JPS5837398A - Diagnosis of condition of pipe line - Google Patents

Abstract

PURPOSE:To grasp the condition of pipe line burried underground by a statistical technique by a method wherein in case where the condition of pipe line is diagnosed, data such as the lowest limit of water pressure used on the pipe line, the number of years for which the pipe line has been burried, the method by which the pipe was manufactured, the surface load applied on the pipe line, the rigidity of the pipe and the nature of the ground are used as explanatory variables and are introduced into an equation expressing as an objective variable the condition of the pipe line obtained by quantitative theoretical engineering. CONSTITUTION:In the equation yi is supposed (wherein y designates an objective variable and x(j) designates an explanatory variable). Then the explanatory variable x(j) is divided into several items x(j, k) and the weight of each of the items is calculated in such a manner that the error (e) between the actual measured value and the calculated value of the item becomes minimum. Note that in the equation, (i) designates a sample number, (j) designates an explanatory variable and (k) designates an item number. The items to be introduced into the equation at least include the water pressure used on the pipe line, the number of years for which the pipe line has been burried, the method by which the pipe was manufactured, the surface load applied on the pipe line, the rigidity of the pipe and the nature of the ground and the objective variable is obtained on the bases of these items.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for indirectly diagnosing the condition of a buried pipeline using statistical techniques.

Conventionally, the method of grasping the current state of existing buried pipelines was to directly grasp the stress state and corrosion status of the pipes, but this method is difficult to determine. In reality, it was not implemented unless there were special circumstances. However, in order to prevent accidents, it is necessary to know the condition of the pipeline even if only in general, and it has been desired to develop a simple method for understanding the condition of the pipeline. In addition, if the condition of the pipeline can be ascertained, it will also be possible to determine how many years the pipeline will remain in use, and to what extent it can withstand usage conditions.

From this point of view, the present invention aims to provide a method for easily diagnosing pipes using statistical methods.

Easily obtainable statistical data for conducting pipe diagnosis using statistical methods is data related to government control, and the data that can be obtained from this includes accident situations, burial conditions, corrosion conditions, etc. There are many things that are relatively difficult to quantify.

Therefore, it is necessary to adopt a method that allows qualitative phenomena to be grasped in a comprehensive manner, and the present invention employs a number theory (2) that is considered to be inappropriate for diagnosis of pipelines.

Similar to multiple regression analysis, the political blitz theory type (1) assumes the following equation of the objective variable y and multiple explanatory variables x(j) (hereinafter referred to as items), and further sets the explanatory variables. This method classifies Cj, k) into several items

Category No., and its item No. and category No. K belong to F'ix, (j, -) = 1
, those that do not belong to xi(j, &)'' are set to 0, so that the explanatory variable does not have to be a numerical value in the multiple regression analysis of X, and a qualitative modification can be used.

Next, one embodiment of the method of the present invention will be described. Table IK attached at the end of the sample #i specification used is shown.

In addition, in Table 1, soil quality (1) is (2) +3
) Soil types that cannot be classified into (4) were classified. Also, soil corrosivity is based on ANSI soil quality evaluation standards! The value is determined by $, and the higher the value, the more corrosive it is. In addition, the objective variables were calculated based on the following criteria, focusing on the expected occurrence of the accident and the magnitude of the accident. In other words, the evaluation of the condition of the pipeline is divided into basic points, supplementary points, and corrosion points.
Based on the addition of those evaluation points, ■ Basic points la) If there is an accident, 75 points (similarly, excluding cases where the accident occurred due to external force due to work 1 (fb)) If a leak occurs in the pipe. 5,0 points (c
l If there is no accident or water leakage 1) points (2) Supplementary points faJ If cracks have occurred in the pipe 10 points (b
l If a small rack has occurred in the pipe
8 points (c) If water leaks from the joint 8 points (d) If the joint has a large body 6 points (e) If the water leaks are very small but occurs in many places 4 points) If corrosion has occurred in the fitting and its accessories 2 points (g) If there is no abnormality other than corrosion of the pipe body or pongee Km
0 points (3) Rotten pot, white (, I) If the pipe has a shell hole, lOα (
b) If the maximum pitting depth on the inner and outer surfaces is 75% or more of the pipe thickness: 8 points: 75-50% 6 points: 150-25%
In case of 4 points 25-5%
In the case of 2 points of 5 to 0%, the model formula obtained by performing the above-mentioned calculation using a large computer based on the conditions of 1 boiled shellfish H is shown in the following formula (2). In addition, in the calculation process, items with a small influence were sequentially omitted in the calculation. Calculation was performed using the categories of soil corrosivity #-1(r) to (4), and the category weight of (1) was A; the value was smaller than that of (2). (1)<2) was set as one category 1 no.

y (estimated value) = (2tx,, + 25.78'x1□+232fyx
13 + 28.63'x14 ) + (1) - X, 2
．． +06(?x22+0.42・X23)+(0-X3
, +0.57・X32) +(0・% 41 3.22 X42) 1−(0・X5
1-1.29・X52-5.36・X53)+(0'X
61-0.07・X61 z2o・X63)−+−(
1,96-'X7, +0-X7□-0,02-X73-
1.35-X74)+(o・xgl 3.16-X8
2-5.95-X838.20・X64)to(0・'x
, 1-3. o6・xg2-4. o2・xg3)+(”
”101-14s・xl (123,29・Xl.3)+
(0”111 3.84・x1L2 1.94・XH3
) ten (-0' X 121 +1.08 ・X 122
)"(0'X131-080'X132-119・X
133)---+2+Also, the difference between the maximum and minimum category weights of each item given by this formula (2), that is, the category weight bandwidth, is 1 & 2 f
It becomes il.

Table 2 The band width of each item category indicates the degree of influence of each item on the objective variable, and by adopting items with large bandwidths as items, it becomes possible to correctly diagnose the conduit. Displayed in descending order of band width, water pressure used, years of burial, manufacturing method, road load, pipe hardness, soil quality, etc. At least these items must be adopted in pipe diagnosis. There is. However, the value and order of the bandwidth of these items, the weight of each category, etc. vary depending on the type of sample, the selection of categories, the concept of setting the calculation criteria for the objective variable, etc. However, as long as pipe diagnosis is to be performed, it seems essential to adopt the above items.

In addition, the multiple correlation coefficient R of the above calculation result is 8832% contribution half 7
It was 800%. , Therefore, it is considered that the objective can be explained by the explanatory variables. $N First, the actual measured value, that is, the objective variable calculated using the calculation criteria described above for the actual state of each sample, and the estimated value, that is, the estimated objective variable calculated by equation (2) based on the explanatory variables of each sample. In FIG. 1, which shows the relationship, 'f11. Hunting is 45
The line is drawn at 17 t fy (dispersion value of the actual measured values), and the dashed-dotted line is drawn at t ty fV'';'.Here, e is the simple correlation coefficient between the actual measured value and the estimated value. The following can be considered.Assuming that the measured values follow a normal distribution, the probability of falling between ±By among n measured values is α68n
, and since σy' supports the variance of the error in the estimated value, the error in the estimated value relative to the actual measured value has a probability of 68%.
;y is considered to be within the range of +/-3.16. When estimated values were calculated for accident cases other than the sample, they matched the actual values at a rate of about 90%.

When applying the above calculation results to actual pipeline diagnosis,
The condition of the conduit can be determined according to the following criteria:

(a) 15±3 or more is the range where a major accident may occur.

(b) The range in which the possibility of an accident occurring is considered to be between 15±3 and 7±3.

(c) 7±3 or less, a range where the possibility of an accident occurring is low.

Note that the calculation criteria for the objective variables were selected through repeated calculations so as to be able to provide such judgment criteria and to maximize the multiple correlation coefficient of the calculation results.
In addition, the force degree was selected by repeating the process of classifying the force into as many categories as possible, calculating them, and then omitting those that have a small effect and recalculating them. However, the samples adopted in the above examples are not a collection of data necessary for establishing the method of the present invention, but use only the data recorded in limited past accident records. Therefore, as the number of samples increases and the data becomes more precise, more accurate judgments will become possible. but,
The method of the present invention is a diagnostic method using a model created using a statistical method, and cannot make absolute judgments about individual pipes, but rather grasps the condition of the pipes as a whole! It doesn't matter.

According to the method of diagnosing the unexploded Ellj pipe W, as is clear from the above explanation, the condition of the pipe can be statistically evaluated by calculating using data that can be obtained relatively easily, such as the water pressure used and the number of years it has been buried. It is extremely important that this can be estimated based on objective evidence and contributes to the prevention of accidents involving buried pipelines.

[Brief explanation of the drawing]

The drawing shows the actual total number 1 of objective variables in the pipeline diagnosis method of the present invention.
It is a graph showing the relationship between bean paste and estimated values. Agent Yoshihiro Morimoto

Claims (1)

[Claims] l Use at least the water pressure used, the number of years buried, the manufacturing method of the pipe body, the road surface load, the hardness of the pipe body, and the soil quality as explanatory variables, and use the condition of the pipe line determined by the failure theory ■ as the objective variable. 1. A method for diagnosing pipes, which comprises estimating the condition of the pipe by inputting data of the pipe to be diagnosed into a calculation formula expressed as