15 parts of viscosity regulator . Among them, waste plastics are made into corrugated perforated sheets. The wavy structure makes the plastic sheet and the cement form a mechanical engagement in the transverse direction, and the holes on the plastic sheet make the cement and the plastic form a mechanical engagement in the thickness direction. Compared with traditional cement, it has higher impact strength and bending strength. The large tensile strain changes the failure mode of cement from brittle failure to ductile failure, and has good application prospects in the fields of earthquake resistance engineering and guardrail engineering.

The invention discloses a novel bearing isolation structure and an installation method thereof, and belongs to the technical field of anti-seismic engineering of buildings. The novel bearing isolationstructure mainly comprises a top plate, a bottom plate, an isolation component and an elastic ball, wherein the isolation component is installed between the top plate and floor, and the isolation component comprises a fixed rod, an upper panel and a groove; the upper panel is installed on the upper surface of the isolation component, the upper panel is provided with a circular hole, and the groove is hemispherical and a groove opening is fixedly connected with the circular hole; second spring fixed holes are separately formed in four corners of the upper surface of the upper panel, and firsthinged supports are separately arranged in four corners of the lower surface of the upper panel; and the fixed rod is fixedly mounted under the groove, a first bolt hole is formed in the fixed rod, the elastic ball is arranged in the groove, a spherical cavity is arranged in the elastic ball, and active oil is injected into the spherical cavity. In summary, the novel bearing isolation structure has the advantages of reasonable structure, low construction cost, good seismic isolation effect and the like.

The invention relates to a method for determining a post-earthquake restoration scheme of a subway underground station, which belongs to the field of earthquake engineering. The invention solves the problem that there is no relevant dynamic visualization technology to show the post-earthquake state in the prior art, which leads to the problem that it is difficult to determine the repair plan for the post-earthquake state of the subway station. Including: establishing the BIM model of the subway station; defining the post-earthquake state of the components in the BIM model; obtaining the post-earthquake initial state of each component of the subway station, and calculating and collecting the repair cost and repair time required for each component in each damage state; Import the defined post-earthquake initial state, repair cost and repair time into the BIM model for display; determine the repair plan based on the displayed post-earthquake initial state, repair cost and repair time; dynamically display the repair plan in the BIM model; repeat N Repeat the above steps to obtain the post-earthquake state of the subway underground station and the expected value of the repair time, which is used for the evaluation of the seismic resilience of the subway station. The invention is used for determining the post-earthquake restoration scheme of the subway station.

The invention discloses a measuring point coordinate processing device for three-dimensional seismic engineering exploration. The device comprises a total mounting disc, wherein cutting assemblies are uniformly arranged at the bottom of the total mounting disc, a drilling opening penetrating through the side wall is formed in the center of the total mounting disc, a regulation and control groove is formed in the inner wall of the drilling opening, a detection assembly is arranged in the regulation and control groove, a plurality of rectangular openings are formed in the total mounting disc, the inner wall of the rectangular opening is rotationally connected with a lead screw column, the lead screw column is connected with an arc-shaped connecting ring through a joint control assembly, connecting pieces used for mutual connection are arranged on the two sides of the arc-shaped connecting ring, and the arc-shaped connecting ring is connected with a drilling machine through an assembling ring. According to the invention, by arranging a drilling machine, a wave detector and a wire arranging cylinder on a total mounting disc, equipment is convenient to carry in the field, and when ground drilling is carried out, supporting of the drilling machine can be achieved by changing the total mounting disc, and the drilling machine can be carried to move downwards continuously, so that it is ensured that the drilling machine is fast, convenient and effective when drilling is carried out.

The invention discloses an ultra-high ductility concrete used in anti-seismic engineering and a preparation method thereof, belonging to the technical field of building materials. Based on the micro-crack evolution effect theory, the present invention starts from the perspective of improving the elastic modulus of the reinforcing material and increasing the development scale of the micro-cracks inside the concrete, and introduces calcium carbonate whiskers with high strength and high elastic modulus as an additional reinforcing material to assist Inorganic gelling agent is used to increase the gain effect of calcium carbonate whiskers on the number of microcracks inside concrete, thus preparing an ultra-high ductility concrete material in which the ultimate tensile strain does not decrease but rises with the increase of strain rate, fundamentally It solves the technical problem that the ultimate tensile strain of PE-UHDC decreases with the increase of strain rate.

The invention discloses a novel bearing isolation structure and an installation method thereof, and belongs to the technical field of anti-seismic engineering of buildings. The novel bearing isolationstructure mainly comprises a top plate, a bottom plate, an isolation component and an elastic ball, wherein the isolation component is installed between the top plate and floor, and the isolation component comprises a fixed rod, an upper panel and a groove; the upper panel is installed on the upper surface of the isolation component, the upper panel is provided with a circular hole, and the groove is hemispherical and a groove opening is fixedly connected with the circular hole; second spring fixed holes are separately formed in four corners of the upper surface of the upper panel, and firsthinged supports are separately arranged in four corners of the lower surface of the upper panel; and the fixed rod is fixedly mounted under the groove, a first bolt hole is formed in the fixed rod, the elastic ball is arranged in the groove, a spherical cavity is arranged in the elastic ball, and active oil is injected into the spherical cavity. In summary, the novel bearing isolation structure has the advantages of reasonable structure, low construction cost, good seismic isolation effect and the like.

The invention relates to an assembly-integrated-type subway station model structure and a manufacturing method thereof and belongs to the field of underground structure earthquake resistance and rocksoil earthquake engineering model testing. The structure comprises prefabricated model components, cast-in-situ model components, model rebars and equivalent connectors; the model rebars are arranged;the equivalent connectors are arranged on the model rebars; prefabricated components are manufactured; the negative-second-layer prefabricated model component is installed; negative-second-layer supports are arranged; a baseplate is cast in situ; middle beams are installed; negative-first-layer supports are arranged; the negative-first-layer prefabricated model component is installed; a middle plate is cast in situ; top beams are installed; a top plate is cast in situ. The assembly-integrated-type subway station model structure can be applied to scale model testing of an assembly-integrated-type subway station, has the advantages that the main structural characteristics of a prototype structure can be reflected and the structure is economical and convenient to manufacture, and has wide application prospects.

The method for predicting the peak acceleration of earthquake motion based on the deep neural network of second-order neurons belongs to the field of earthquake engineering, and it aims to solve the technical problem of low accuracy of earthquake motion prediction. Prediction method of peak ground acceleration: 1. Select magnitude, projection distance, shear wave velocity, area, overburden layer thickness, fault type and cycle as input parameters in the data set, and the corresponding peak ground acceleration as target parameter; 2. A deep neural network consisting of three hidden layers is established. The neurons are all second-order elements. The hyperbolic tangent function is used as the activation function. The mean square error function and Adam adaptive optimization function are used for backpropagation. The mean absolute error function is Evaluation function; 3. Deep neural network model training; 4. Peak acceleration prediction. The invention adopts a multi-input structure and a second-order neuron network, which can not only improve the accuracy of predicting the peak acceleration of earthquake, but also ensure the applicability of the deep neural network model.

The invention relates to a method for solving one-dimensional wave equations under different working conditions based on a neural network. The invention belongs to the field of earthquake engineering. One-dimensional wave equation solution method: 1. Select the 1-dimensional wave equation as the equation to be solved; 2. Determine the solution domain of input variables and the number of residual points; 3. Establish a fully connected layer neural network including 6 hidden layers; 4. Design a specific loss function; 5. Pre-training and refined training of the neural network. The present invention takes the wave velocity as an input, and proposes a neural network-based one-dimensional wave equation solution method, so that the model can learn the influence of different working conditions on the equation solution. On the premise of maintaining high solution accuracy, the general equation and stress condition The addition of also increases the interpretability of the solution method.

A shock-absorbing and energy-dissipating device for shield tunnel bolt joints belongs to the technical field of shield tunnel anti-seismic engineering. The shock absorption and energy dissipation device of the shield tunnel bolt joint includes a joint gasket, a first rubber gasket, a first sleeve, a second rubber gasket, a second sleeve and a third rubber gasket arranged coaxially in sequence. A sleeve is provided with a first elastic waterproof sealing ring and a second elastic waterproof sealing ring. The first sleeve is evenly provided with several protrusions along the circumferential direction, and each protrusion is equipped with a shock-absorbing spring. The springs are each provided with circular gaskets, the first rubber washer is provided with several through holes corresponding to the protrusions, the second sleeve is provided with a third step groove, and the third rubber washer is provided with a third step corresponding to the third step groove. There are three step protrusions, and the third step groove is provided with a third elastic waterproof sealing ring. The shock-absorbing and energy-dissipating device for shield tunnel bolt joints is easy to install, has strong shock-absorbing and energy-dissipating effects, can effectively attenuate vibrations, and also has a good waterproof effect.

The present invention proposes a method for classifying earthquake damage levels of buildings based on deep learning, which relates to the field of earthquake engineering. First, based on the shooting of individual buildings in the area, the earthquake damage and damage images are obtained, and then the nerves are segmented through images based on deep learning. The network quickly and accurately segments the buildings in the captured image, that is, extracts the earthquake-affected buildings from the complex image background; and then uses the image classification neural network based on deep learning to classify the earthquake damage of the segmented buildings. The invention can quickly and efficiently classify the damage of buildings in the earthquake-stricken area accurately. The image classification method based on deep learning can use the computing power and fast decision-making ability of the deep neural network to replace manual evaluation on site and greatly reduce the time. And energy consumption, under the condition of ensuring the accuracy rate, it can efficiently judge whether the buildings in the large-scale earthquake-stricken area collapse.