JP2017090238A - Prediction method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To predict the shape of a metal structure after being corroded with the passage of time and determine life with high accuracy.SOLUTION: According to the present invention, a derivation unit derives the secular change function of a parameter distribution function showing the distribution of prescribed parameter values that represent a surface shape in a discretionary area on the basis of surface shape data in a discretionary area on the surface of each of a plurality of metal samples buried in the ground and differing in elapsed years (steps S101-S104), and a generation unit arranges the values of parameters in a discretionary area in accordance with the parameter distribution function in a prescribed number of elapsed years derived from the secular change function of the parameter distribution function, and generates a shape model representing the predicted surface shapes of the metal samples in the elapsed years (steps S105-S107).SELECTED DRAWING: Figure 11

Description

  The present invention relates to a prediction method.

  Metal structures represented by infrastructure equipment such as steel pipe columns, supporting anchors and piping are used in a state where the whole or a part thereof is buried in the ground. Usually, metal structures corrode because they come into contact with soil or groundwater, and the thickness of the metal structures decreases with age (see Non-Patent Documents 1 and 2).

  For metal structures, it is necessary to take some measures such as repairs and replacements before the functions that should originally be secured are no longer exhibited. However, since underground metal structures are often difficult to inspect directly by humans or machines, metal structures that are partially embedded in the ground are only checked on the ground or on the ground. In many cases, the underground part was not confirmed. For this reason, it has been difficult to appropriately determine the time for exchanging the metal structure with the underground part buried in the ground.

  As a method for confirming the corrosion state of the underground part, for example, a method is known in which leakage of a fluid flowing in a pipe pipe is detected to identify a corroded part. However, this method is only a post-conservation method.

  In addition, as a method of confirming the corrosion state of other underground parts, a method of nondestructively inspecting using a percussion method or an ultrasonic method is also known. However, these methods are not sufficiently general and are not general-purpose methods because the accuracy is not sufficient and the inspection object is limited to a simple structure.

  Furthermore, a method of predicting the corrosion state from the environment, the elapsed years, etc. is known instead of directly confirming the corrosion state of the underground part. For example, for equipment such as piping that is determined to have a life when a through-hole is opened due to corrosion, the thickness of the through-hole can be determined by dividing the thickness of the equipment at a corrosion rate that is empirically specified or calculated by experiment or analysis. The time of occurrence, that is, the life of the equipment is estimated. However, this method has very low accuracy because it greatly simplifies the corrosion mechanism.

  A method for estimating the lifetime of equipment such as piping by means of various experiments and extreme statistical analysis using corrosion data of actual machines is also known. Extremum statistical analysis estimates the maximum local corrosion depth in a large area using the maximum local corrosion depth collected from a small-area measurement section of a sample. A highly accurate estimate is obtained. However, this extreme value statistical analysis is also applied only when the criteria for determining the lifetime in the shape of the facility is clear, such as when a through hole is opened. Non-Patent Document 3 discloses an algorithm for generating a surface shape by numerical calculation.

Morio Kadoi, Shoaki Takahashi, Kotaro Yano, "Studies on soil corrosion of metal materials (1st report)", Corrosion protection technology, 1967, Vol.16, No.6, pp.10-18 Yoshikazu Miyata, Shuji Asakura, “Measurement of Soil Corrosion Focusing on Electrochemical Method (Part 2)”, Materials and Environment, 1997, Vol.46, pp.610-619 Michimasa Uchinada, “Generation and Application of Practical Surface Topography Data”, Journal of Japan Society for Precision Engineering, 2010, Vol.76, No.9, pp.1003-1006

  However, in some metal structures, the criteria for determining the lifetime in terms of shape are often not clear. For example, a pole anchor or the like that supports a power pole or the like is not immediately determined to have a life even if a through hole is opened in a certain portion of the underground portion. For such metal structures, the allowable shape varies depending on the installation conditions such as soil and tension, and the surface shape such as uneven shape and distribution as well as average thickness reduction and through hole diameter and distribution Etc. are judged. As described above, when the lifetime of the metal structure is determined based on various conditions of the shape of the metal structure after corrosion with the lapse of years, it is difficult to predict the lifetime.

  The present invention has been made in view of the above, and an object of the present invention is to predict the shape of a metal structure after corrosion with the lapse of years and determine the life with high accuracy.

  In order to solve the above-described problems and achieve the object, the prediction method according to the present invention is applied to surface shape data in an arbitrary region of each surface of a plurality of metal samples embedded in the ground and having different elapsed years. A derivation step of deriving a function of change over time of a parameter distribution function representing a distribution of values of predetermined parameters representing a surface shape in an arbitrary region, and a predetermined function derived from the function of change over time of the parameter distribution function Generating a shape model that represents the predicted shape of the surface of the metal sample in the elapsed years by arranging the value of the parameter in an arbitrary region according to the parameter distribution function in the elapsed years, To do.

  ADVANTAGE OF THE INVENTION According to this invention, the shape of the metal structure after corrosion with the elapsed years can be estimated, and a lifetime can be determined with high precision.

FIG. 1 is a schematic diagram showing a schematic configuration of a prediction apparatus according to an embodiment of the present invention. FIG. 2 is an explanatory diagram for explaining processing of the derivation unit of the present embodiment. FIG. 3 is an explanatory diagram for explaining processing of the derivation unit of the present embodiment. FIG. 4 is an explanatory diagram for explaining processing of the derivation unit of the present embodiment. FIG. 5 is an explanatory diagram for explaining processing of the derivation unit of the present embodiment. FIG. 6 is an explanatory diagram for explaining the processing of the derivation unit of the present embodiment. FIG. 7 is an explanatory diagram for explaining processing of the generation unit of the present embodiment. FIG. 8 is an explanatory diagram for explaining the processing of the generation unit of the present embodiment. FIG. 9 is an explanatory diagram for explaining processing of the generation unit of the present embodiment. FIG. 10 is an explanatory diagram for explaining processing of the generation unit of the present embodiment. FIG. 11 is a flowchart showing the prediction processing procedure of the present embodiment.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited by this embodiment. Moreover, in description of drawing, the same code | symbol is attached | subjected and shown to the same part.

[Configuration of prediction device]
FIG. 1 is a schematic diagram illustrating a schematic configuration of a prediction apparatus according to the present embodiment. As illustrated in FIG. 1, the prediction device 1 is realized by a general-purpose computer such as a personal computer, and includes an input unit 11, an output unit 12, a storage unit 13, and a control unit 14.

  The input unit 11 is realized by using an input device such as a mouse or a keyboard, and inputs various instruction information, measurement data, and the like to the control unit 14 in response to an input operation by the operator. The output unit 12 is realized by a display device such as a liquid crystal display, a printing device such as a printer, an information communication device, and the like, and outputs a shape model generated by a prediction process described later.

  The storage unit 13 is realized by a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. Stores the function of the derived parameter distribution function over time. The storage unit 13 may be configured to communicate with the control unit 14 via an electric communication line such as a LAN or the Internet.

  The control unit 14 is realized by an arithmetic processing device such as a CPU (Central Processing Unit). The prediction device 1 executes a prediction process, which will be described later, under the control of the control unit 14, and generates a shape model that represents the predicted shape of the metal sample in an arbitrary age. The control unit 14 functions as the derivation unit 141 and the generation unit 142 by executing the processing program stored in the memory.

  The deriving unit 141 has a predetermined parameter value representing a surface shape in an arbitrary region based on surface shape data in an arbitrary region of each surface of a plurality of metal samples with different elapsed years embedded in the ground. A function of change over time of the parameter distribution function representing the distribution is derived.

  Specifically, the deriving unit 141 first acquires, via the input unit 11, surface shape data of an arbitrary region of the surface obtained for each of the metal samples embedded in the ground and having different elapsed years.

  Here, although each metal sample is not particularly limited, it is preferable that the material and the shape are similar to each other and the metal structure to be predicted, and only the elapsed years embedded in the ground are different. That is, it is preferable that the material and shape of each metal sample and the embedded soil condition and installation environment are close to the actual environment of the metal structure to be predicted. Further, it is preferable that the values of the elapsed years are distributed from relatively small new ones to large old ones without bias. This increases the reliability of the prediction process.

  Moreover, it is preferable that the surface area of the arbitrary area | region of the surface of a metal sample is about 1% or more of the surface area of a metal sample. The shape data is represented by, for example, a three-dimensional coordinate position. That is, the shape data is acquired using, for example, a laser displacement meter, and, as illustrated in FIG. 2, at a coordinate point (x, y) on the XY plane represented by the orthogonal X direction axis and Y direction axis. It is represented by height z (x, y). Hereinafter, an area on the XY plane corresponding to an arbitrary area on the surface is referred to as a data acquisition area.

  Next, the deriving unit 141 divides the data acquisition area into predetermined unit areas, and calculates predetermined parameter values for each unit area. Here, the shape of the unit region is not limited, but the number of divisions is preferably increased to some extent. For example, as illustrated in FIG. 2, the shape data is represented by the height z (x, y) at the coordinate point (x, y) on the XY plane represented by the orthogonal X direction axis and Y direction axis. In this case, a unit section on the XY plane is set as a unit area.

  The parameter means a parameter representing the surface shape of the metal sample. Examples of parameters include average plate thickness, maximum peak height, maximum valley depth, arithmetic average height, root mean square height, skewness value, or height distribution information such as kurtosis calculated from shape data. Is done.

  In the present embodiment, a case will be described in which the parameter is a skewness value representing the surface unevenness state. The skewness value Ssk is calculated by the following equation (1) using the shape data, that is, the height z (x, y) at the coordinate point (x, y) on the XY plane. Here, A is the area of the data acquisition area.

  As illustrated in FIG. 3, the deriving unit 141 calculates a skewness value for each unit region. In the example of FIG. 3, the skewness value for one unit region is calculated as a, and the skewness value for another unit region is calculated as b.

  Next, the derivation unit 141 uses a parameter value calculated for each unit region of each metal sample as a regression function to represent a parameter distribution function that represents the distribution of parameter values in the data acquisition region of each metal sample. To derive.

  For example, the calculated skewness value for the data acquisition area is distributed like a histogram illustrated in FIG. Further, it is known that the skewness value is distributed according to a normal distribution function expressed by the following equation (2) as shown by a curve in FIG. Here, μ is an average, and σ is a standard deviation.

  Therefore, the deriving unit 141 recursively specifies the average μ and the standard deviation σ of the above equation (2) using the histogram of the calculated skewness value illustrated in FIG. In this way, the deriving unit 141 derives a skewness value distribution function. The function form of the parameter distribution function is not limited to the normal distribution function, and is determined for each parameter.

  Next, the deriving unit 141 derives a function of change over time of the parameter distribution function, that is, a relational expression between the elapsed year and the parameter distribution function, using the parameter distribution function derived for each of the plurality of metal samples having different elapsed years. To do.

  Specifically, as illustrated in FIG. 5, the deriving unit 141 derives a skewness value distribution function according to the above procedure for each of the metal samples having different elapsed years T. In the example of FIG. 5, distribution functions of three skewness values with elapsed years x1, x2, and x3 are derived. Then, the derivation unit 141 derives the relationship between the elapsed years T and the parameter distribution function, as illustrated in FIG. In the example of FIG. 6, the average μ of the skewness value distribution function over time is obtained from three points indicating the relationship between the elapsed years (x1, x2, x3) and the average of the three skewness value distribution functions illustrated in FIG. Assuming that the change is a straight line, the following equation (3) is derived as a relational expression between the elapsed years T and the average μ. Here, α and β are constants.

  Similarly, the deriving unit 141 derives a relational expression between the elapsed years T and the standard deviation σ of the skewness value distribution function. Thus, if the elapsed year T is specified, the mean μ and the standard deviation σ of the skewness value distribution function that changes with time are specified.

  Note that the number of parameters is not limited to one. When there are a plurality of types of parameters, a function of change over time of the parameter distribution function is derived by the above processing for each of the plurality of types of parameters.

  The generation unit 142 arranges the value of the parameter in an arbitrary region in accordance with the parameter distribution function in a predetermined elapsed time derived from the function of change over time of the parameter distribution function, and predicts the surface of the metal sample in the elapsed time A shape model representing the shape is generated.

  Specifically, the generation unit 142 first specifies the elapsed years T and specifies the parameter distribution function in the elapsed years T. Next, the generation unit 142 distributes parameter values in each unit region in an arbitrary region so as to follow the specified parameter distribution function. That is, the generation unit 142 specifies the frequency of the skewness values a, b, c, d, and e using the distribution function of the skewness value in the specified elapsed years x as illustrated by a curve in FIG. Then, as illustrated in FIG. 8, the generation unit 142 distributes the skewness values a, b, c, d, and e to each unit region in an arbitrary region according to the specified frequency.

  When the generating unit 142 arranges the parameter value in an arbitrary area, it may be randomly arranged using a random number or the like. For example, the parameter value b is arranged next to the parameter value c. You may arrange according to the rule. Regarding corrosion, for example, if the average plate thickness is small in a certain unit region, the value is often small in a unit region adjacent to the unit region. As described above, if the arrangement is performed according to a rule that takes into account the relevance of each parameter related to corrosion in adjacent unit regions, the accuracy of the prediction process is increased.

  As illustrated in FIG. 9, the generation unit 142 distributes parameter values for a plurality of types of parameters through a similar process. In the example of FIG. 9, a plurality of parameters including three types of average plate thickness, skewness value, and kurtosis are applied. Here, the average plate thickness a1, the skewness value b1, the kurtosis c1, etc. are arranged in a certain unit region, and the average plate thickness a2, the skewness value b2, the kurtosis c2, etc. are arranged in another unit region.

  Next, the production | generation part 142 produces | generates the shape model showing the predicted shape of the surface in the elapsed years T of a metal sample using the distributed parameter value. Specifically, the generation unit 142 generates a surface shape model as illustrated in FIG. 10 using the parameter values of the plurality of types of unit regions illustrated in FIG. For example, the generation unit 142 applies an algorithm such as a fractal model, a convolution with a coefficient matrix, or an inverse Fourier transform to each unit region using information such as a parameter value, the number of data points, and a pitch (Non-Patent Document 3). (Refer to) A three-dimensional shape model is generated. The generation unit 142 repeats this process for all unit regions in the region where the parameter values are arranged, and generates a three-dimensional shape model of an arbitrary region.

  Note that when the shape model is discontinuous between adjacent unit regions, the generation unit 142 may smooth the numerical values of the boundary portions of the unit regions to be continuous.

[Prediction processing]
Next, a prediction processing procedure in the prediction device 1 will be described with reference to the flowchart of FIG. First, the derivation unit 141 acquires surface shape data in each arbitrary region of a plurality of metal samples embedded in the ground with different elapsed years via the input unit 11 (step S101).

  Next, the deriving unit 141 divides a data acquisition region corresponding to an arbitrary region on the surface into predetermined unit regions using the shape data, and calculates a predetermined parameter value for each unit region (step S102). Then, the deriving unit 141 derives a parameter distribution function in the data acquisition region of each metal sample using the parameter value calculated for each unit region of each metal sample (step S103). Subsequently, the deriving unit 141 derives a function of temporal change of the parameter distribution function using the parameter distribution function derived for each of the plurality of metal samples having different elapsed years (step S104).

  Thereafter, the generation unit 142 specifies the elapsed years, and derives a parameter distribution function for the specified predetermined elapsed years (step S105). Then, the generation unit 142 arranges parameter values in each unit region in an arbitrary region so as to follow the derived parameter distribution function (step S106). Finally, the generation unit 142 generates a shape model that represents a predicted shape of an arbitrary region of the surface of the metal sample in a predetermined elapsed age using the arranged parameter values (step S107).

  As described above, in the prediction device 1 of the present embodiment, the derivation unit 141 is based on surface shape data in an arbitrary region of each surface of a plurality of metal samples embedded in the ground and having different elapsed years. Then, a function of change over time of a parameter distribution function representing a distribution of values of predetermined parameters representing the surface shape in an arbitrary region is derived. Further, the generating unit 142 arranges the value of the parameter in an arbitrary region in accordance with the parameter distribution function in a predetermined elapsed time derived from the function of change over time of the parameter distribution function, and the surface of the metal sample in the elapsed age A shape model representing the predicted shape is generated.

  Thereby, the shape of the metal structure after corrosion with the passage of years can be predicted. Also, using the generated shape model, for example, the resistance force against the tension of the metal structure is calculated by dynamic simulation, and the life of the metal structure is determined with high accuracy from the decrease in the resistance force. Evaluation can be made.

  As mentioned above, although embodiment which applied the invention made | formed by this inventor was described, this invention is not limited with the description and drawing which make a part of indication of this invention by this embodiment. That is, other embodiments, examples, operational techniques, and the like made by those skilled in the art based on this embodiment are all included in the scope of the present invention.

DESCRIPTION OF SYMBOLS 1 Prediction apparatus 11 Input part 12 Output part 13 Storage part 14 Control part 141 Derivation part 142 Generation part

Claims (3)

A parameter distribution that represents the distribution of the value of a predetermined parameter that represents the surface shape in an arbitrary region based on the surface shape data in the arbitrary region of each surface of a plurality of metal samples that are buried in the ground and that have different age A derivation process for deriving a function of function change over time;
A shape that represents the predicted shape of the surface of the metal sample in the elapsed time by arranging the value of the parameter in an arbitrary region in accordance with the parameter distribution function in a predetermined elapsed time derived from the function of change over time of the parameter distribution function A generation process for generating a model;
The prediction method characterized by including.   The prediction method according to claim 1, wherein the plurality of metal samples are similar in material and shape to each other.   The prediction method according to claim 1, wherein the parameter is height distribution information in the arbitrary region.