JP6664777B1 - Method and system for creating state discrimination model of structure - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method and a system for creating a state determination model that can accurately determine whether the current state of the structure is normal or abnormal by inputting vibration data measured from the structure. Kind Code: A1 A method and system for creating a state determination model of a structure that inputs vibration data for a structure and diagnoses the vibration data. The system of the present invention comprises a data generating device having a built-in vibration model of a structure and a calculation unit including a model creating device for creating a state determination model, and a database for storing vibration data generated by the data generating device. The data generation device inputs normal vibration data and abnormal vibration data by inputting simulated vibration to the vibration model and performs calculation by vibration analysis, and transfers it to the database. A state discrimination model for a structure is created by applying machine learning including artificial intelligence and statistical methods to big data including. [Selection diagram] Fig. 1

Description

  The present invention relates to a method and a system for creating a state discrimination model of a structure for inputting vibration data for the structure and diagnosing the vibration data.

  BACKGROUND ART A structure such as a building may be deteriorated due to fatigue or aging due to stationary vibration generated by a transportation means or the like traveling nearby or non-stationary vibration such as an earthquake. Actual structure design is often performed using structural analysis using vibration data so as to withstand suddenly large vibrations such as deterioration due to aging and earthquakes for a long period of time.

  As a vibration test technology to acquire data for designing a large structure using analysis by numerical simulation for such a large structure, by combining structural analysis by numerical simulation with the measured value of vibration, 2. Description of the Related Art Techniques for predicting response characteristics and damage of structures are known. Design of specific structures by CAE using these prediction data has also been put to practical use.

  When designing a structure by CAE, in many cases, structural analysis and vibration analysis are performed using an analysis model of the structure created using CAD, and the analysis results are used as the design data of the actual structure. A method is adopted to reflect this. According to this method, it is possible to easily change the design such as the shape and the material without manufacturing an actual structure or a product imitating the structure.

  As a method of creating or improving such an analysis model, for example, Patent Document 1 discloses a process of dividing a model into elements of a predetermined size and an eigenvalue of a node coordinate vector related to a node coordinate vector related to an element of the model. A process of performing an analysis to calculate an eigenvector, a process of generating a design variable that determines a candidate for a shape change from the calculated eigenvector, and a process of performing a structural optimization analysis using the generated design variable. A structure optimization method in a design system for optimizing a structure by correcting a shape of a model having a process of outputting a result of the structure optimization is disclosed. According to this method, a computer automatically sets design variables when a design change is made, so that an optimum shape can be derived without depending on the skill of a designer.

  Further, in Patent Document 2, when the existing analysis model is a composite analysis model including a plurality of partial analysis models, any of the above partial analysis models is added to the composite analysis model as a partial analysis model to be changed. An adjacent part analysis model interlocking unit that performs an interlocking change process of adding a shape change in conjunction with the shape change to the adjacent part analysis model by reflecting the changed shape in the adjacent part analysis model adjacent to the change target partial analysis model An analysis model creation system for creating a target analysis model by adding a desired shape change to an existing analysis model is disclosed. According to this system, when creating a target composite analysis model using an existing analysis model, distortion of a mesh due to a partial shape change or the like is suppressed, and generation of the target analysis model is further reduced. It is said that it is possible to do it.

JP 2002-149717 A JP 2008-52336 A

  The creation of an analysis model using conventional CAD is often performed by using or referring to design data of a structure similar to the structure to be designed. For example, the design data and vibration analysis of a conventional structure are often used. A method of designing using big data in which data and the like are aggregated and applying a statistical method or the like can be considered. In an analysis model designed based on such conventional structure design data and vibration analysis data, measurement data in a state where the structure is operating or functioning normally is applied. Cannot be determined based on the calculation result when "." Is input.

  For this reason, in order to determine whether the current state of the structure is normal or abnormal based on the calculation result by the analysis model, specific measurement data at the time of abnormality such as the limit of failure of the structure is acquired, and the analysis model is used. A method of comparing the calculation result with the measurement data at the time of the abnormality can be considered. However, since it is extremely difficult to acquire data at the time of such an abnormality, the determination is made based on measurement data in a state where the structure is normally operating or functioning, and as a result, the structure There is a problem that the accuracy of determining the current state has to be sacrificed.

  Therefore, an object of the present invention is to provide a method and a system for creating a state determination model capable of accurately inputting vibration data measured from a structure and determining whether the current state of the structure is normal or abnormal. is there.

  In order to solve the above problems, a method for creating a state discrimination model of a structure for inputting vibration data for a structure and diagnosing the vibration data according to the present invention includes inputting simulated vibration to a vibration model of the structure. Normal vibration data and abnormal vibration data by performing calculation by vibration analysis, and then machine learning and statistical methods including artificial intelligence in the obtained big data including the normal vibration data and the abnormal vibration data Is applied to create a state determination model of the structure.

  Further, according to the present invention, a system for creating a state discrimination model of a structure for inputting vibration data for a structure and diagnosing the vibration data includes a data generation device incorporating the vibration model of the structure and a state discrimination model. An arithmetic unit including a model creation device to be created, and a database for storing the vibration data generated by the data generation device, wherein the data generation device inputs a simulated vibration to the vibration model and performs a vibration analysis. By performing the calculation, the normal vibration data and the abnormal vibration data are obtained and transferred to the database, and the model creation device performs machine learning and statistics including artificial intelligence in the big data including the normal vibration data and the abnormal vibration data. The method is characterized in that a state determination model for the structure is created by applying a dynamic method.

  According to this invention, machine learning and statistical methods including artificial intelligence are applied to big data including normal vibration data and abnormal vibration data of a structure obtained by vibration analysis, and the state of the structure is applied. By creating the discrimination model, the created state discrimination model can obtain a calculation result in consideration of the abnormal vibration data, so that it is possible to accurately determine whether the current state of the structure is normal or abnormal.

1 is a schematic diagram showing a system for creating a state determination model of a structure according to a representative example of the present invention. 2A and 2B are schematic diagrams illustrating a configuration of the analysis model illustrated in FIG. 1, wherein FIG. 2A illustrates a structural analysis model, and FIG. 2B illustrates a state determination model. It is a flowchart which shows the procedure which the creation method of the state discrimination model of a structure by a typical example of this invention performs. It is the schematic which shows the typical application example of the preparation system of the state discrimination model of a structure by this invention.

  Hereinafter, a method and system for creating a state determination model of a structure according to a representative example of the present invention will be described with reference to the drawings.

  FIG. 1 is a schematic diagram showing a system for creating a state determination model of a structure according to a representative example of the present invention. FIG. 2 is a schematic diagram showing the configuration of the analysis model shown in FIG. 1. FIG. 2A shows a structural analysis model, and FIG. 2B shows a state determination model. A system 100 for creating a state determination model of a structure according to a representative example of the present invention includes a vibration sensor 110 that measures vibration data of a structure to be measured and a state determination model for determining abnormality of the structure. An arithmetic unit 120 to be created and a database 130 to store vibration data are included. Further, as shown in FIG. 4 described later, the creation system 100 includes an input device 140 for setting a measurement start command and measurement conditions by the vibration sensor 110, and a measurement condition and a structure input by the input device 140. And a display device 150 that displays the various information and the determination result.

  As the vibration sensor 110, any of a type that directly attaches to a measurement point of a structure to measure vibration and a type that can detect the vibration without contacting the structure can be applied. A type in which vibration displacement is converted into an electric signal and detected is used. In addition, in the specific example shown in FIG. 1, the vibration sensor 110 transmits the measured vibration data to the arithmetic unit 120 via a connection line 112 obtained by adding unique information such as the acquisition position (location) and the measurement time.

  The arithmetic unit 120 includes a data generation device 122 including a structural analysis model SM simulating a structure to be measured, a model generation device 124 that generates a state determination model DM using vibration data stored in a database 130, including. The arithmetic unit 120 may include a central control device (not shown) that controls the operation of the entire system 100 for creating a state determination model of a structure illustrated in FIG.

  The database 130 is configured as a storage unit that stores the big data 132 including the normal vibration data 134 and the abnormal vibration data 136. The normal vibration data 134 and the abnormal vibration data 136 are those output by the structural analysis model SM of the data generator 122, actual vibration data acquired in the past for the same structure, or the target structure. And data generated or measured under various conditions or environments, such as vibration data when an abnormality such as a failure occurs in a structure similar to the above.

  Here, the database 130 stores, as the big data 132, the installation position of the vibration sensor 110 that measured the vibration data together with the vibration data, the abnormality occurrence position (range) when the determination result is abnormal, or the abnormality Information such as the degree of severity may be added and stored. These additional information are referred to when generating the normal vibration data 134 or abnormal vibration data 136 with the structural analysis model SM or when generating the state determination model DM.

  The data generation device 122 included in the arithmetic unit 120 inputs simulated vibration (for example, input data from the vibration sensor 110) to the structural analysis model SM, and outputs a result of a numerical operation to the database 130 as vibration data. Here, as shown in FIG. 2A, the structural analysis model SM is a model simulating a structure by a numerical analysis method such as a finite element method, and is based on a design drawing of the structure by CAE or the like. The structure includes a normal vibration analysis model SM1 in which there is no defect such as a defect in the structure and an abnormal vibration analysis model SM2 in which a defect such as a defect is intentionally formed in a part of the structure.

  The vibration data output from the structural analysis model SM of the data generation device 122 is big data including normal vibration data 134 based on the calculation result of the normal vibration analysis model SM1 and abnormal vibration data 136 based on the calculation result of the abnormal vibration analysis model SM2. 132 and transmitted to the database 130. The data generation device 122 may further have a function of adding information indicating the severity of the abnormality to the abnormal vibration data 136 when the output value of the abnormal vibration data 136 exceeds a predetermined threshold. At this time, data indicating the weight of the abnormality may be further added to the abnormal vibration data 136 according to the importance or the seriousness of the defective portion.

  Normally, the normal vibration data 134 is vibration data output when the structure is functioning or operating normally, and can be calculated using a model to which the design data of the structure is applied as it is. On the other hand, it is difficult to acquire the abnormal vibration data 136 without simulating a state where an abnormal location or an overload has occurred in the structure. Therefore, in the present invention, for example, a model in which a defective part is intentionally included in a part of the vibration model is created as the abnormal vibration analysis model SM2, and a simulated vibration is input to the model to perform a calculation by the vibration analysis. The abnormal vibration data 136 is obtained. At this time, the above-described defective portion may be provided at a plurality of locations in the vibration model.

  In addition, as a method of obtaining the abnormal vibration data 136, a calculation based on vibration analysis may be performed in a state where excessive or small simulated vibration is input to the abnormality diagnosis analysis model SM2. With these techniques, it is possible to simulate abnormal vibration data 136 of various patterns.

  On the other hand, the model creation device 124 accesses the database 130 described later, reads the normal vibration data 134 and the abnormal vibration data 136 from the big data 132, and performs machine learning and statistical methods including artificial intelligence on these vibration data. Is applied to create a state determination model DM corresponding to the structure. Here, in creating the state determination model DM, the model creation device 124 uses a big data 132 composed of a large amount of vibration data to perform a method based on statistical analysis, a data mining, a machine learning, a deep learning, and the like. Various methods such as a method using a neural network can be applied in combination. Further, the model creation device 124 may periodically access the big data 132 updated as needed to modify or update the state determination model DM.

  As shown in FIG. 2B, the state determination model DM includes a vibration propagation model DM1 for calculating output vibration data when the input vibration data propagates through the structure and outputs the vibration data. And a state output model DM2 for calculating a load, a displacement, and the like that the vibration gives to the structure when the light propagates. In this embodiment, the case where the vibration propagation model DM1 and the state output model DM2 are defined as separate models is illustrated. However, a model simulating a structure is shared, and the output vibration data described above is used for the model. And a function of calculating the state output of the structure may be provided.

  The vibration propagation model DM1 reads in normal vibration data 134 and abnormal vibration data 136 for a number of vibration data acquisition locations in advance, and generates a calculation rule by applying machine learning and statistical methods including artificial intelligence thereto. I do. As a result, for example, when some vibration data (vibration data measured by the vibration sensor 110) is input to a predetermined input position of a model imitating a structure, the vibration propagates through the structure and a predetermined The output vibration data when output to the output position is calculated and output. At this time, the vibration propagation model DM1 determines whether the calculated output vibration data is normal or abnormal depending on whether or not the calculated output vibration data exceeds a predetermined threshold value, or determines whether the normal vibration data 134 or the abnormal vibration data 134 already stored in the database 130. It may have a function of determining whether it is normal or abnormal in comparison with 136 and adding the information to the output vibration data.

  On the other hand, the state output model DM2 is obtained by giving the vibration data acquired from the large number of vibration sensors 110 to the vibration propagation model DM1 and performing the calculation, and the information of the individual vibration sensors stored in the database 130 (sensor information). Position, the degree of influence on the entire structure when it is determined to be abnormal vibration data, or additional information such as the severity of abnormal vibration data). A calculation rule is generated by applying a statistical method. As a result, for example, when the vibration data measured by the vibration sensor 110 is input, the input vibration data is referred to the unique information such as the measurement position and the measurement time, and the input vibration data is used to determine the shape and material of the structure. It calculates and outputs whether it represents a normal vibration state or an abnormal vibration state in light of the characteristics.

  At this time, the state output model DM2 includes information on the individual vibration sensors stored in the database 130 (the position of the sensor, the degree of influence on the entire structure when it is determined that the vibration data is abnormal, or the abnormal vibration data). In consideration of the additional information (e.g., the degree of seriousness of the structure), the site (range) where the abnormality occurs in the structure and the severity thereof are determined. Further, the state output model DM2 may be configured to have a function of simultaneously inputting vibration data at a plurality of positions and integrating the calculation results at the plurality of positions as an evaluation of the entire structure.

  FIG. 3 is a flowchart showing a procedure executed by a method for creating a state determination model of a structure according to a representative example of the present invention. As shown in FIG. 3, the arithmetic unit 120 first acquires simulated vibration data input to the vibration analysis model SM (Step S1). At this time, as the simulated vibration, for example, a vibration obtained from one or a plurality of vibration sensors 110 attached to a structure can be applied.

  Subsequently, the data generation device 122 of the arithmetic unit 120 reads out the normal vibration analysis model SM1 of the vibration analysis model SM (Step S2), inputs the simulated vibration data into the normal vibration analysis model SM1, and calculates the normal vibration data 134. And output it to the database 130 (step S3).

  Next, the data generation device 122 reads out the abnormal vibration analysis model SM2 of the vibration analysis model SM (Step S4), inputs the simulated vibration data to the abnormal vibration analysis model SM2, calculates the abnormal vibration data 136, and makes the database 130 (Step S5). At this time, a plurality of abnormal vibration models SM2 may be prepared, and a plurality of abnormal vibration data may be generated from one simulated vibration data. Then, the database 130 updates the data by adding the normal vibration data 134 and the abnormal vibration data 136 output from the data generating device 122 to the big data 132 (step S6).

  Subsequently, the arithmetic unit 120 reads the big data 132 including the normal vibration data 134 and the abnormal vibration data 136 from the database 130 into the model creation device 124 (Step S7). At this time, as the vibration data to be read, all of the big data 132 may be read, or, for example, data which has been subjected to predetermined filtering such as only data when vibration is input to a specific portion of a structure may be read.

  Next, the model creation device 124 applies a machine learning and a statistical method including artificial intelligence to the read normal vibration data 134 and abnormal vibration data 136, and thereby the state determination model DM corresponding to the structure. Is created (step S8), and the creation of the state determination model DM ends. At this time, if a state determination model DM is newly created for the target structure, this is newly saved, and if the state determination model DM that has already been created is held, Overwrite and save the newly created model. Thus, in the method for creating a state determination model of a structure according to the present invention, a state determination model that takes into account not only normal vibration data but also abnormal vibration data of a structure is created.

  FIG. 4 is a schematic diagram showing a typical application example of a system for creating a state determination model of a structure according to the present invention. As shown in FIG. 4, a system 100 for creating a state determination model of a structure according to the present invention is applied to a building such as a tunnel 10 or a building 20 or a transportation means such as a traveling bus 30. In this embodiment, the vibration sensor 110 of the creation system 100 is attached to a location where vibration can be transmitted and measured, such as the tunnel 10, the wall surface of the building 20, or the body of the bus 30.

  Note that the signal from the vibration sensor 110 may be transmitted by a wire via the arithmetic unit 120 and the connection line 112 as shown in FIG. 1, or may be transmitted by a wireless connection such as an electromagnetic wave as shown in FIG. Is also good. Thus, even when the vibration measurement position is located at a remote place distant from the arithmetic unit 120, a state determination model of the structure can be created using the vibration data measured at the remote place.

  The vibration data acquired by the vibration sensor 110 is transmitted to the arithmetic unit 120, and the arithmetic unit 120 that has received the vibration data executes the creation of the state determination model of the structure to be measured according to the procedure shown in FIG. I do. Here, the arithmetic unit 120 uses the created state determination model DM to diagnose whether the vibration data measured by the vibration sensor 110 is normal or abnormal, or uses the measured vibration data to construct a structure. It may be configured to further include a determination device (not shown) that determines the current state of the object. At this time, it is preferable that the discriminating device has a function of discriminating a site (range) where the abnormality occurs in the structure and its severity.

  Further, the arithmetic unit 120 may be configured to further include an alarm device (not shown) that issues an alarm when the vibration data diagnosed by the above-described determination device is determined to be abnormal. As the alarm device, for example, a sound device such as a buzzer or a siren, a display device such as a lamp or a monitor, or a combination thereof can be applied.

  Further, in the case of a transportation means or the like in which the measurement target moves, the state determination model creation system 100 shown in FIG. 4 may be configured to be mounted on a structure (a transportation means such as the bus 30). With such a configuration, even in an environment where it is difficult to transmit the measured vibration data to a remote location, the state determination model DM can be constantly created or updated based on the latest vibration data. .

  At this time, a lightweight configuration in which only the arithmetic unit 120 is mounted on the structure may be selected. With such a lightweight configuration, it can be configured as one function of an in-vehicle device or the like.

  With the above-described configuration, the method and system for creating a state determination model of a structure according to the present invention provide an artificial model based on big data including normal vibration data and abnormal vibration data stored in a database. A state discrimination model is created by using machine learning and statistical techniques including intelligence. Therefore, the created state determination model can accurately determine whether the current state of the structure is normal or abnormal by obtaining a calculation result in consideration of the abnormal vibration data.

  The embodiments according to the present invention and the modifications based thereon have been described above, but the present invention is not necessarily limited to these examples. Those skilled in the art will also be able to find various alternative embodiments and modifications without departing from the spirit of the invention or the scope of the appended claims.

  Further, in the specific examples described above, the case where the system for generating a state determination model of a structure according to the present invention is applied to a building, a vehicle, or the like is illustrated. The present invention may be applied to measurement of seismic vibration or vibration measurement of transportation equipment other than vehicles such as aircraft and ships.

10 Tunnel 20 Building 30 Bus 100 Structural state determination model creation system 110 Vibration sensor 112 Connection line 120 Operation unit 122 Data generation device 124 Model creation device 130 Database 132 Big data 134 Normal vibration data 136 Abnormal vibration data 140 Input device 150 Display device

Claims (10)

A method for creating a state discrimination model of a structure in which vibration data for a structure is input and diagnosis of the vibration data is performed,
Normal vibration data and abnormal vibration data are obtained by inputting simulated vibration to the vibration model of the structure and performing calculation by vibration analysis,
Thereafter, by applying machine learning and statistical techniques including artificial intelligence to the obtained big data including the normal vibration data and the abnormal vibration data, a state determination model of the structure is created. Method for creating a state determination model of a structure to be changed.   2. The structure according to claim 1, wherein the abnormal vibration data is obtained by performing a calculation based on vibration analysis in a state where a defective portion is intentionally included in a part of the vibration model. 3. A method for creating a state determination model of an object.   The method according to claim 2, wherein the defective portion is provided at a plurality of locations in the vibration model.   2. The state determination of the structure according to claim 1, wherein the abnormal vibration data is obtained by performing a calculation based on vibration analysis in a state where an excessively large or small simulated vibration is input to the vibration model. 3. Model creation method.   The method according to any one of claims 1 to 4, wherein when an output value of the abnormal vibration data exceeds a predetermined threshold, information indicating the severity of the abnormality is added to the abnormal vibration data. A method for creating a state determination model of a structure. A system for creating a state determination model of a structure for inputting vibration data for a structure and diagnosing the vibration data,
An arithmetic unit including a data generation device incorporating a vibration model of the structure and a model generation device for generating a state determination model, and a database for storing the vibration data generated by the data generation device,
The data generation device obtains normal vibration data and abnormal vibration data by inputting a simulated vibration to the vibration model and performing calculation by vibration analysis, and transfers the data to the database.
The model creation device creates a state determination model for the structure by applying a machine learning and statistical method including artificial intelligence to big data including the normal vibration data and the abnormal vibration data. A system for creating a state determination model of a structure.   7. The system according to claim 6, wherein the vibration model intentionally includes a defective part in a part thereof.   The system according to claim 7, wherein the defective portion is provided at a plurality of locations of the vibration model.   7. The structure according to claim 6, wherein the data generation device obtains the abnormal vibration data by performing a calculation based on vibration analysis in a state where excessive or small simulated vibration is input to the vibration model. A system for creating a state determination model.   7. The data generating apparatus according to claim 6, further comprising a function of adding information indicating a degree of abnormality to the abnormal vibration data when an output value of the abnormal vibration data exceeds a predetermined threshold. 10. The system for creating a state discrimination model of a structure according to any one of items 9 to 9.

Images