WO2007077467A1

WO2007077467A1 - Method and apparatus for planning renewals and maintenance of utility piping systems in order of succession

Info

Publication number: WO2007077467A1
Application number: PCT/HU2006/000001
Authority: WO
Inventors: Béla TOLNAI
Original assignee: Tova-Partner Kft.
Priority date: 2006-01-06
Filing date: 2006-01-06
Publication date: 2007-07-12

Abstract

Method for planning renewal or maintenance of utility piping systems in order of succession. During the method decisions are made by estimating the physical parameters and the environmental conditions of the system components. The gathered information is registered and is used for generating generalized operating data for entering in a fuzzy logic arrangement (FL) containing multi-level evaluating blocks (1-12). Evaluating rules relating to said generalized operating data are assigned to a respective evaluating block for determining failure probability factors. The failure probability factor is multiplied by the coefficient of damage assigned to a respective system component in order to determine the risk (RSK). The system component having the higher risk are listed first in the maintenance plan.

Description

A

METHOD AND APPARATUS FOR PLANNING RENEWALS AND MAINTENANCE OF UTILITY PIPING SYSTEMS IN ORDER OF SUCCESSION

The present invention relates to a method for planning renewal or maintenance of utility piping systems in order of succession; during the method decisions relating to the components of the utility piping system are made by estimating the state of the components; the physical parameters of the components, the environmental conditions of the components and the history of the individual components are registered and are used for generating generalized operating data.

Public utilities such as water-pipe systems of big cities are usually large-sized. The same holds true of the gas-distributing pipelines, electric cables or sewer systems as well. Typically, the presently existing pipe-systems have been developed over long years, therefore these pipelines are extremely heterogeneous in respect of their age and the material used. Vague information relating to the state of the pipes laid under ground can be obtained only indirectly. Particularly, when concerning the whole system, this mass of data is incomplete for the most part.

Generally, water-pipe systems or networks are positioned in public properties. Water-pipes or other conduits may run for example under busy roads. A possible crack or other failures of the pipe may cause significant loss. When making decision on the rehabilitation of the network, considering merely the technical elements is not enough, because due to the local circumstances different factors must be allowed for. Therefore decision making is highly complex, all the viewpoints must be considered as far as possible.

Practically, instead of quantitative relations only qualitative estimates can be established among the input variables. Rules are based on empirical factors. However, these empirical factors are available only in a short time domain considering the life cycle of the pipe relative to the lifetime of a human being. The useful life of a pipe is approximately 100 years, in case of some pipe-materials it may be less or even more. Usually, the amount of money that can be spent on rehabilitation is limited. In case of pipes which can not be marked out for rehabilitation it is important to know the extent of the risk involved. A deliberate decision can be made by the maintainer of the system if estimate data relating to the whole system are available.

There is a need for implemeting an information technology which guarantees the maximal exploitation of the network infrastructure investment. A complex planning method and means which form an integral part of the management system of the company operating the public utility system can solve the problem. Preferably, since the matter in question is planning or scheduling rehabilitation of a public utility system, the aforementioned means is provided in geographic information system. In the sense of information technology this implies integration of the company management and the technical information system based on digital map.

Patent application US 2001001149 discloses a system in which fuzzy logic is used for determining a possible boiler tube leak event. This solution is applicable in case of system components running above ground as their state can be estimated easily, a possible failure can be well prognosticated. When these system components are positioned under ground, empirical and probability factors are of greater importance and maintenance of the components requires different technology.

Risk assessment is very important for making a decision on maintenance-work or renewals. Usually the investment plan of a company can be put into action as long as the expenses do not exceed the estimated costs.

The object of the present invention is to provide a method in which evaluating rules can be made by using available generalized operating data and on the basis of these rules output data can be generated which is used for featuring the risk when planning renewal of a utility piping system and determining the optimal order of the work. In one aspect the present invention provides a method for planning renewal or maintenance of utility piping systems in order of succession, during the method decisions relating to each component of the system are made by estimating the state of the components, the method comprising the steps of: a)registering the physical parameters, the environmental conditions of each component, registering the events associated with the individual components, and using the registered information for generating generalized operating data; b) entering the generalized operating data as input data in a fuzzy-logic arrangement comprising multi-level logical evaluating blocks; c) generating evaluating rules relating to the generalized operating data, each of the rules are assigned to a respective evaluating block; d) determining - on the basis of the evaluating rules - failure probability factors by means of the evaluating blocks (1-12) using the generalized operating data, and determining the failure probability by fuzzy summing-up said factors; e) assigning a coefficient of damage to each component of the system, the coefficient of damage being proportionate to the expected total value of the direct and indirect damages caused by failures; f) multiplying the failure probability by the coefficient of damage and assigning the product as risk to the respective component of the system; g) determining the plan containing the order of renewal or maintenance of the utility piping system wherein the components having the higher risk are listed first.

In another aspect the present invention provides an apparatus for implementing the above method. The apparatus is adapted to receive generalized operating data as input data, and contains multi-level logical evaluating blocks based on fuzzy logic, a respective evaluating rule is assigned to each of the evaluating blocks, all the evaluating blocks have at least one output signal representing failure probability, the output signal is transmitted to one of the inputs of a multiplier, and a signal representing the coefficient of damage assigned to the components of the system is transmitted to the other input of the multiplier, and a signal representing the risk is generated at the output of the multiplier.

Yet in another aspect the present invention provides a computer program product comprising program means for performing the steps of the method.

If the evaluating method according to the invention is considered as a black box then it can be described as a system having several inputs and a single output. Besides the great number of inputs another characteristic feature is that a considerable part of the data to be entered can not be measured. Practically, a few of the data are derived from measurement, while in case of the others only qualitative classification can be determined which is often subjective. These parameters can merely be characterized as "low-average-high". A further problem is that the regularities which can be drawn up between the different variables are incoherent.

To solve problems of this type models built on the basis of fuzzy sets can be well used.

The stability of non-linear regulations by means of a suitably designed fuzzy logic can be guaranteed, therefore a system based on fuzzy logic is used in the present invention.

For carrying out the evaluating model according to the invention failure probability factors are determined. Then a coefficient of damage is assigned to each component of the system proportionately to the expected total value of the direct and indirect damages caused by failures, and the plan in which the renewal or maintenance of the utility piping system is scheduled is determined on the basis of the result produced by multiplying the failure probability by the coefficient of damage wherein the result represents the measure of risk.

The structure of the model is similar to other logical circuits. Fuzzy evaluating blocks are known in the art. They are multiple-valued logical elements and they are different from binary logic in that they use not only "1" and "0" but also other numbers between "1" and "0". Signals corresponding to parameters (referred to as generalized operating data) which belong to data representing three main types of the individual system components are transmitted to inputs of the evaluating blocks. Data describing the system components in any respects may represent generalized operating data. However, three main types of the generalized operating data are determined according to the invention. They are as follows: a) the physical parameters of the system components (e.g.: material, age, size, type, etc.); b) the environmental parameters of the system components especially the parameters of the soil (e.g.: type of soil, depth, aboveground traffic load, etc.); c) the history of the system components (e.g.: failure, replacement, repairs, etc.) The number of these parameters can be as high as 40-80.

Preferably, the generalized operating data may be stored and at the same time displayed in a geographic information system. The geographic information system approach is advantageous because the individual system components can be directly correlated in such a manner that the relative position of the components is also considered.

A detailed description of the method according to the invention will now be disclosed with reference to the accompanying drawings in which:

Figure 1 shows the method used in the company management for planning renewals and maintenance of utility piping systems;

Figure 2 is a table used for determining the measure of a risk as the function of the failure probability factor and the coefficient of damage in accordance with the invention;

Figure 3 shows the structure of an exemplary fuzzy logic for planning reconstruction of a water-pipe system in accordance with the invention;

Figure 4 shows integration of the evaluating cycles into the technical data environment; and Figure 5 is a table showing the generalized operating data and evaluating rules in accordance with the method of Figure 3.

The company management of the public utility works may be considered as methodology during which the jobs to be done with the system components and the order of the jobs are determined. A system component having a higher failure probability factor will be placed forward. The jobs must be planned on company level. Scheduling or planning the maintenance-work is based on risk assessment performed for the system components simultaneously.

Figure 1 shows the method used in the company management for planning renewal and maintenance of utility piping systems. In the first step the state of the system components is registered and monitored. In addition to registering the state of the system components other circumstances (e.g.: load conditions) are also registered. Events are used for pointing to the state indirectly(e.g.: pipe-crack statistics). Environmental or external features (e.g.: electro-corrosion effects, traffic load, works done on public properties) may serve as data for predection of a possible pipe-crack. Internal analyses (e.g.: analysis of hidraulics, the level of shutdown and the consumers affected by the shutdown) are also important for evaluation. Aimed survey of state serve for making the indirect information more precise. In case of utility piping systems laid under ground the aimed survey of state involves uncovering, disjoining the pipes which is an expensive way for gathering information, therefore information may be obtained during repairs or occurrences.

The obtained data are assigned to a related system component (e.g.: a section of a pipe) and used for the evaluation. These data are collected and updated on a platform of geographic information system. Evaluation of the system components takes place simultaneously using the same evaluating method. The risk is determined during evaluation and the plan of maintenance and investment is drawn up on the basis of the risk factor. A system component having a higher risk factor will have higher priority in the order of succession of renewals or maintenance-work. The sum of the risks calculated in this manner may be used for making a long-term development plan.

According to the invention risk is calculated by multiplying the failure probability by the damage caused by the failure. Naturally, risk may be calculated by using other functions, for example products raised to a power may be used, where the risk would be exponentially proportional to one of the multiplicands. However, according to the invention every kind of weighting is performed before multiplication.

The table of Figure 2 shows the risk as a function of the failure probability factor and the coefficient of damage for a given system component. As it can be seen, the risk is high even if the failure probability factor is low, but the extent of damage caused by the failure is great. The risk is high when the failure probability is high, even if the extent of damage is small.

Both direct and indirect damage are taken into consideration during assessment of the damage. Direct damage is to be understood as the damage resulting from the shutdown and the cost of repairs, indirect damage means the damage caused in the environment and in the neighbouring objects.

The outputs of the fundamental units applied in the fuzzy-logic arrangement FL, i.e. evaluating blocks 11-12 or multiplier 13, in other words the result of the evaluation may be used as inputs to a next level of evaluation. In this manner an optionally composed apparatus may be constructed.

A model of this type is shown in Figure 3 in which a simplified structure can be seen where the utility piping system is a water-system. The fuzzy-logic arrangement FL is adapted for receiving generalized operating data GD as input data. In a multilevel logical structure it comprises evaluating blocks 1-12. A respective evaluating rule is assigned to each of the evaluating blocks 1-12. In the present example blocks 2 and 3 evaluate parameters of operation and service parameters, block 4 handles financial losses including direct damages resulting from a possible shutdown and failure correction. Block 1 evaluates the shut off time of the system component. The output of block 1 is connected to one of the inputs of block 5 which is arranged on a next level. Block 5 has several inputs and controls the extent of possible damage caused in the environment and in the neighbouring objects, i.e. the indirect damage which depends on the shut off time handled by block 1.

Blocks 6, 7, 8, 9 and 10 evaluate the pipe characteristics, operating characteristics, corrosive effects, external effects and dynamic load respectively. The outputs of these blocks are connected to the inputs of evaluating block 12 which determines the failure probability which is the probability of a crack according to the present example. The evaluating blocks of Figure 3 are determined in the following table:

Similarly, outputs of evaluating blocks 2-5 are connected to inputs of evaluating block 11 which determines the coefficient of damage. In the present example it determines the direct and indirect damages caused by the crack.

Multiplier 13 multiplies the failure probability by the coefficient of damage. Weighting of the two multiplicands has been performed by evaluating blocks 1-10 on the preceding level. The product represents the risk RSK.

Internal functioning of evaluating blocks 1-12 may also be determined by rules expressed in words. For example: "If inputi is A and/or input2 is B, then the output is C", where A₁B and C can be assigned to respective fuzzy-logic values.

Properly speaking, the rules are based on empirical factors, experiences. According to the strategy of the company the individual rules may have weighting factors in order to highlight their priority. Application of benchmarking elements can be useful when determining the rules and their priority.

The labellings of the individual evaluating blocks shown in Figure 3 represent the typical parameters taken into consideration during evaluation. Practically, the number of the input data may be as high as 40-80.

In Figure 4 integration of the evaluating cycles into the technical data environment is shown. The evaluating method and the implementing apparatus i.e. the fuzzy-logic arrangement FL obtains data from the database of a technical information system TD through an interface, and transmits the calculated risk RSK values to a technical planning platform TP through an interface. It is done cyclically as many times as many system components, e.g.: maintenance units, pipe sections, etc. exist.

After determining the risk RSK the data may be transmitted through the technical planning platform TP to a geographic information subsystem for example within the frame of the maintenance plan. The geographic information subsystem handles these data included in a map. In this manner in case of under ground piping systems modifications of the system e.g.: replacement of a system component or extension of the system will be available for other applications.

Figure 5 is a table showing the generalized operating data and evaluating rules in accordance with the exemplary method of Figure 3 adapted for a water-pipe system.

It will be readily appreciated by those skilled in the art that a similar evaluating system may also be used with pipes other than water-pipes. It is particularly advantageous in case of under ground utility piping systems which can not be examined directly and the operator of the system can obtain information only indirectly or by considering the empirical factors. When integrated with geographic information system it can be an effective means for the engineer planning the rehabilitation. Due to the continuous feedback and recording of conditions the system can learn and can be made dynamic. As the number of the empirical factors increases, supplementary information for modification can be added to the evaluating model according to the invention.

Claims

1. Method for planning renewal or maintenance of utility piping systems in order of succession, during the method decisions relating to each component of the system are made by estimating the state of the system components, the method comprising the steps of: a) registering the physical parameters, the environmental conditions of each system component, registering the events associated with the individual system components, and using the registered information for generating generalized operating data; characterized in that b) entering said generalized operating data as input data in a fuzzy logic arrangement (FL) comprising multi-level logical evaluating blocks (1-12); c) generating evaluating rules relating to said generalized operating data, each of said rules are assigned to a respective evaluating block; d) determining - on the basis of said evaluating rules - failure probability factors by means of said evaluating blocks (1-12) using said generalized operating data, and determining the failure probability by fuzzy summing-up said factors; e) assigning a coefficient of damage to each system component, said coefficient of damage being proportionate to the expected total value of the direct and indirect damages caused by failures; f) multiplying said failure probability by said coefficient of damage and assigning the product as risk (RSK) to the respective system component; g) determining the plan containing the order of renewal or maintenance of said utility piping system wherein the system components having the higher risk are listed first.

2. Method according to claim 1 characterized in that said utility piping system is an under ground water-pipe sytem or gas-pipe system or sewer system or electric cable system, and said system components are pipes or pipe sections or fittings customarily used for connecting, detaching, distributing the pipes.

3. Method according to claims 1 or 2 characterized in that said generalized operating data are handled in a geographic information system.

4. Method according to any of claims 1-3 characterized in that one or more of the steps taken in b)-e) are repeated.

5. Method according to any of claims 1-4 characterized in that said generalized operating data assigned to said system components are updated when an operation is effected on said system components.

6. Method according to claims 4 or 5 characterized in that said updated generalized operating data are entered in said geographic information system.

7. Method according to any of claims 1-6 characterized in that said rules applied in said evaluating blocks are determined on the basis of information gathered experientially in case of a given type of utility piping system.

8. Apparatus comprising means for executing the steps of the method according to any of claims 1-7.

9. Apparatus according to claim 8 characterized in that it comprises a fuzzy-logic arrangement (FL) adapted to receive generalized operating data as input data and contains multi-level logical evaluating blocks (1-12), a respective evaluating rule is assigned to each of the evaluating blocks (1-12), all the evaluating blocks (1-12) have at least one output signal representing failure probability, the output signal is transmitted to one of the inputs of a multiplier (13), and a signal representing the coefficient of damage assigned to a system component is transmitted to the other input of the multiplier (13), and a signal representing the risk (RSK) is generated at the output of the multiplier (13).

10. Computer program product comprising program means for executing the steps of the method according to any of claims 1-7.