JP2022089440A - Structure deterioration prediction device, structure deterioration prediction method, and structure deterioration prediction program - Google Patents

Abstract

To facilitate understanding the degree of improvement in the accuracy of structure deterioration prediction.SOLUTION: A deterioration prediction device includes: a deterioration distribution creation unit that creates deterioration distribution of a structure divided into a plurality of surface elements obtained based on n=first inspection of the structure; a first deterioration distribution setting unit that sets first deterioration distribution in consideration of measurement error; a second deterioration distribution setting unit that sets second deterioration distribution in consideration of the measurement error and time error considering the time error; a deterioration speed correction coefficient multiplication unit that can evenly distribute up to 0 multiplication speed by multiplying a deterioration speed correction coefficient so that a degree of deterioration that is most deteriorated can be covered in the second deterioration distribution of an inspection result in consideration of the measurement error and the time error set by the second deterioration distribution setting unit; and a deterioration expression rate correction unit that corrects a deterioration expression rate of deterioration prediction so that the deterioration distribution in consideration of the measurement error and the time error set by the second deterioration distribution setting unit and the second deterioration distribution set by the second deterioration distribution setting unit match.SELECTED DRAWING: Figure 1

Description

The present invention relates to a technique for predicting deterioration of a structure. In particular, the present invention relates to a technique for predicting future deterioration of the structure by applying a measurement error and a temporal error to the inspection result performed after a predetermined number of years have passed since the structure was constructed.

In Japan, structures (infrastructure) such as actual bridges constructed after the high-growth period are expected to deteriorate rapidly in the future. It is necessary to systematically maintain and renew aging structures to ensure the safety and security of the people and to reduce and level the total cost of maintenance and renewal. The Ministry of Land, Infrastructure, Transport and Tourism introduced the concept of preventive maintenance as a plan to clarify the direction of medium- to long-term efforts to steadily promote the maintenance and renewal of structures. We are making efforts such as formulating a "plan (action plan)" in May 2004 ahead of other ministries and agencies.

For this reason, from the conventional "ex post-management" of dealing with damage after it occurs, "prevention" of inspecting in advance and promptly taking measures before a fatal defect appears when an abnormality is confirmed or predicted. Life cycle cost (Life Cycle Cost,) by shifting to "conservative management" and strategically implementing maintenance to protect the lives and property of the people, ensure safety and security, and extend the life of the facility. Hereinafter referred to as "LCC") is required to be reduced.

As a strategic maintenance method for bridges, a bridge management system (Bridge Management System, hereinafter referred to as “BMS”) is attracting attention, and research for improving the accuracy of BMS is underway. In order to improve the accuracy of BMS, it is necessary to improve the accuracy of LCC calculation, and for that purpose, the method and timing selection that minimizes LCC from various repair methods by matching the deterioration prediction with the actual deterioration. It is essential to calculate the repair quantity with good method and accuracy.
As a deterioration prediction model, there is a model (typically a Markov chain model) that statistically predicts the deterioration based on the discrimination result between the physical model and the inspection data (see Non-Patent Document 1 below).

In the specification of the present application, it is decided to use a physical model that is expected to improve the accuracy of deterioration prediction due to future technological progress. Since the deterioration state of an actual structure varies depending on the member and location, it is necessary to vary the deterioration prediction so that the deterioration distribution of the actual structure can be predicted. However, the deterioration distribution of the deterioration prediction and the deterioration distribution of the actual bridge do not match just by making them scattered.
Therefore, it is necessary to correct the distribution of deterioration prediction so that it matches the distribution of the actual bridge.

The deterioration distribution of the constructed structure varies due to various factors. For example, the deterioration distribution of an actual bridge, which is an actual bridge (bridge) that has already been constructed, varies depending on the temperature, humidity, variation in the amount of flying salt, variation in materials, and the way in which wind and rain are received due to differences in the location of members. The cause is the difference in temperature and humidity due to the slight difference in the terrain.

In order to consider these causes individually, it is necessary to identify the characteristics of various causes for each constructed structure. For example, it is necessary to identify the characteristics of these causes for each bridge.
Furthermore, uncertain error factors that cannot be taken into consideration in the current research results are also possible, so even if the characteristics of variations in material conditions and environmental conditions are identified for each structure, for example, bridges, deterioration prediction and actual results are possible. It is difficult to match the deterioration distribution of the structure.

Here, the deterioration rate (mg / m ² / hour) of the structure is affected by the input conditions such as material conditions and environmental conditions.
Therefore, rather than considering the amount of flying salt, temperature, humidity, and variations in materials and surrounding environment separately, it is possible to simplify the model by considering the variations in the "deterioration rate" that is the result of considering them.
Deterioration rate (mg / m ² / hour) is a function of time and cumulative corrosion amount (mg / m ² ). The degree of deterioration is determined from the cumulative corrosion amount (mg / m ² ), and the cumulative corrosion amount (mg / m ² ) can be converted.

The inventors have already proposed to consider the variation in the deterioration prediction with the variation in the "deterioration rate" (Non-Patent Document 2 below).
The deterioration distribution of the actual bridge can be obtained from the inspection results, but the inspection results include the measurement error of the engineer who carried out the inspection. In addition, even if the crack is measured, it is unknown when the crack was in that state, so an error (time error) between the time when the crack occurred and the time when the inspection was performed is also included.

In Patent Document 1 below, the inventors give variation to a definite prediction, correct the distribution of variation in deterioration prediction to the distribution of variation in inspection result using the inspection result, and then disperse it. A weight of "deterioration expression rate" is given to the deterioration prediction, and it is corrected to match the distribution of the inspection results. On the other hand, the "measurement error" included in the inspection results and when the deterioration state was reached. We propose a method to correct the deterioration prediction by setting the range surrounded by the range of "time error" as the "error box". In addition, the deterioration prediction corrected by the first inspection result is compared with the second inspection result to verify the effect of the correction.

Japanese Unexamined Patent Publication No. 2019-15528

Naokatsu Tsuda, Kiyoyuki Kaido, Kazuya Aoki, Kiyoshi Kobayashi: Estimating Markov Transition for Predicting Bridge Deterioration, JSCE Proceedings, No.801 / I-73, pp.69-82, 2005. Tetsuro Kudo, Bonkogge Sakurunatacorn, Seigo Nasu: Structure repair scenario considering variation in deterioration, JSCE Proceedings E2, Vol.68 No. 4, pp.316-329, 2012.

However, even when the correction technique for predicting deterioration of a structure such as an actual bridge described in Patent Document 1 is used, there is a problem that the accuracy of predicting deterioration of the structure is not sufficiently high.
Further, in the technique described in Patent Document 1, since the "deterioration score" is used to indicate the degree of deterioration, there is a problem that it is difficult to understand the degree of improvement in the accuracy of deterioration prediction.
Therefore, an object of the present invention is to improve the accuracy of deterioration prediction of a structure.
Another object of the present invention is to make it easy to understand the degree of improvement in the accuracy of deterioration prediction.

According to one aspect of the present invention, it is a deterioration prediction device for a structure.
A deterioration distribution creation unit that creates a deterioration distribution of the structure divided into a plurality of surface elements, which is obtained based on n = first inspection of the structure.
In the deterioration distribution obtained by the deterioration distribution creating unit, the first deterioration distribution setting unit that sets the first deterioration distribution in consideration of the measurement error, and the deterioration distribution setting unit.
In the first deterioration distribution in consideration of the measurement error, the second deterioration distribution setting unit for setting the second deterioration distribution in consideration of the time error and the measurement error and the time error, and the second deterioration distribution setting unit.
"Deterioration rate correction" so as to cover the degree of deterioration that is most advanced in the second deterioration distribution of the inspection result considering the measurement error and the time error set by the second deterioration distribution setting unit. Deterioration speed correction coefficient multiplication unit that can be evenly distributed up to 0x speed by multiplying by "coefficient",
The deterioration expression rate of the deterioration prediction is corrected so that the deterioration distribution considering the measurement error and the time error, which is set by the second deterioration distribution setting unit, and the deterioration distribution calculated by the deterioration prediction match. Deterioration expression rate correction unit and
Have,
Hereinafter, a structure deterioration prediction device is provided, wherein n + 1 is used, and the process is continued until n = m (m is the number of final inspections of 2 or more) and then the process is terminated.
Here, the ratio of the surface area expressed by one deterioration prediction is the "deterioration expression rate". In addition, there is a possibility that there is a part (element) in the actual bridge that does not deteriorate at all, and in order to express that, the slowest deterioration speed is referred to as "0x speed". That is, in the present invention, it is necessary to match the "deterioration distribution considering measurement error and time error" with the "deterioration distribution calculated by deterioration prediction".

The first deterioration distribution setting unit is
It is characterized in that a measurement error correction matrix is used for the deterioration distribution obtained by the deterioration distribution creating unit to set a first deterioration distribution in consideration of the measurement error.

The deterioration distribution creation unit
It is characterized in that the degree of deterioration of each of the plurality of surface elements, which is the state of variation of the deterioration of each of the plurality of surface elements, which is grasped by the inspection of the actual structure divided into the plurality of surface elements, is obtained.

Here, in the present invention, variation is given to the deterministic prediction, and the distribution of the variation in the deterioration prediction is corrected to the distribution of the variation in the inspection result by using the inspection result. The method of considering the variation is a method of varying the definite deterioration prediction from 0 times to an arbitrary multiple that can cover the deterioration of the actual bridge.
In addition, since it does not match the deterioration distribution of the inspection results just by making them scattered, a weight called "deterioration expression rate" is given to the various deterioration predictions, and it is corrected so as to match the distribution of the inspection results.

From the above, by changing the density of deterioration prediction, it is possible to predict deterioration consistent with the deterioration distribution of the inspection results.
In the present invention, since the inspection result used for the correction includes a measurement error and a time error, the measurement error is taken into consideration for the inspection result, and the time error when the deterioration state is reached is taken into consideration. ..
This makes it possible to comprehensively predict the deterioration rate that may occur as a physical phenomenon.

The number of deterioration prediction elements is 100 to 1000 elements.
The structure is characterized by being a real bridge.

According to another aspect of the present invention, it is a method for predicting deterioration of a structure.
a) The step of conducting the first inspection and grasping the deterioration distribution created in the deterioration distribution creation step, and
b) With the step that the deterioration distribution setting unit sets the first deterioration distribution in consideration of the measurement error by using the measurement error correction matrix for the obtained deterioration distribution.
c) A step in which the deterioration distribution setting unit considers the time error in the deterioration distribution considering the measurement error and sets the second deterioration distribution in consideration of the measurement error and the time error.
d) Deterioration rate correction coefficient multiplication unit, until the definite deterioration prediction can cover the degree of deterioration that is most deteriorated in the "deterioration distribution of inspection results considering measurement error and time error", "deterioration rate" A step that multiplies the "correction coefficient" and evenly distributes up to 0x speed,
e) The deterioration expression rate correction unit corrects the "deterioration expression rate" of the deterioration prediction so that the deterioration distribution of the inspection result considering the measurement error and the time error and the deterioration distribution calculated by the deterioration prediction match. Steps to do and
Have,
Hereinafter, a method for predicting deterioration of a structure is provided, wherein the process is continued until n = m (m is the number of final inspections of 2 or more), and then the process is terminated, where n + 1.
The "deterioration expression rate" is as described above.

The step of setting the first deterioration distribution is
A method for predicting deterioration of a structure is provided, which comprises setting a first deterioration distribution in consideration of a measurement error by using a measurement error correction matrix for the deterioration distribution obtained by the deterioration distribution creation step. That is, in the present invention, it is necessary to match the "deterioration distribution considering measurement error and time error" with the "deterioration distribution calculated by deterioration prediction".

The deterioration distribution creation step is
It is characterized in that the degree of deterioration of each of the plurality of surface elements, which is the state of variation of the deterioration of each of the plurality of surface elements, which is grasped by the inspection of the actual structure divided into the plurality of surface elements, is obtained.

Further, the present invention is a program for causing a computer to sequentially execute the processes in the structure deterioration prediction method described above.

According to the present invention, the accuracy of deterioration prediction of a structure can be improved.
Further, according to the present invention, it is possible to easily understand the degree of improvement in the accuracy of deterioration prediction.

It is a flowchart which shows the flow of the deterioration prediction processing of a structure by embodiment of this invention. It is a functional block diagram which shows one configuration example of the deterioration prediction processing apparatus of a structure by embodiment of this invention. It is a figure which compares the degree of deterioration judged by the staff and the degree of deterioration judged by an expert, and shows the tendency of the measurement error of the staff in the damage evaluation distribution map. It is a figure which shows the deterioration prediction example when the deterioration prediction is made arbitrary, the horizontal axis shows time, and the vertical axis shows the degree of deterioration. It is a figure which shows the calculation example of the deterioration expression rate in consideration of a measurement error. It is a figure which shows the image example of the deterioration degree distribution in consideration of a time error. It is a figure which shows the image example of the deterioration distribution which considered both the measurement error and the time error. It is a figure which shows the relationship between the spread of a distribution due to a measurement error and a time error when the 1st and 2nd inspections are considered, and the improvement of accuracy by increasing the number of inspections used for correction. It is a figure which shows the relationship between the spread of a distribution due to a measurement error and a time error when the 1st to 3rd inspections are considered, and the improvement of accuracy by increasing the number of inspections used for correction. It is a figure which shows the deterioration degree definition in the Kochi prefecture inspection manual in Example 1 of this invention. It is a figure following FIG. 10, and is the figure which shows the classification of the degree of damage and the classification of the degree of damage. It is a map showing the position of the bridge to be analyzed. It is a figure which shows the deterioration prediction which considered the time error of 4 years ago in Katakasucho Ohashi, and is the figure which shows the example of the case where the deterioration rate correction coefficient is 1 times. It is a figure which shows the deterioration prediction which considered the time error of 4 years ago in Katakasucho Ohashi, and is the figure which shows the example of the case where the deterioration rate correction coefficient is 7 times. It is a figure which shows the comparison result of the inspection result and deterioration prediction in Katakasucho Ohashi. It is a figure which shows the consistency of the inspection result and deterioration prediction in Katakasucho Ohashi (100 elements). It is a figure which shows the consistency of the inspection result and deterioration prediction in Katakasucho Ohashi (1000 elements). It is a figure which shows the consistency of the inspection result and deterioration prediction in Katakasucho Ohashi (comparison of 100 elements and 1000 elements). It is a figure which shows the consistency of the inspection result and deterioration prediction in the Narutaki Bridge (100 elements). It is a figure which shows the consistency of the inspection result and deterioration prediction in Todoroki Bridge (100 elements). It is a figure which shows the consistency of the inspection result and deterioration prediction in Kumano Jindai Bridge (100 elements).

In the specification of the present application, the inspection result is corrected by using the inspection result after considering the measurement error and the time error. The inspection data shall be explained using the data provided by Kochi Prefecture.
The description and drawings of Patent Document 1, which is a related invention of the present application, will be described as appropriate. Therefore, the description, drawings, etc. of Patent Document 1 are also included in the contents of the present specification.

(1) Method of considering variation in deterioration prediction model 1-1) Focus on consideration of variation Deterioration of actual bridges varies due to variations in temperature, humidity, flying salt content, variations in materials, and differences in member locations. The cause is the difference in temperature and humidity due to the way of receiving wind and rain and the slight difference in the topography around the bridge. In order to consider these causes individually, it is necessary to identify the characteristics of these causes for each bridge. Furthermore, since uncertain error factors that cannot be taken into consideration in the current research results are possible, deterioration prediction and deterioration distribution of actual structures can be identified even if the characteristics of variations in material conditions and environmental conditions are identified for each bridge. Is difficult to match. On the other hand, the deterioration rate is affected by the input conditions such as material conditions and environmental conditions. From the above, it is possible to simplify the model by considering the variation in the "deterioration rate" which is the result of considering them, rather than considering the variation in the amount of flying salt, temperature, humidity, material and surrounding environment separately. Therefore, the variation in the deterioration prediction is considered with the variation in the "deterioration rate".

1-2) Consideration method and correction policy of variation distribution When the actual structure (bridge in this paper) is inspected, the deterioration state that varies in the plane (each element) as shown in the example of FIG. 1 of Patent Document 1 is obtained. Be done. Further, the deterioration rate for each element is an image as shown in FIG. 2 of Patent Document 1. Here, the vertical axis of FIG. 2 of Patent Document 1 is the cumulative corrosion amount (mg / m ² ), but it is shown that the corrosion rate (mg / m ² / hour) changes with time. In addition, the "deterioration degree" is obtained from the cumulative corrosion amount, and an image obtained by converting the cumulative corrosion amount into the deterioration degree is also shown.

The deterioration rate (function of time, cumulative corrosion amount, and degree of deterioration) in FIG. 2 of Patent Document 1 varies depending on the material conditions and environmental conditions. Since the example of FIG. 1 of Patent Document 1 has 20 elements in total, it can be said that the ratio of 5% in area per one is expressed. In the present specification, the ratio of the surface area expressed by each deterioration prediction is defined as "deterioration expression rate". Since the deterioration rate based on the conventional deterioration prediction is definite as shown in FIG. 3 of Patent Document 1, it is not possible to consider the variation in deterioration. Therefore, as shown below, the deterministically calculated deterioration rate is set to 1x and can be dispersed in any multiple.

1-2-1)
There is a possibility that there is a part (element) that does not deteriorate at all in the actual bridge, and in order to express that, the slowest deterioration speed is set to "0x speed".

1-2-2)
As the fastest deterioration rate, after grasping the deterioration distribution of the actual bridge, it is compared with the deterioration rate calculated deterministically, and the coefficient is multiplied so that the deterioration distribution of the actual bridge can be sufficiently covered. This coefficient is defined as the "deterioration rate correction coefficient".

1-2-3)
The interval of the multiple is an interval that can sufficiently cover the deterioration distribution of the actual bridge between the slowest deterioration rate (0 times speed) and the fastest deterioration rate. Here, the deterioration degree distribution of the actual bridge varies between a to e in FIG. 1 of Patent Document 1, but if the interval between the magnifications is too large, the deterioration becomes b, c or d between a and e. In some cases, there is no speed.

Here, "a range that can be sufficiently covered" means that all the deterioration degrees can be expressed. FIG. 4 of Patent Document 1 is a deterioration image dispersed in the above manner, and is dispersed in 0.4x increments from 0x to 4x with respect to a definite deterioration prediction. Since it is scattered among 10 lines, the "deterioration expression rate" of 1 line is 10%.

However, when the deterioration prediction is arbitrarily dispersed by this method, the variation becomes even, and therefore, it does not match the deterioration degree distribution of the actual bridge, which is the uneven variation shown in FIG. 2 of Patent Document 1. Therefore, correction is required.

As a correction method, by correcting the "deterioration expression rate" per deterioration curve, it can be matched with the actual bridge. FIG. 5 of Patent Document 1 shows an image of "deterioration expression rate" in deterioration prediction, but as shown in an image of deterioration rate correction in FIG. 6 of Patent Document 1, the "deterioration expression rate (ratio of bridge area)" is shown. ) ”Is corrected to match the inspection result.
It should be noted that the deterioration prediction in the present specification predicts the rate of deterioration distribution, and cannot indicate a specific part or the like.

1-3) Specific correction method A specific correction method is shown using the actual bridge model shown in FIG. 1 of Patent Document 1 and the deterioration prediction shown in FIG. 4 of Patent Document 1. Since FIG. 4 of Patent Document 1 disperses the deterioration curve into 10 lines, 10% of the bridge area per line is expressed when considered evenly. However, in that case, as described above, since it does not match the deterioration degree distribution of the inspection result, the area per one is corrected from 10% so as to match the inspection result. The correction method is shown below.

1-3-1)
Since the degree of deterioration "a" in the inspection result is 10% and the "deterioration expression rate" of the deterioration prediction is also 10%, the ratio of the inspection result and the deterioration prediction is consistent with respect to "a", and correction is performed. do not have to.

1-3-2)
The degree of deterioration "b" in the inspection result is 40%. Since there are two deterioration rates that are "b" in the deterioration prediction, the "deterioration expression rate" when considered evenly is 20%. However, in order to match the inspection result with 40%, the "deterioration expression rate" is corrected from 10% to 20% (40/2 = 20%) for each deterioration rate. As a result, the inspection results and the deterioration degree distribution of the deterioration prediction match.

1-3-3)
Similarly, since there is only one deterioration rate at which the degree of deterioration is "c" in the deterioration prediction, the ratio is 10% when considered evenly, but 25% is present in the inspection result. Therefore, by correcting the "deterioration expression rate" per one from 10% to 25%, the ratio of the deterioration degree "c" is consistent with the inspection result.

1-3-4)
Similarly, since there are four deterioration rates at which the degree of deterioration is "d" in the deterioration prediction, the ratio is 40% when considered evenly, but only 20% is present in the inspection result.
Therefore, by correcting the "deterioration expression rate" per one from 10% to 5% (20/4 = 5%), the ratio of the deterioration degree "d" is consistent with the inspection result.

1-3-5)
Finally, since there is one deterioration rate in which the degree of deterioration is "e" in the deterioration prediction, the ratio is 10%, but the inspection result shows only 5%. Therefore, by correcting this one "deterioration expression rate" from 10% to 5%, the ratio of the deterioration degree "e" is consistent with the inspection result.

By the above method, the deterioration prediction and the deterioration degree distribution of the inspection result are matched. By correcting the "deterioration expression rate" per deterioration prediction and then predicting future deterioration, it is possible to make predictions in consideration of variations in deterioration of the actual structure.

Although the planar elements shown in FIG. 1 of Patent Document 1 are equally divided, the area of the actual bridge inspection may differ for each element as shown in FIG. 7 of Patent Document 1. In this case as well, if the ratio of the degree of deterioration in consideration of the area can be obtained by inspection, the correction of the deterioration prediction can be corrected by using the above-mentioned idea.

Hereinafter, the structure deterioration prediction technique according to the embodiment of the present invention will be described in detail with reference to the drawings.

(2) Method for considering measurement error of inspection result 2-1) Outline In 1 above, a method for correcting deterioration prediction using inspection results has been described.
However, in the inspection result, an error occurs in the deterioration prediction due to various factors such as the inspection method, the inspection condition, and the defined degree of deterioration. There is a method to consider all of these factors, but in the current research, there are uncertainties among these factors, which complicates the model and limits the improvement of accuracy.

Therefore, in the present invention, attention is paid to the "measurement error" in the deterioration prediction technique of the structure.
Here, the "measurement error" can be treated as an error in measuring the crack width of the actual bridge or an error determined by the engineer when the engineer measures the degree of deterioration of the structure as an action of various error factors.
In the present invention, the deterioration prediction is corrected for the deterioration distribution obtained by correcting the "measurement error" for the deterioration distribution obtained from the inspection.

2-2) Consideration of measurement error In Kochi Prefecture, regular inspections of actual bridges are carried out by staff. As part of staff education and training, the degree of deterioration judged by staff is compared with the degree of deterioration judged by experts, and the tendency of staff measurement error is analyzed.

In FY2016, when the analysis results are available, the tendency is shown in Fig. 3. Here, since the deteriorations targeted in the research related to the present invention are "cracking", "reinforcing bar exposure, peeling", and "free lime, swelling", the analysis results for these damages are used.

The measurement error described herein takes into account the measurement error using the trends in this material. In terms of the tendency of measurement error, the positive evaluation is 88%, the evaluation on the 3-step dangerous side is 2%, the evaluation on the 2-step dangerous side is 3%, the evaluation on the 1-step dangerous side is 3%, and the evaluation on the 1-step safe side is. It was 4%.

Here, the case where it is judged that the degree of deterioration (a to e) evaluated by the staff is more advanced than the expert's judgment is expressed as "safety side", and the case where it is not advanced is expressed as "dangerous side". There is. As for the specific method of considering error correction using this tendency, if the staff evaluates a, the risk side cannot be evaluated any more. Therefore, the ratio of the positive evaluation and the risk side evaluation is "correct evaluation 88 + 3 + 3 ++ 2". = 96% ". In addition, the ratio of b evaluation on the one-step safety side is 4%. b Evaluation and below are also corrected in the same manner as above. Table 1 shows the correction matrix in which the above are arranged.

(3) Specific examples of measurement error 3-1) Examples of consideration of measurement error when the number of inspections is one An example of correction considering measurement error is shown in the inspection results. First, as in Patent Document 1, a bridge having two inspection elements in the case where the number of inspections is one will be described.
If the inspection result of this bridge is element 1 = a and element 2 = c, the deterioration distribution considering the measurement error is as shown in Table 2.

Here, the total deterioration distribution ratio in consideration of the measurement error is 1.045 and not 1.000 because the judgment of the staff is biased. Therefore, the inventor came up with the idea of rebating the whole so that the total deterioration distribution becomes 1.000.
Table 3 shows the deterioration distribution when the measurement error based on the above is taken into consideration.

From the above, when the Kochi Prefecture staff inspects the two-element bridge and determines that element 1 = a and element 2 = c, the final ratio of deterioration degree a is 0.459 and the ratio of deterioration degree b is 0.057. The ratio of the degree of deterioration c is 0.445, the ratio of the degree of deterioration d is 0.024, and the ratio of the degree of deterioration e is 0.014. Further, if the deterioration prediction of this bridge is the result shown in FIG. 4, the deterioration expression rate is as shown in FIG.

3-2) Example of Consideration of Measurement Error When The Number of Inspections is Two The embodiment of the present invention is characterized in that the inspection results of two or more inspections are used.
First, consider a bridge in which the number of inspections is two and the inspection elements are two elements. If the inspection result of this bridge is element 1 = a, element 2 = c in the first inspection, element 1 = b, and element 2 = c in the second inspection, the deterioration distribution considering the measurement error is as shown in Table 4. It becomes.

Here, the results of the two inspections are 5 × 5 = 25 combinations, but if the correct deterioration judgment is made, the deterioration cannot be improved naturally, so b → a, c → b and c → a. Etc. can be ignored. However, if repairs are carried out during the period between inspections, improvement is expected, so improvement in the degree of deterioration should be considered.

In addition, the deterioration prediction speed, which is considered to be the maximum after considering the measurement error and the time error described later with respect to the first inspection result, is also predicted. This is considered to be the maximum deterioration rate based on the inspection results.

Therefore, it is considered that a deterioration degree exceeding this deterioration prediction speed does not occur as a physical phenomenon, and is excluded from the deterioration prediction target. For example, an inspection result such as b → d or b → e in which the degree of deterioration progresses by two or more stages may be obtained after five years. If the environmental conditions are severe in the deterioration prediction, the degree of deterioration may progress by two or more stages in five years. In that case, such inspection results are also subject to the deterioration prediction.

In addition, in the deterioration prediction, if there is no speed to advance by two or more steps in 5 years, the deterioration prediction this time is performed by the physical model, and it is considered that the deterioration in the range exceeding this physical deterioration prediction model does not actually occur. Therefore, it is considered to be deterioration or damage due to different factors, and is not subject to the amendment of the present invention.

Further, when the measurement error is taken into consideration, the degree of deterioration may progress significantly as in the case of a → d, a → e, and the like. In the example of Table 4, a → c, a → d, a → e, b → d, b → e, and c → e are considered to be in the range exceeding the deterioration prediction, and are therefore excluded from the study.

Table 5 shows the results of adjusting the rebate so that the total of the results of the second inspection in consideration of the above is 1.000. In this example, as a result, a → a, a → b, b → b, b → c, c → c and c → d remain. The deterioration distribution can be expressed by considering the inspection results from the second time onward in the same manner.

(4) Method to consider the time error of the inspection result 4-1) Outline Since the periodic inspection is decided to be once every 5 years, it is unknown when the "deterioration" occurred. In other words, the degree of deterioration of the actual bridge changes over time, and when the damage such as "cracking" and "reinforcing bar exposure and peeling" found in the inspection became that state, etc., at the time of inspection and at the time of damage occurrence, etc. There will be a time error. The error is based on the fact that the inspection is carried out once every five years. It is considered that there is an error of up to 4 years.

The inspection result needs to be obtained as a deterioration distribution considering this "time error". Here, a method of considering "time error" in the inspection result will be described.
Consideration of time error gives a deterioration distribution in the inspection at the time of inspection, but assuming that this deterioration distribution occurred 1 to 4 years before the inspection, it was predicted under the initial conditions as shown in FIG. A deterioration curve having a faster deterioration rate than the deterioration curve can be obtained. This is a deterioration curve considering the time error, and the deterioration distribution obtained from this deterioration curve is the deterioration distribution of the time error.

In the deterioration prediction, the deterioration distribution due to the above-mentioned measurement error and the deterioration distribution due to this time error are taken into consideration in the deterioration distribution obtained from the inspection result.

At present, the probability of causing a time error is unknown, but the cumulative probability of deterioration distribution at the time of inspection results is "0" at the time of 5 years ago and "1" at the time of inspection. .. At present, it is unknown at what point the deterioration occurs, so if this probability distribution function is assumed to be the simplest linear function, Eq. (1) is obtained.
P (x) = 0.2x (1)
Here, P (x) is the cumulative probability, and x is the time (year) elapsed since the previous inspection.

4-2) Consideration of time error When the equation (1) is differentiated, the right side becomes a constant of 0.2, so the probability of deterioration distribution at the time of inspection is 20% for each year from 4 years ago to the time of inspection. It becomes. As described above, five deterioration prediction models are derived for each bridge due to the time error, and the probability of each deterioration prediction is 20%. Therefore, it is necessary to take this probability into consideration in the "deterioration expression rate". be.
However, if the time error is taken into consideration, the deterioration prediction with a high deterioration rate is taken into consideration, but the deterioration rate may be too fast to obtain a deterioration curve that does not match the inspection result. In that case, the deterioration expression rate of the deterioration curve is 0%.

5) Summary of predicted deterioration distribution 5-1) Impact of consideration of deterioration distribution on accuracy So far, for definite deterioration prediction, distribution due to variation has been given, and corrected distribution due to measurement error of inspection. We have considered the correction distribution due to time error. Images of deterioration distribution considering both measurement error and time error are shown in FIGS. 7 to 9. As shown in FIGS. 7 to 9, at the time of the first inspection, when the correction distribution due to the measurement error and the time error distribution due to the time error are taken into consideration with respect to the initial deterioration prediction distribution, a wider range than the initial deterioration prediction distribution is obtained. It turns out that it will have to be predicted. Further, it can be seen that the prediction range is narrower than that at the time of the first inspection, although the range is wider than the initial deterioration prediction distribution when the measurement error and the time error are similarly taken into consideration at the time of the second inspection. This deterioration prediction is a method of predicting by giving a density called deterioration expression rate to the deterioration curve.

Therefore, narrowing the range of deterioration prediction means that the density of the deterioration curve is high, and therefore, the higher the density of the deterioration curve, the higher the prediction accuracy. From the above, it is considered that the prediction range will be expanded once by considering the measurement error and the time error, but the accuracy will be improved by repeating the correction by the inspection.

5-2) Flow of inspection result correction and deterioration prediction Based on the above, the flow of inspection result correction and deterioration prediction correction is shown below. FIG. 1 is a flowchart showing a flow of a structure deterioration prediction process according to the present embodiment. Further, FIG. 2 is a functional block diagram showing a configuration example of the deterioration prediction processing device A for the structure according to the present embodiment.

The processing flow will be described below with reference to FIGS. 1 and 2.
a) The first inspection is carried out, and the deterioration distribution created by the deterioration distribution creation unit 1 is grasped (n = 1 in step S1, step S2).
b) Using the measurement error correction matrix for the deterioration distribution obtained in step S2, the deterioration distribution setting unit 3 sets the deterioration distribution in consideration of the measurement error (step S3).

c) The deterioration distribution setting unit 5 considers the time error in the deterioration distribution in consideration of the measurement error, and sets the deterioration distribution in consideration of the measurement error and the time error (step S4).
d) The deterioration rate correction coefficient multiplication unit 7 allows the definite deterioration prediction to cover the degree of deterioration in which the deterioration is most advanced in the "deterioration distribution of inspection results considering measurement error and time error". By multiplying by the "deterioration speed correction coefficient", the speed can be evenly distributed up to 0 times speed (step S5). That is, the deterioration rate is multiplied by the deterioration rate correction coefficient, and the deterioration rate is evenly distributed from 0 times speed to cover the deterioration distribution with the most advanced deterioration degree.

e) The deterioration expression rate correction unit 11 corrects the "deterioration expression rate" of the deterioration prediction so that the "deterioration distribution of the inspection result considering the measurement error and the time error" and the deterioration distribution match (step S6).
That is, the "deterioration expression rate" of the deterioration prediction is corrected so that the deterioration distribution obtained from the deterioration prediction and the deterioration distribution of the inspection result considering the error match.
Hereinafter, the process is continued as n + 1 (step S8) until n = m (m is the number of final inspections) (step S7), and then the process is terminated (end).

The above processing may be performed by invention, hardware processing, or software processing.
In the present invention, it is hypothesized that the accuracy of deterioration prediction is improved by performing the above adjustments every time an inspection is performed.

[Example 1]
Hereinafter, an example of deterioration prediction using the structure deterioration prediction processing device A according to the embodiment will be described.

(Example of deterioration prediction in a real bridge)
In the above, the correction method of deterioration prediction using the inspection result and the hypothesis that the error range becomes narrower and the accuracy of deterioration prediction improves as the inspection result increases. Here, using inspection data from Kochi Prefecture, we will verify the accuracy of actual bridge deterioration prediction and correction and deterioration prediction using inspection results.

(1) Outline of inspection system in Kochi Prefecture The inspection data used is the data of regular inspections conducted in Kochi prefecture once every five years. In Kochi Prefecture, inspections are being carried out by staff, and as of March 2019, the second order (second) inspection is being carried out.

In the inspection system of Kochi prefecture, the degree of deterioration is determined by the indexes shown in FIGS. 10 and 11. By aggregating the degree of deterioration for each element as shown in FIGS. 1 and 4 (application) of Patent Document 1, it is possible to obtain information on the variation in deterioration for each bridge. In the analysis of this paper, among the bridges in the jurisdiction of Kochi prefecture, the PC bridges whose inspection results are registered in the database more than three times are "Katakasu Ohashi", "Kataki Bridge", "Todoroki Bridge", and "Todoroki Bridge". It was carried out for "Kumano Jindai Bridge" and so on. The figure shown in the map of those real bridges is FIG.
The deterioration prediction process shows a correction process of deterioration prediction based on the inspection result after considering the measurement error and the time error of the inspection result.

In addition, deterioration prediction not corrected by inspection, deterioration prediction corrected by the first inspection, deterioration prediction corrected by the first and second inspection results, and deterioration prediction result corrected by the first to third inspection results and the third inspection The accuracy verification by comparison with the result will be described.

(2) Inspection results at the actual bridge 2-1) Bridge (analysis) specifications "Katakasu Ohashi", "Kataki Bridge", "Todoroki Bridge", and "Kumano Jindai Bridge" were erected from 1973 to 1998. It is a PC bridge that was built. In addition, since there is no detailed structural drawing or bar arrangement drawing, it was designed by the Ministry of Construction standard drawing collection, the same specification (S36 prestressed concrete road bridge specification), and S43 prestressed concrete road bridge specification. The analysis will be carried out after assuming the necessary specifications with reference to the design drawings of bridges of similar scale. Table 6 shows the analysis specifications.

2-2) Results of the first inspection considering measurement errors Table 7 shows the year of inspection of the four bridges to be analyzed.

The first inspection was in 2008 (24 years in service) at "Katakasu Ohashi", 2009 (33 years in service) for "Rakutaki Bridge", 2010 (37 years in service) for "Todoroki Bridge", and "Kumano Jindai Bridge". The first inspection was carried out in 2008 (10 years of service).
The second and subsequent inspections are basically carried out every 5 years after the first inspection, but there are some bridges that are less than 5 years old due to the inspection plan.

As described above, since the inspection result obtained here is considered to include the measurement error, the obtained inspection result is corrected by the correction matrix of the measurement error shown in Table 1.
Since the deterioration rate of the deterioration prediction of the present invention is set by using the first inspection result, the deterioration degree distribution in consideration of the measurement error is set in the first inspection result.
As an example of correction, Table 8 shows the result of the first inspection at Katakasucho Ohashi, and Table 9 shows the inspection result corrected by the correction matrix of the measurement error.

Further, Table 10 shows the results of the first to third inspections, which are the targets of deterioration prediction of the present invention, and Table 11 shows the results of the first to third inspections corrected in consideration of the measurement error.

Further, as described above, it is necessary to correct the results of the first to third inspections in consideration of the measurement error.
For correction, in the 1st to 2nd inspection, or from the 2nd to 3rd inspection, it is confirmed whether the deterioration degree is within the range or out of the range of the physical phenomenon, and the inspection result outside the range is. exclude.
The confirmation method is performed by comparing with the deterioration prediction considering the maximum possible deterioration rate.

Table 11 shows the number of deterioration predictions, and makes it possible to compare the results of deterioration progressed by two or more stages with the deterioration predictions.
Here, it is considered that the deterioration prediction of 0 (bold portion) corresponding to the phenomenon that the degree of deterioration progresses by two or more stages is out of the range. Table 12 shows the results of rebating so that the total deterioration degree distribution is 1.000, excluding the outside of the range.

(3) Correction of deterioration prediction using inspection results 3-1) Outline Next, correction of deterioration prediction is carried out for the bridge to be analyzed using the inspection results. As the method for correcting the deterioration prediction using the inspection result, the method described in the above embodiment is used.
In addition, here, the deterioration prediction that Katakasucho Ohashi is not corrected by the inspection, the deterioration prediction that was corrected by the first inspection, the deterioration prediction that was corrected by the first and second inspection results, and the correction by the first to third inspection results were corrected. The deterioration prediction is shown by a concrete example to compare with the inspection result. Then, the accuracy of the number of inspections to be corrected gives to the deterioration prediction is verified.
Other bridges to be analyzed will be briefly described.

3-2) Setting the deterioration rate coefficient in the initial analysis The present invention proposes to correct the variation of the actual bridge by giving an arbitrary multiple to the definite deterioration prediction. Therefore, the definite deterioration prediction of the bridge to be analyzed is multiplied by "0x speed to double speed within the range that can cover the deterioration distribution of the actual bridge" and dispersed arbitrarily, and the deterioration prediction is performed using the inspection result corrected by the measurement error matrix. It becomes a flow to correct.

Here, the above-mentioned "double speed within the range that can cover the deterioration distribution of the actual bridge" means the most advanced deterioration degree in the scattered deterioration prediction and the most advanced deterioration degree in the inspection result of the actual bridge. Double speed that becomes the same.

The deterioration distribution of the actual bridge can be obtained from the inspection results, but it is necessary to consider the measurement error and time error in the inspection results as described above. In other words, the "double speed within the range that can cover the deterioration distribution of the actual bridge" is determined by comparing the deterioration distribution of the inspection result, which goes back four years from the time of the inspection, with the deterioration distribution of the deterioration prediction. It is necessary to consider the possibility of the deterioration distribution obtained from the first inspection result in the year.

In the result of the first inspection, as shown in Table 9, the ratio of e is 6.1%. Therefore, when considering the time error, it is necessary to consider that the degree of deterioration e exists in the 20th year of service. On the other hand, in the deterioration prediction, e does not exist because the deterioration rate is slow as shown in FIG. 13 when the 20th year of service is predicted at 1x speed.

Therefore, it is necessary to multiply the deterioration rate by an arbitrary multiple so that the deterioration distribution of the actual bridge can be covered. As a result of gradually increasing the multiple, it is necessary to set the multiple to 7 in order for the degree of deterioration e to exist in the 20th year of service. Therefore, in Katakasucho Ohashi, by "variing from 0 times to 7.0 times at 0.07 times intervals (100 elements)", "0 times speed to double speed within the range that can cover the deterioration distribution of the actual bridge". It will be scattered.

FIG. 14 is a diagram showing deterioration prediction in consideration of the time error of Katakasucho Ohashi four years ago, and is a diagram in the case of a deterioration rate correction coefficient of 7 times.
As shown in FIG. 14, the deterioration distribution of the analysis result can cover the deterioration distribution of the inspection result. In the present invention, this arbitrary multiple is referred to as a "deterioration rate correction coefficient".
Similarly, as a result of setting the deterioration rate correction coefficient for other bridges to be analyzed, the deterioration rate coefficient of Narutaki Bridge is 0.1, the deterioration rate coefficient of Todoroki Bridge is 5, and the deterioration rate coefficient of Kumano Jindai Bridge. Was 717.

3-3) Accuracy of deterioration prediction of Katakasucho Ohashi Next, the goodness of fit of deterioration prediction is confirmed for the first to third inspection results. As shown in bold, there is a degree of deterioration for which there is no deterioration prediction for the target inspection results shown in Table 12. This indicates that the prediction interval was large and could not be predicted in the deterioration prediction of 100 lines. Table 13 shows the results of totaling the deterioration degree distribution excluding the unpredictable deterioration degree.

As shown in Table 13, the total degree of deterioration is 0.913. This is the accuracy of the deterioration prediction when the deterioration prediction is distributed to 100 lines.

3-4) Deterioration prediction results of Katakasucho Ohashi For Katakasucho Ohashi, the above-mentioned results will be organized and the accuracy of deterioration prediction will be verified. Deterioration prediction not corrected by inspection, deterioration prediction corrected by the first inspection, deterioration prediction corrected by the first and second inspection results, and deterioration corrected by the first to third inspection results using the method described above. The results of the prediction are shown in Tables 14 to 17.

In addition, Fig. 15 shows a graph in which the inspection results of Katakasucho Ohashi shown in Table 12 and the results of each deterioration prediction are superimposed, and the ratio of the inspection results and the deterioration degree of the deterioration prediction is small value ÷ large value × inspection result. FIG. 16 shows the consistency of the sum of the probability functions.

Regarding the accuracy of deterioration prediction, as shown in Table 14, the deterioration prediction that is not corrected by the inspection result first has the same initial value of 0.010 as the deterioration expression rate for each of the 100 deterioration predictions. As shown in FIG. 15, the degree of consistency with the inspection result is small as a whole, and as shown in FIG. 16, the degree of consistency is 0.235.

Similarly, it can be seen that the deterioration prediction corrected only by the first inspection is significantly improved in accuracy with respect to the inspection result not corrected by the inspection result. It can be seen that there is some consistency with respect to the degree of deterioration in which the ratios of aaa, dee and ee are relatively large.
However, while there is almost no ratio in the inspection results for the relatively small degree of deterioration such as bbb and bcc, the deterioration prediction shows a ratio of about 3.0%, which is inconsistent.

Since the consistency in FIG. 16 is 0.802, the accuracy is about 80%.
Looking at FIG. 15, the deterioration prediction corrected in the first and second inspections is improved against the inconsistency of the low degree of deterioration such as bbb and bcc seen in the deterioration prediction corrected in the first inspection. It has been seen, and the results are almost the same as the inspection results.
However, there is some inconsistency with the inspection results for aaa, dee, and eee, which have a large rate of deterioration.

As shown in FIG. 16, the overall consistency is 0.818, and as a result, the accuracy is not significantly improved as compared with the deterioration prediction corrected in the first inspection.
Finally, the deterioration prediction corrected in the first to third inspections is consistent with the inspection results for any degree of deterioration. However, although there is a ratio in the inspection results to the degree of deterioration such as aab and abc, the degree of deterioration could not be predicted, so the consistency shown in FIG. 16 is 0.913.

However, at Katakasucho Ohashi, the accuracy of deterioration prediction has been improved by correcting the deterioration prediction using the inspection results, and the accuracy of deterioration prediction has improved by increasing the number of inspections to be corrected. I understood.

Next, for the problem that the degree of deterioration of Katakasucho Ohashi, such as aab and abc, which was corrected in the first to third inspection results, could not be predicted, the effect on the accuracy when the number of elements was increased and the prediction was made with 1000 elements. To verify.
Table 18 shows the analysis results, and Fig. 17 shows the comparison between the inspection results and the deterioration prediction of 100 elements and 1000 elements corrected by the 1st to 3rd inspection results. Consistency between 100 elements and 1000 elements. Is shown in FIG.

Comparing the deterioration prediction corrected by the inspection results of the 1st to 3rd inspections with the case of 100 elements and 1000 elements, looking at Table 17, Table 18 and FIG. 17, the tendency is almost the same, but the tendency is increased to 1000 elements. As a result, it is possible to predict the degree of deterioration with a low rate that could not be predicted by 100 elements such as abb and bbc, but by 1000 elements. However, the ratio is about 0.066 in total, and as shown in FIG. 18, the consistency, which was 0.913 for 100 elements, is 0.920 for 1000 elements, which is a result that only a slight improvement in accuracy can be expected.

From the above results, it was found that in the case of Katakasucho Ohashi, even if the number of elements is increased from 100 elements to 1000 elements, a significant improvement in accuracy cannot be expected. Therefore, it was found that a division of about 100 elements is sufficient.

3-5) Correction results for Narutaki Bridge, Todoroki Bridge, and Kumano Jindai Bridge For Narutaki Bridge, Todoroki Bridge, and Kumano Jindai Bridge, as with Katagasu Ohashi, before and after the correction of the 1st to 3rd inspection results. And the deterioration prediction not corrected by the inspection, the deterioration prediction corrected by the first inspection, the deterioration prediction corrected by the first and second inspection results, and the deterioration prediction result and the inspection result corrected by the first to third inspection results. The accuracy was verified by comparison.
FIG. 19 is a diagram showing the consistency between the inspection result and the deterioration prediction at the Narutaki Bridge (100 elements).
FIG. 20 is a diagram showing the consistency between the inspection result and the deterioration prediction in Todoro Bridge (100 elements). FIG. 21 is a diagram showing the consistency between the inspection result and the deterioration prediction at Kumano Jindai Bridge (100 elements). It is a figure corresponding to FIG. 16, respectively.

Although details are omitted, when the inspection results were compared before and after the correction, and the deterioration prediction results were compared, deterioration prediction was made when the inspection results were not corrected for all of the Narutaki Bridge, Todoroki Bridge, and Kumano Jindai Bridge. It turned out that the accuracy of was not good.
On the other hand, the prediction accuracy is greatly improved by making corrections based on the results of the first inspection, and the prediction accuracy is improved each time the corrections are made based on the results of the first and second inspections and the results of the first to third inspections. It was found that they showed a similar tendency. Further, the consistency between the inspection result and the deterioration prediction in the actual bridge is the same as in FIG. Therefore, it was found that the prediction correction technique of the present invention has versatility.

(summary)
The analysis results and inspection results of the above deterioration prediction are summarized.
(1) The four bridges, Katakasucho Bridge, Narutaki Bridge, Todoroki Bridge, and Kumano Jindai Bridge, are not corrected by the inspection results, and the accuracy of deterioration prediction is poor.
This is because the deterioration prediction technique of this embodiment targets a wide range of predictions in consideration of measurement error and time error, and sets the deterioration rate coefficient for the one with the fastest deterioration rate among the wide range of deterioration distributions. , Disperses definitive deterioration predictions.

In the actual deterioration distribution of these four bridges, the rate of high deterioration rate is small, but the deterioration prediction that is not corrected by the inspection results is an equal rate, so it is considered that the deviation from the actual deterioration distribution tends to be large.

(2) It was found that the accuracy of deterioration prediction corrected by more inspection results was improved for 4 bridges. As shown in FIGS. 8 and 9, the process of deterioration becomes clear as the number of inspection results increases. The deterioration prediction of the present invention increases the density while leaving the deterioration prediction consistent with the inspection result, and excludes the inconsistent deterioration prediction.

Therefore, it is considered that the accuracy is improved by leaving the deterioration prediction in which the history of deterioration is consistent and increasing the density.

(3) At Katakasucho Ohashi, the prediction accuracy was compared between the case where the number of deterioration predictions was 100 and the case where the number was 1000, and it was confirmed that there was no big difference.

This is because, as shown in Tables 17 and 18, the deterioration prediction of the present invention is multiplied by the deterioration rate coefficient, so that the number of deterioration predictions is larger than the distribution in which the deterioration rate is fast, and in that case, the deterioration rate is slow. The number of predicted deterioration tends to decrease with respect to the distribution.

In the process of deterioration by three inspections such as Katakasucho Ohashi, the number of deterioration predictions is small for the range where the deterioration rate is slow such as "ab", "abb" and "abc", and the deterioration prediction is It tends to be difficult to match the details of the inspection results. It is considered that this is because the accuracy is improved to some extent even if the number of deterioration predictions is changed from 100 to 1000, but the tendency does not change.

(Conclusion)
The present invention considers variations in deterioration prediction, which has been one of the causes of problems in developing a practical BMS, and further uses inspection results in consideration of measurement errors and time errors to create a deterioration prediction correction model. It is to provide.

(1) In the present invention, a method is presented in which variations are given to a definite prediction and the distribution of variations in deterioration prediction is corrected to the distribution of variations in inspection results using inspection results. As a method for considering the variation, we propose a method for diversifying the definite deterioration prediction from 0 times to an arbitrary multiple that can cover the deterioration of the actual bridge.
In addition, since it does not match the deterioration distribution of the inspection results just by making them scattered, we propose a method of giving a weight called "deterioration expression rate" to the various deterioration predictions and correcting it so that it matches the distribution of the inspection results. do.

From the above, by changing the density of deterioration prediction, it is possible to predict deterioration consistent with the deterioration distribution of inspection results.

(2) Since the inspection result used for correction includes measurement error and time error, the measurement error was taken into consideration for the inspection result, and the time error when the deterioration state was reached was taken into consideration.
From this, it was found that the deterioration rate that can occur as a physical phenomenon can be comprehensively predicted.

(3) Focusing on 4 bridges including Katagasu Ohashi, which is being inspected for the third time, from among the bridges damaged by salt in the Kochi prefecture jurisdiction, analysis is performed by applying the deterioration prediction model of the present invention. rice field.
It was found that the deterioration prediction corrected by the inspection result has higher prediction accuracy than the deterioration prediction not corrected by the inspection result, while showing an arbitrary multiple given to the definite deterioration prediction.
It was also found that the prediction accuracy of the correction based on the inspection result improves as the number of times increases.

(Supplement) Deterioration prediction model used in the present invention The deterioration prediction model used in the present invention is constructed by using the results of previous research. The deterioration prediction model consists of a flying salt crack model and a peeling model. A one-dimensional mass transfer equation is used for the flying salt content calculation model, chloride ion transfer model, corrosion model, and water transfer model. The quantity calculation model is the Kokubo model (Yukie Kokubo, Hajime Okamura: Calculation model of the amount of flying chloride ions focusing on the generation process of seawater droplets, JSCE Proceedings B, Vol.65 No.4, pp.259-268, 2009.8 ) Is used.

The chloride ion / corrosion model uses an electrochemical corrosion model (Koichi Maekawa, Tetsuya Ishida, Toshiharu Kishi: Multi-Scale Modeling of Structural Concrete 9) that can take into consideration changes in the environment, electrocorrosion protection, and repair effects.

The crack model is a proposed formula obtained from FEM analysis and corrosion crack experiment (Lukuan Q., Hiroshi Seki: Study on crack occurrence and crack width of concrete due to reinforcing bar corrosion, JSCE Proceedings, No.669 / V-50, pp. .161-171, 2001.) is used.
The exfoliation model is a proposal formula obtained from FEM analysis and experiments that the fog exfoliates when the corrosion depth exceeds the limit value (Seiichi Tottori, Toyoaki Miyagawa: Deterioration prediction related to reinforcing bar corrosion when affected by initial chloride ions, Proceedings of the Society of Civil Engineers, No. 781 / V-65, pp.157-170, 2005. (Reception on August 14, 2020) are used.

The processing and control in this specification are software processing by CPU (Central Processing Unit) and GPU (Graphics Processing Unit), ASIC (Application Specific Integrated Circuit) and FPGA (Field Programmable). can.

Further, in the above-described embodiment, the configuration and the like shown in the illustration are not limited to these, and can be appropriately changed within the range in which the effect of the present invention is exhibited. In addition, it can be appropriately modified and implemented as long as it does not deviate from the scope of the object of the present invention.

In addition, each component of the present invention can be arbitrarily selected, and an invention having the selected configuration is also included in the present invention.
Further, the program for realizing the function described in the present embodiment is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into the computer system and executed to process each part. May be done. The term "computer system" as used herein includes hardware such as an OS and peripheral devices.

Further, the "computer system" includes the homepage providing environment (or display environment) if the WWW system is used.
Further, the "computer-readable recording medium" refers to a portable medium such as a flexible disk, a magneto-optical disk, a ROM, or a CD-ROM, and a storage device such as a hard disk built in a computer system. Further, a "computer-readable recording medium" is a communication line for transmitting a program via a network such as the Internet or a communication line such as a telephone line, and dynamically holds the program for a short period of time. In that case, it also includes those that hold the program for a certain period of time, such as the volatile memory inside the computer system that is the server or client. Further, the program may be for realizing a part of the above-mentioned functions, and may be further realized for realizing the above-mentioned functions in combination with a program already recorded in the computer system. At least some of the functions may be realized by hardware such as integrated circuits.

(Additional note)
The present invention includes the following disclosure regarding Patent Document 1 and the disclosure of the improved invention based on the above description.

(reference)
(Invention 1)
The amount of corrosion (mg / m) that varies among the plurality of surface elements, which is a state of variation in deterioration for each of the plurality of surface elements, ascertained by the inspection performed by the inspector for the actual structure divided into the plurality of surface elements. ² ) The process of grasping and
Deterioration prediction rate (mg / m ² / hour) based on the amount of corrosion (mg / m ² ) for each of the divided plurality of surface elements and the elapsed time (year) from the time of construction to the time of the inspection. The deterioration prediction curve, which is a curve representing, is divided into a plurality of surface elements with the horizontal axis representing the elapsed time (year) from the time of construction and the vertical axis representing the cumulative corrosion amount (mg / m ² ). The process represented by the inspector based on the above inspection,
The ratio of the surface area in the actual structure represented by one of the deterioration prediction curves for each of the divided plurality of surface elements is grasped as the deterioration expression rate, and the deterministically calculated deterioration prediction speed is set to 1x speed. The slowest predicted deterioration rate that has not deteriorated at all when the above inspection is performed is set to 0 times the speed, and the fastest deterioration rate is compared with the deterioration rate deterministically calculated after grasping the deterioration distribution of the actual structure. The cumulative corrosion amount (mg / m ² ) obtained from the cumulative corrosion amount (mg / m ² ) is converted, and the degree of deterioration, which is the deterioration state of the actual structure, can be expressed over all of the divided plurality of surface elements. The speed is doubled, and the deterministically calculated deterioration prediction speed is multiplied by the magnification of the predetermined interval corresponding to the interval corresponding to the deterioration distribution of the actual structure between the slowest deterioration rate and the fastest deterioration rate, respectively. A process of obtaining a plurality of deterioration expression rates that are scattered at a magnification of the predetermined intervals from the slow deterioration rate to the fastest deterioration rate.
The actual structure is inspected by correcting the dispersed deterioration expression rate per one deterioration prediction curve so as to match the deterioration expression rate of the inspection result performed by the inspector for the actual structure. A step of obtaining a deterioration expression rate obtained by correcting the various deterioration expression rates per one deterioration prediction curve by making corrections that are consistent with the inspection results performed by the person.
A predetermined score is defined for each of a plurality of deterioration degrees representing the deterioration state of the actual structure, and the score for each defined deterioration degree is multiplied by the area ratio of the actual structure for each deterioration degree, and the total is calculated. The process of scoring deterioration and
The process of expressing the deterioration score for each of a plurality of divided surface elements as a deterioration score curve representing the deterioration prediction speed, the horizontal axis as the elapsed time (year) from the time of construction, and the vertical axis as the deterioration score.
At each inspection performed every n years (n is an integer of 2 or more) of the deterioration score curve,
Compare the degree of deterioration judged by the inspector with the degree of deterioration judged by the expert regarding the "measurement error due to the inspection result" that occurs during the inspection performed after a predetermined number of years have passed since the construction of the structure. The "measurement error due to the inspection result" obtained by correcting the inspection result for the degree of deterioration judged by the inspector using the tendency of the measurement error by the inspector, which is analyzed in advance and grasped in advance. "When,
The probability of deterioration prediction is {(1 / n) × 100}, where 0 is the cumulative probability n years before the deterioration distribution at the time of inspection and 1 is the cumulative probability at the time of the inspection which is the deterioration distribution at the time of inspection. "Time error" obtained by setting to%
The process of setting the range surrounded by the range as an "error box" and
The process of predicting the future deterioration of the structure based on the deterioration prediction included in the range of the "error box" is provided. The future deterioration of the structure using the inspection result considering the measurement error and the time error can be determined. How to predict.

(Invention 2)
It is a deterioration prediction system composed of a computer system.
The deterioration prediction curve, which is a curve representing the deterioration prediction rate (mg / m ² / hour) for each of a plurality of divided surface elements in the actual structure, the horizontal axis is the elapsed time (year) from the time of construction, and the vertical axis is cumulative. A deterioration prediction curve creation processing unit that performs processing based on an inspection performed by an inspector on the actual structure divided into a plurality of surface elements as a corrosion amount (mg / m ² ).
The ratio of the surface area in the actual structure expressed by one deterioration prediction curve for each of a plurality of divided surface elements in the actual structure is grasped as the deterioration expression rate, and the deterministically calculated deterioration prediction speed is 1x. As a result, the slowest deterioration prediction speed that has not deteriorated at all when the inspection is performed is set to 0 times the speed, and the fastest deterioration speed is compared with the deterioration speed deterministically calculated after grasping the deterioration distribution of the actual structure. Then, the cumulative corrosion amount (mg / m ² ) obtained from the cumulative corrosion amount (mg / m ² ) is converted, and the deterioration degree, which is the deterioration state of the actual structure, is spread over all of the divided surface elements. It is a double speed that can be expressed, and the deterministically calculated deterioration prediction speed is multiplied by a magnification of a predetermined interval corresponding to the interval corresponding to the deterioration distribution of the actual structure between the slowest deterioration rate and the fastest deterioration rate. , A modified deterioration expression rate creation processing unit that performs a process of creating a plurality of deterioration expression rates scattered at a magnification of the predetermined interval from the slowest deterioration rate to the fastest deterioration rate.
The actual structure is inspected by correcting the dispersed deterioration expression rate per one deterioration prediction curve so as to match the deterioration expression rate of the inspection result performed by the inspector for the actual structure. A deterioration prediction curve correction processing unit that obtains a deterioration expression rate obtained by correcting the various deterioration expression rates per one deterioration prediction curve by making corrections that match the inspection results performed by the person.
A predetermined score is defined for each of a plurality of deterioration degrees representing the deterioration state of the actual structure, and the score for each defined deterioration degree is multiplied by the area ratio of the actual structure for each deterioration degree, and the total is calculated. A deterioration score calculation processing unit that performs processing to be a deterioration score,
Deterioration score curve creation process in which the deterioration score for each of a plurality of divided surface elements is represented as a deterioration score curve representing the deterioration prediction speed, the horizontal axis is the elapsed time (year) from the time of construction, and the vertical axis is the deterioration score. Department and
At each inspection performed every n years (n is an integer of 2 or more) of the deterioration score curve,
Compare the degree of deterioration judged by the inspector with the degree of deterioration judged by the expert regarding the "measurement error due to the inspection result" that occurs during the inspection performed after a predetermined number of years have passed since the construction of the structure. The "measurement error due to the inspection result" obtained by correcting the inspection result for the degree of deterioration judged by the inspector using the tendency of the measurement error by the inspector, which is analyzed in advance and grasped in advance. "When,
The probability of deterioration prediction is {(1 / n) × 100}, where 0 is the cumulative probability n years before the deterioration distribution at the time of inspection and 1 is the cumulative probability at the time of the inspection which is the deterioration distribution at the time of inspection. "Time error" obtained by setting to%
An "error box" setting processing unit that performs processing to set the range surrounded by the range as an "error box",
A deterioration prediction system including a processing unit for predicting future deterioration of a structure based on the deterioration prediction included in the range of the “error box”.

The present invention can be used as a deterioration prediction device for structures.

A Deterioration prediction processing device for structures 1 Deterioration distribution creation unit 3 Deterioration distribution setting unit 5 Deterioration distribution setting unit 7 Deterioration rate correction coefficient multiplication unit 11 Deterioration expression rate correction unit

Claims (9)

It is a deterioration prediction device for structures.
A deterioration distribution creation unit that creates a deterioration distribution of the structure divided into a plurality of surface elements, which is obtained based on n = first inspection of the structure.
In the deterioration distribution obtained by the deterioration distribution creating unit, the first deterioration distribution setting unit that sets the first deterioration distribution in consideration of the measurement error, and the deterioration distribution setting unit.
In the first deterioration distribution in consideration of the measurement error, the second deterioration distribution setting unit for setting the second deterioration distribution in consideration of the time error and the measurement error and the time error, and the second deterioration distribution setting unit.
"Deterioration rate correction" so as to cover the degree of deterioration that is most advanced in the second deterioration distribution of the inspection result considering the measurement error and the time error set by the second deterioration distribution setting unit. Deterioration speed correction coefficient multiplication unit that can be evenly distributed up to 0x speed by multiplying by "coefficient",
The deterioration distribution in consideration of the measurement error and the time error set by the second deterioration distribution setting unit is matched with the second deterioration distribution set by the second deterioration distribution setting unit. , Deterioration expression rate correction unit that corrects the deterioration expression rate of deterioration prediction,
Have,
Hereinafter, a structure deterioration prediction device, wherein n + 1 is used, and the process is continued until n = m (m is the number of final inspections of 2 or more) and then the process is terminated.
Here, the ratio of the surface area expressed by one deterioration prediction is the "deterioration expression rate". The first deterioration distribution setting unit is
The deterioration prediction device for a structure according to claim 1, wherein a first deterioration distribution in consideration of a measurement error is set by using a measurement error correction matrix for the deterioration distribution obtained by the deterioration distribution creating unit. .. The deterioration distribution creation unit
The invention according to claim 1 or 2, wherein the degree of deterioration of the actual structure divided into the plurality of surface elements, which is the state of variation of the deterioration of the plurality of surface elements ascertained by inspection, is obtained. Deterioration prediction device for structures. The deterioration prediction device for a structure according to any one of claims 1 to 3, wherein the number of deterioration prediction elements is 100 to 1000 elements. The structure deterioration prediction device according to any one of claims 1 to 4, wherein the structure is an actual bridge. It is a method of predicting deterioration of structures.
a) The step of conducting the first inspection and grasping the deterioration distribution created in the deterioration distribution creation step, and
b) In the obtained deterioration distribution, the step of setting the first deterioration distribution in consideration of the measurement error, and
c) A step in which the deterioration distribution setting unit considers the time error in the deterioration distribution considering the measurement error and sets the second deterioration distribution in consideration of the measurement error and the time error.
d) Deterioration rate correction coefficient multiplication unit, until the definite deterioration prediction can cover the degree of deterioration that is most deteriorated in the "deterioration distribution of inspection results considering measurement error and time error", "deterioration rate" A step that multiplies the "correction coefficient" and evenly distributes up to 0x speed,
e) The deterioration expression rate correction unit sets the "deterioration expression rate" of the deterioration prediction so that the deterioration distribution of the inspection result considering the measurement error and time error and the deterioration distribution of the inspection result considering the error match. Steps to correct and
Have,
Hereinafter, a method for predicting deterioration of a structure, wherein n + 1 is used, and the process is continued until n = m (m is the number of final inspections of 2 or more), and then the process is terminated.
Here, the ratio of the surface area expressed by one deterioration prediction is the "deterioration expression rate". The step of setting the first deterioration distribution is
The deterioration prediction method for a structure according to claim 6, wherein a first deterioration distribution in consideration of a measurement error is set for the deterioration distribution obtained by the deterioration distribution creation step by using a measurement error correction matrix. .. The deterioration distribution creation step is
The claim is characterized in that the degree of deterioration of each of the plurality of surface elements, which is the state of variation of the deterioration of each of the plurality of surface elements, which is grasped by the inspection of the actual structure divided into the plurality of surface elements, is obtained. The method for predicting deterioration of a structure according to 6 or 7. A program for causing a computer to sequentially execute the processes in the structure deterioration prediction method according to any one of claims 6 to 8.

Images