WO2017043958A1

WO2017043958A1 - Method for obtaining failure prognostic information of electrical power equipment

Info

Publication number: WO2017043958A1
Application number: PCT/MY2016/050054
Authority: WO
Inventors: Hui Mun Looe; Gordon Rees Jones; Anthony Grayham DEAKIN; Joseph William Spencer; Ezzaldeen Yousef ELZAGZOUG
Original assignee: Tnb Research Sdn. Bhd.; University Of Liverpool
Priority date: 2015-09-09
Filing date: 2016-09-05
Publication date: 2017-03-16
Also published as: MY175958A

Abstract

The present invention provides a method (100, 200) for obtaining failure prognostic information of electrical power equipment based on degradation of dielectric insulation fluid that is immersed in components of said power equipment that include electrical windings and tap-changers comprising a step (107) of computing a rate of change with respect to time of each of a plurality of sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters obtained for a plurality of time instances and which have been grouped according to parameter type of insulation fluid degradation, such as "Dissolved Gas Analysis (DGA)", "Acidity, Water Content, Dielectric Strength (AWD)" and a step (108) of assigning scores to said plurality of sets of tertiary (106a, 106b, 106c) and secondary (105a. 105b, 105c) first order chromatic parameters at said plurality of time instances and the rate of change with respect to time of said sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters.

Description

METHOD FOR OBTAINING FAILURE PROGNOSTIC INFORMATION OF ELECTRICAL POWER EQUIPMENT

The present invention relates to the field of equipment condition monitoring. More particularly, the present invention relates to the field of electrical equipment condition monitoring. Most particularly, the present invention relates to a method for obtaining failure prognostic information of electrical equipment that utilize dielectric insulation fluid. BACKGROUND OF THE INVENTION

The use of fluid filled electrical power equipment is widespread in the electrical utility industry. Such electrical power equipment include but are not limited to step-down and step-up power transformers, load tap changers, switch gears and the like, which are usually filled with a dielectric insulation fluid such as insulation oil.

Operational faults in said dielectric insulation fluid filled electric power equipment is predicated by a degradation of said dielectric insulation fluid. The degradation of said dielectric insulation fluid, is usually detected by the presence of certain dissolved gasses in the insulation fluid which result from the process of degradation of said dielectric insulation fluid, which in the power or electric utility industry is determined by way of a process known as "Dissolved Gas Analysis" or better known as DGA in abbreviated form. Aforementioned presence of dissolved gases in dielectric insulation fluid may indicate operational faults such as arcing, pyrolysis, corona discharge and the like. Dissolved Gas Analysis (DGA) performed on electrical power equipment insulation fluid, on its own is often inconclusive in relation to providing a measure of insulation fluid degradation. In particular, as mentioned in the background of US 6,928, 861 B1 , equipment breakdown indicated by operational faults such as arcing and corona discharge as well as resulting deterioration of equipment components, is not directly indicated by the presence of dissolved gasses in the insulation fluid. As such other tests such as tests to indicate moisture, acidity, dielectric-breakdown and colour of insulation fluid are carried out on electrical power equipment insulation fluid along with said Dissolved Gas Analysis (DGA) to provide a more conclusive indication of the state of the insulating fluid.

In addition to the above, it is usually difficult to interpret the combined test results of Dissolved Gas Analysis as well as tests to indicate acidity, moisture content and dielectric strength to determine the likelihood that the electrical windings or conductive components of an electrical power equipment immersed in insulation fluid, fails. This difficulty stems from the fact that a large number of parameters must be considered and evaluated by a skilled and experienced engineer before an acceptable prognosis of the electrical power equipment having conductive components immersed in said insulation fluid can be provided.

The above problems of prior art methods of failure prediction, i.e. prognosis of faults in electrical power equipment have led to research involving new methods of failure prediction of aforesaid electrical power equipment to obtain better results in terms of failure prediction. Foremost among these methods, is chromaticity processing. Chromaticity processing is the name given to the application of sets of non-orthogonal weighted integrals to signals distributed across a measurement range and the subsequent transformation of the integral quantities obtained to give parameters summarising certain characteristic of the distribution.

The many advantages of chromaticity processing include useful information retention with high robustness to noise and reduction of the complexity of prognostic monitoring of electrical power equipment based on dielectric insulation fluid degradation data, as it reduces the individual consideration of the large number of parameters required for prognostic monitoring such as Dissolved Gas Analysis data, acid number, dielectric strength and moisture content data as well as colour data of said dielectric insulation fluid to a manageable number of chromatic parameters which hence eliminate the need for considerable skill and experience required in order to provide an acceptable prognosis of an electrical power equipment based on dielectric insulation fluid degradation. Prior art methods, in essence disclose chromaticity processing of data to provide information for prognostic monitoring of a system or process in general. Said prior art methods, however, do not provide a method directed specifically toward prognostic monitoring of electrical power equipment based on insulation fluid degradation that takes into consideration not only of Dissolved Gas Analysis (DGA) data, but also data from other tests such as tests that indicate acidity, dielectric strength and moisture content of said insulation fluid as well as colour of said insulation fluid.

Moreover prior art methods for non-orthogonal prognostic monitoring appear to be complicated as these methods entail the second order chromaticity processing of a plurality of first order chromatic parameters at different instances of time to provide a second set of chromatic parameters and chromatic maps as well as polar plots of said second set of chromatic parameters, which would inadvertently lead to a computationally complex method when implemented as a computer based implementation of the method, which in turn leads to increased cost. This computational complexity stems from the large amount of first order chromatic parameters for a plurality of time instances that have to be subject to second order chromaticity processing.

In view of the above, it would be advantageous if a method for obtaining prognostic information of electrical power equipment based on insulation fluid degradation that takes into consideration not only Dissolved Gas Analysis (DGA) data, but further takes into consideration Acidity, Water content, Dielectric Strength (AWD) and colour information utilizing chromaticity processing to provide acceptable prognosis of said electrical power equipment based on insulation fluid degradation is conceived.

It would be further advantageous if a simpler method for obtaining prognostic information of electrical power equipment based on insulation fluid degradation that is simpler in terms of computational complexity than prior art methods that utilize chromaticity processing be conceived. SUMMARY OF THE INVENTION

The following presents a simplified summary of the invention in order to provide a basic understanding of some aspects of the invention. This summary is not an extensive overview of the invention. This summary is not intended to identify key/critical elements or essential features of the claimed subject matter. Its sole purpose is to present some concepts of the invention in a simplified form as a prelude to the more detailed description that is presented later. It is an advantage of the present invention to provide a method for obtaining failure prognostic information of electrical power equipment based on dielectric insulation fluid degradation that takes into consideration not only Dissolved Gas Analysis (DGA) data, but further takes into consideration AWD ( Acidity, Water/Moisture Content and Dielectric Strength) data of insulation fluid and/or insulation fluid colour information utilizing chromaticity processing to provide an accurate prognosis of said electrical power equipment based on insulation fluid degradation.

It is an advantage of the present invention to provide a method for obtaining failure prognostic information of electrical power equipment based on dielectric insulation fluid degradation that is simpler in terms of computational complexity than prior art methods based on chromaticity processing.

In one aspect, the present invention provides a method for obtaining failure prognostic information of electrical power equipment based on degradation of dielectric insulation fluid that is immersed in components of said power equipment that include electrical windings and tap-changers comprising of: a first stage in which a plurality of discrete signals are grouped according to parameter type across one or more parameter types of insulation fluid degradation and chromatically processed to obtain a plurality of sets of primary, secondary and tertiary first order chromatic parameters that correspond to each of said one or more parameter types of insulation fluid degradation at a plurality of time instances and in which scores are assigned to each chromatic parameter and a rate of change with respect to time of each chromatic parameter of said plurality of sets of secondary and tertiary first order chromatic parameters at a plurality of time instances; and a second stage in which the scores of each chromatic parameter and the rate of change of each chromatic parameter with respect to time of said plurality of sets of secondary and tertiary first order chromatic parameters at a plurality of instances of time are combined across a given parameter type of insulation fluid degradation and said combined score for each parameter type are collectively subjected to second order chromaticity processing to provide a set of second order chromatic parameters at a plurality of time instances, that are utilized to provide an indication of the insulation fluid degradation and the health of components of an electrical power equipment immersed in the insulation fluid being monitored. In accordance to a preferable embodiment, the present invention provides a method for obtaining failure prognostic information of electrical power equipment based on degradation of dielectric insulation fluid that is immersed in components of said electrical power equipment that include electrical windings and tap-changers comprising the steps of: a first step of obtaining a plurality of discrete signals obtained from a plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters that include discrete signals corresponding to "Dissolved Gas Analysis" parameters, acidity of said insulation fluid, dielectric strength of said insulation fluid, moisture content of said insulation fluid and/or colour of said insulation fluid, at a plurality of time instances; a step of grouping the plurality of discrete signals obtained from said plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters, based on parameter type such as Dissolved Gas Analysis (DGA) parameters, AWD (Acidity, Water Content and Dielectric Strength) parameters, and/or colour index (CI) parameters; a step of normalizing the plurality of discrete signals obtained from said plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters that correspond to the AWD (Acidity, Water/Moisture Content and Dielectric Strength) parameter type; a step of applying a plurality of non-orthogonal Gaussian approximated responses and computing the result of multiplication of the weighted integral of said plurality of non-orthogonal Gaussian approximated responses and said discrete signals grouped according to parameter type obtained in the second and third steps for said plurality of time instances to hence provide a plurality of sets of primary first order chromatic parameters at said plurality of time instances that correspond to the discrete signals grouped based on parameter type of insulation fluid degradation; a step of algorithmically converting the plurality of sets of primary first order chromatic parameters at said plurality of time instances that are grouped according to parameters based on parameter type of insulation fluid degradation into corresponding sets of secondary first order chromatic parameters that provide better interpretability of information through partitioning each primary first order chromatic parameter of said sets of primary first order chromatic parameters into components of distinct character; a step of scaling each individual chromatic parameter of a given set of primary first order chromatic parameters that correspond to a given parameter type with a reciprocal of an integer multiple of a secondary first order chromatic parameter of a corresponding set of secondary first order chromatic parameters at said plurality of time instances to obtain a tertiary set of first order chromatic parameters for each parameter type of insulation fluid degradation parameters at said plurality of time instances; and a step of computing a rate of change with respect to time of each of said sets of tertiary and secondary first order chromatic parameters obtained for said plurality of time instances and which have been grouped according to parameter type of insulation fluid degradation.

In accordance to a preferable embodiment, of the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the method further includes a step of assigning scores to said plurality of sets of tertiary and secondary first order chromatic parameters at said plurality of time instances and the rate of change with respect to time of said sets of tertiary and secondary first order chromatic parameters, said sets of tertiary and secondary first order chromatic parameters and corresponding rate of change with respect to time of said sets of tertiary and secondary first order chromatic parameters categorized according to parameter type of insulation fluid degradation and are assigned scores based on empirical data of parameters categorized according to said parameter type of insulation fluid degradation.

In accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the method further includes a step of combining scores assigned to said plurality of sets of tertiary and secondary first order chromatic parameters as well as the rate of change with respect to time of said plurality of sets of tertiary and secondary first order chromatic parameters at a plurality of time instances that fall under a given group of insulation fluid parameter type (DGA, AWD and/or CI) to obtain a combined score of insulation fluid degradation for each parameter type at a given instance of time and computing the result of multiplication of the weighted integral of a plurality of non-orthogonal Gaussian approximated responses and said combined score for each parameter type to provide a set of second order chromatic parameters at a given instance of time that in turn provides an overall indication of insulation fluid degradation and hence health of components such as electrical windings and tap changers which are immersed in the insulation fluid under test, of a given electrical power equipment.

In accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the number of Gaussian approximated non-orthogonal responses utilized is numerically equal to three responses, which consequently result in three primary first order chromatic parameters, 'R' , 'G' , 'B' for each plurality of discrete signals that are grouped according to a parameter type of dielectric insulation fluid degradation parameters (i.e. such as DGA , AWD and CI).

In accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the three primary first order chromatic parameters are converted into corresponding three secondary first order chromatic parameters namely 'Hue', 'Lightness' and 'Saturation' denoted as Ή', 'L' and 'S' , that provide better interpretability of information through partitioning said primary first order chromatic parameters into components of distinct character. More particularly in accordance to said preferable embodiment, the conversion of said three primary first order chromatic parameters, 'R', 'G', 'B' for each plurality of discrete signals that are grouped according to a parameter type of insulation fluid degradation parameters (i.e. DGA, AWD and/or CI) into three corresponding secondary first order chromatic parameters, is executed utilizing the following transformation formulae:

C 60 (G - S)/(max(ff, G, B) - min(ff, G, B)), if max(ff, G, B) = R

H = \ 60(2 + (B - ff))/(max(ff, G, B) - min(ff, G, B)) , if max(ff, G, B) = G (60(4 + (ff - G))/(max(ff, G, S) - min(ff, G, S)), if max(R, G, B) = B

/R + G + B\

L = (—3—) max(ff, G, B) - min(ff, G, B)

~ max(ff, G, B) + min(ff, G, B)

In accordance to a preferable embodiment, the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the step of scaling each individual primary chromatic parameter of a given set of primary first order chromatic parameters that correspond to a given parameter type comprises of multiplying each of said individual primary chromatic parameters with a reciprocal of "3L", in which "L" represents the "Lightness" secondary first order chromatic parameter of a set of secondary first order chromatic parameters that correspond to said set of primary first order chromatic parameters. More particularly, in accordance to said preferable embodiment, the conversion of said three primary first order chromatic parameters, 'R', 'G', 'B' for each plurality of discrete signals that are grouped according to a parameter type of insulation fluid degradation parameters (i.e. DGA, AWD and CL) into three corresponding tertiary first order chromatic parameters ( i.e. R', G', B') is executed utilizing the following transformation formulae:

< _ ^R < _ ^G < _ ^B

R^~ L ^{; G ~} U ^{; B} - L In accordance to a preferable embodiment, the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, further includes a step of mapping each of the sets of tertiary first order chromatic parameters that respectively correspond to a given parameter type of insulation fluid degradation parameter type for a plurality of time instances, in Cartesian space.

In accordance to a preferable embodiment, the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, further includes a step of plotting trend- lines of each secondary first order chromatic parameter of the sets of secondary first order chromatic parameters that respectively correspond to a given parameter type of insulation fluid degradation parameter type for a plurality of time instances. In accordance to a preferable embodiment, the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention includes a step of mapping the set of second order chromatic parameters in Cartesian space. In accordance to a preferable embodiment, the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, empirical data is utilized to determine assignment of scores to said plurality of sets of tertiary and secondary first order chromatic parameters at said plurality of time instances and the rate of change with respect to time of said sets of tertiary and secondary first order chromatic parameters which have been categorized according to parameter type of insulation fluid degradation, is based on the CIGRE (International Council On Large Electric Systems) levels of said insulation fluid degradation parameter types. CIGRE levels of said insulation fluid degradation parameter types are further referenced to provide upper limits of said insulation fluid data that corresponds to healthy components such as electrical windings and tap changers of an electrical power equipment that is immersed in said insulation fluid under test. In accordance to a preferable embodiment, the method for obtaining failure prognostic information of the electrical power equipment based on insulation fluid degradation of the present invention, the plurality of sensor/detectors include gas detectors/sensors for detecting the presence of gasses such as Methane (ChU), Hydrogen (H2), Ethane (C2H6), Acetylene (C2H2), Carbon dioxide (CO2), Carbon monoxide (CO) and Ethylene (C2H4). In accordance to said preferable embodiment, the plurality of sensors/detectors further comprise of a sensor/detector for detecting a level of acidity of an insulation fluid, a sensor/detector for detecting a dielectric strength of insulation fluid, and a sensor/detector for detecting the presence of water in said insulation fluid. In accordance to said preferable embodiment, the plurality of sensors/detectors yet further include a tri-stimulus sensor system comprising of three colour photo detectors for detecting and providing a measure of the colour of an insulation fluid under test in terms of a set of primary first order chromatic parameters. This and other objects and advantages of the invention will become apparent to one of ordinary skill in the art upon reading the following specification and appended claims.

BRI EF DESCRIPTION OF THE DRAWI NGS

The above and other objects, features and other advantages of the present invention will be more clearly understood from the detailed description taken in conjunction with the accompanying drawings, in which: Figure 1 is a diagram illustrating a plurality of discrete signals that correspond to the Dissolved Gas Analysis (DGA) parameter type of an electrical power equipment insulation fluid degradation that are subjected to first order chromaticity processing by a tri-stimulus processing system comprising of three processors that have non-orthogonal Gaussian approximated responses in accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment based on degradation of insulation fluid that immerses components of said electrical power equipment that include electrical windings and tap-changers;

Figure 2 is a chromatic map of the set of tertiary first order chromatic parameters, i.e. ' R'DGA ', ' G'DGA ', ' B'DGA ' obtained by chromaticity processing of the plurality of discrete signals depicted in figure 1 , in which said tertiary first order chromatic parameters ' R'DGA ', ' G'DGA ', ' B'DGA ' are mapped into Cartesian space, in accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment of the present invention;

Figure 3 is a diagram illustrating a plurality of discrete signals that correspond to the parameter type AWD (Acidity, Water Content and Dielectric Strength) of an electrical power equipment insulation fluid degradation that is subjected to first order chromaticity processing by a tri-stimulus processing system comprising of three processors that have non-orthogonal Gaussian approximated responses in accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment of the present invention;

Figure 4 is a chromatic map of the set of tertiary first order chromatic parameters i.e. ' R'AWD ', y = ' G'AWD ', z = ' B'AWD ' obtained by chromaticity processing of the plurality of discrete signals depicted in figure 3, in which said tertiary first order chromatic parameters ' R'AWD ', ' G'AWD ', ' B'AWD ' are mapped into Cartesian space;

Figure 5 is a diagram illustrating a discrete signal that corresponds to the parameter type of insulation fluid colour index (CI) of an electrical power insulation fluid degradation that is subjected to first order chromaticity processing in accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment of the present invention;

Figure 6 is a diagram illustrating the sequence of steps executed in accordance to a preferable embodiment of the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention;

Figure 7 is a diagram illustrating the steps executed in second order chromaticity processing of the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation in accordance to a preferable embodiment of the present invention;

Figure 8 is a flow-chart illustrating the sequence of steps executed by the method for obtaining failure prognostic information of electric power equipment in accordance to a preferable embodiment of the present invention.

Figure 9 is a schematic diagram of an exemplary computer based system that executes the method for obtaining failure prognostic information of electrical power equipment in accordance to a preferable embodiment of the present invention;

Figure 10 is a screen-shot of a trend line plot of the secondary first order chromatic parameter "LDGA" with respect to a plurality of time instances executed by a software application that executes the method for obtaining failure prognostic information of electrical power equipment in accordance to a preferable embodiment of the present invention;

Figure 1 1 is a screen shot of a pair of tables, in which primary and secondary first order chromatic parameters grouped according to parameter types of "Dissolved Gas Analysis (DGA)" and "Acidity, Water Content and Dielectric Strength (AWD) are organized as weighting factors in relation to electrical power equipment insulation fluid degradation as tabulated by a software application that executes the method for obtaining failure prognostic information of electrical power equipment in accordance to a preferable embodiment of the present invention; and

Figure 12 is a screen shot of a trend-line plot of the second order chromatic parameter "L 2nd Order with respect to a plurality of time instances executed by a software application that executes the method for obtaining failure prognostic information of electrical power equipment in accordance to a preferable embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

The detailed description set forth below in connection with the appended drawings is intended as a description of one or more exemplary embodiments and is not intended to represent the only forms in which the embodiments may be constructed and/or utilized. The description sets forth the functions and the sequence for constructing one or more exemplary embodiments. However, it is to be understood that the same or equivalent functions and sequences may be accomplished by different embodiments that are also intended to be encompassed within the scope of this disclosure.

The method 100, 200 of obtaining failure prognostic information of electrical power equipment utilizing insulation fluid degradation parameters in accordance to a preferable embodiment of the present invention, will now be described with reference to figures 1 to 1 1 appended herein.

Before proceeding to a detailed description of the method 100, 200 of obtaining failure prognostic information of electrical power equipment utilizing insulation fluid degradation parameters in accordance to a preferable embodiment of the present invention, an introduction to chromaticity processing is detailed in the subsequent passages.

By way of introduction, chromaticity processing is the name given to the application of sets of non-orthogonal weighted integrals to signals distributed across a measurement range and the subsequent transformation of the integral quantities obtained to give parameters summarising certain characteristic of the distribution. The name chromaticity processing is derived from the method's origin in broadband optics and colour science, in which the technique is applied to light intensity over optical spectrum. However, despite its origins, chromaticity processing can be applied to measurements of any quantity distributed across another variable such as acoustic intensity versus frequency or temperature versus spatial position. The strength of chromaticity processing for processing of continuous or discrete signals to obtain information provided by distribution of such signals across a given measurement range is the provision of parameters, i.e. chromatic parameters, that provide for the convenient visual correlation of such signals to an incidence of a given anomalous and/or desired event in relation to the measured quantity over the given measurement range. Chromaticity processing further provides the advantage of being maximally information preserving as useful information retention is maximized due to the use of signal processor/filters having non-orthogonal responses. Chromaticity processing of signals yet further provides the advantage of useful information retention with high robustness to noise with the use of at least three detectors/sensors that each have Gaussian approximated responses that are non-orthogonal with respect to each other. It should be understood here that the phrase "non-orthogonal responses", refers to the fact that the responses of the signal processors/filters overlap over a given measurement range.

In chromaticity processing as applied in accordance to a preferable embodiment of the method 100, 200 for obtaining failure prognostic information for electrical power equipment that utilize insulation fluid, of the present invention, a plurality of non-orthogonal weighted integrals are applied to a plurality of discrete signals which correspond to a distribution of a measured quantity across a given variable. The result of said application of non- orthogonal weighted integrals to said discrete signals (which correspond to a distribution of a measured quantity across a given variable) is viewed as the application of a plurality of filters/signal processors with non-orthogonal responses to filter/process aforementioned discrete signals.

The abovementioned operation, is essentially the first stage of chromaticity processing. The resulting output of the abovementioned operation, for a given signal processor/ filter with a given non-orthogonal response, is a weighted integral of the processor response multiplied by the amplitude of the discrete signal that is overlaid by said processor response, which for a plurality of processors/filters with non-orthogonal responses results in a plurality of sets of chromatic parameters which are subsequently mapped into chromatic maps which represents a Cartesian space. The use of chromaticity processing further brings the advantage of reducing the complexity of prognostic monitoring of electrical power equipment based on dielectric insulation fluid data, as it reduces the large number of parameters required for prognostic monitoring such as Dissolved Gas Analysis data, acidity, dielectric strength and moisture content data as well as colour information of said dielectric insulation fluid to a manageable number of chromatic parameters which hence eliminates the need for considerable skill and experience required by a technician or engineer in determining when corrective action which may include replacement of components such as windings or tap-changers immersed in a given insulation fluid under test of a given electrical power equipment, is required.

In the second stage of chromaticity processing, one or more sets of primary first order chromatic parameters are converted to secondary first order chromatic parameters by way of a transformation from Cartesian space to a space referenced by a new set of parameters that provides for better operator interpretability of information represented by said one or more sets of primary first order chromatic parameters by partitioning said primary first order chromatic parameters in any one of said one or more sets into components of distinct character.

For simplicity and in view of the method 100, 200 of obtaining failure prognostic information of electrical power equipment in accordance to a preferable embodiment of the present invention, chromaticity processing as practiced by the method of the present invention utilizes at least three non-orthogonal Gaussian approximated responses applied to a given set of discrete signals distributed over a given measurement range that is applied by at least three signal processors/filters which result in at least three primary first order chromatic parameters and consequently, by way of an additional algorithmic transformations at least three secondary first order chromatic parameters and at least three tertiary first order chromatic parameters. In application in colour science in which exemplarily, three detectors/sensors/filters/signal processors (i.e. e.g. colour photo detectors in CCD cameras) which have Gaussian approximated responses that are non- orthogonal to each other over the exemplary measurement range of frequency, the result of chromaticity processing yields three primary first order chromatic parameters, i.e. 'R', 'G', 'B' in Cartesian space and consequently three secondary first order chromatic parameters, i.e. V, Ή', 'S' into a new space which have been found to render significant operator interpretability of said primary first order chromatic parameters through partitioning of said primary first order chromatic parameters into components of distinct character. More particularly, said transformation from Cartesian space of said primary first order chromatic parameters 'R', 'G', 'B' is executed, by way of example only, as follows: f 60(G - B)/( ax(R, G, B) - in(R, G, B)), if ax(R, G, B) = R

H = \ 60(2 + (B - R))/ ax(R, G, B) - min(fl, G, B)) , if ax(R, G, B) = G (1) 60(4 + (R - G))/(max(fi, G, B) - min(fl, G, B)), if max(R, G, B) = B

_{= (i±f±i)}

(2) max(fl, G, B) - min(fl, G, B)

max(R, G, B) + min(fl, G, B)

In the above mentioned transformation, the secondary first order chromatic parameters Ή' alludes to 'Hue' and corresponds to an effective value of a distribution variable about which measurements are spread. In the abovementioned transformation, the secondary chromatic parameter Ή', i.e. 'Hue' is specified as an angle (given in degrees by the above formula). Referring to aforementioned transformation again, the secondary parameters 'L' and 'S' correspond respectively to 'Lightness' and 'Saturation' and further respectively represent the nominal amplitude of the original measurements summed across the range of their distribution variable and an indication of a degree to which the measurements are spread throughout the range of the distribution. We will now proceed with a description of the method 100, 200 of obtaining failure prognostic information of electrical power equipment in accordance to a preferable embodiment of the present invention. More particularly, with reference to figures 6 to 8, in one aspect, the present invention provides a method 100, 200 for obtaining prognostic information of electrical power equipment based on degradation of dielectric insulation fluid that immerses components of said electrical power equipment that include electrical windings and tap-changers comprising of: a first stage 100 in which a plurality of discrete signals are grouped according to parameter type across one or more parameter types of insulation fluid degradation and chromatically processed (i.e. subjected to first order chromaticity processing 104) to obtain a plurality of sets of primary 104a, 104b, 104c, secondary 105a, 105b, 105c and tertiary 106a, 106b, 106c first order chromatic parameters that correspond to each of said one or more parameter types of insulation fluid degradation at a plurality of time instances and in which scores are assigned to each chromatic parameter and a rate of change with respect to time of each chromatic parameter of said plurality of sets of secondary 105a, 105b, 105c and tertiary 106a, 106b, 106c first order chromatic parameters at a plurality of time instances; and a second stage 200 in which the scores of each chromatic parameter and the rate of change of each chromatic parameter with respect to time of said plurality of sets of secondary 105a, 105b, 105c and tertiary 106a, 106b, 106c first order chromatic parameters at a plurality of instances of time are combined across a given parameter type of insulation fluid degradation and said combined score 201a, 201 b, 201 c for each parameter type are collectively subjected to second order chromaticity processing 202 to provide a set of second order chromatic parameters 202a, 202b, 202c at a plurality of time instances, that are utilized to provide an indication of the insulation fluid degradation and the health of components of an electrical power equipment immersed in the insulation fluid being monitored. In accordance to a preferable embodiment of the method 100, 200 of the present invention, the second stage 200 comprises of a step of combining scores of each parameter type of insulation fluid degradation of each chromatic parameter and the rate of change of each chromatic parameter with respect to time of said plurality of sets of secondary 105a, 105b, 105c and tertiary 106a, 106b, 106c first order chromatic parameters at a plurality of instances of time are combined across a given parameter type of insulation fluid degradation and said combined score 201 a, 201 b, 201 c for each parameter type; and a score assigned to a discrete signal corresponding to an insulation fluid degradation parameter type are collectively subjected to second order chromaticity processing 202 to provide a set of second order chromatic parameters 202a, 202b, 202c at a plurality of time instances, that are utilized to provide an indication of the insulation fluid degradation and the health of components of an electrical power equipment immersed in the insulation fluid being monitored.

With reference to figures 2, 4, 10, and 12, it should be understood here, that prognostic information obtained, being information that enables one of ordinary skill in the art of electrical power equipment condition monitoring to determine whether the insulation fluid has degraded to such an extent as to warrant de- commissioning of the electrical power equipment that utilizes said insulation fluid. It should be further understood, that said prognostic information, in accordance to a preferable embodiment of the method of the present invention, is obtained by analysis of chromatic maps 20, 40 that result from the mapping of said plurality of sets of tertiary first order chromatic parameters 106a, 106b, 106c into Cartesian space; analysis of a trend-line plot of a secondary first order chromatic parameter from any one of the plurality of sets of secondary first order chromatic parameters 105a, 105b, 105c, of a given parameter type of insulation fluid degradation over a plurality of time instances; analysis of a trend-line plot of the rate of change with respect to time of a secondary first order chromatic parameter of any one of the plurality of sets of secondary first order chromatic parameters 105a, 105b, 105c, of a given parameter type of insulation fluid degradation over a plurality of time instances; and analysis of a trend-line plot of a second order chromatic parameter from a set of second order chromatic parameters 202a, over a plurality of time instances. Said set of second order chromatic parameters 202a, obtained from chromaticity processing of combined scores of tertiary and/or secondary chromatic parameters as well as time rate of changes of said tertiary and/or secondary chromatic parameters of said plurality of sets of tertiary and/or secondary chromatic parameters 106a, 106b, 106c, 105a, 105b, 105c that correspond respectively to the insulation fluid degradation parameter types of an insulation fluid.

In accordance to a preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the parameter types are Dissolved Gas Analysis (DGA) data, Acidity, Water Content and Dielectric Strength (AWD) data, as well as Colour Index (CI) data of an insulation fluid under test. In accordance to said preferable embodiment of the method 100, 200 of the present invention, said DGA, AWD and Colour Index (CI) data of an insulation fluid under test, can be embodied and be thought of as discrete signals.

More particularly, with reference to said figures 6 to 8, in accordance to a preferable embodiment of the method 100, 200 of the present invention the method 100, 200 comprises the steps of: a first step 101 of obtaining a plurality of discrete signals 10b, 30b, 50b obtained from a plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters that include discrete signals corresponding to "Dissolved Gas Analysis" parameters, acidity of said insulation fluid, dielectric strength of said insulation fluid, moisture content of said insulation fluid as well as colour of said insulation fluid, at a plurality of time instances; a step 102 of grouping the plurality of discrete signals 10b, 30b, 50b obtained from said plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters based on parameter type such as Dissolved Gas Analysis (DGA), AWD (Acidity, Water Content and Dielectric Strength), and insulation fluid colour Index (CI); a step 103 of normalizing the plurality of discrete signals 10b, 30b, 50b obtained from said plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters that correspond to the AWD (Acidity, Water Content and Dielectric Strength) parameter type; a step 104 of applying a plurality of non-orthogonal Gaussian approximated responses 10a, 30a, 50a and computing the result of multiplication of the weighted integral of said plurality of non-orthogonal Gaussian approximated responses 10a, 20a, 30a and said discrete signals 10b, 30b, 50b grouped according to parameter type obtained in the second 102 and third 103 steps for said plurality of time instances to hence provide a plurality of sets of primary first order chromatic parameters 104a, 104b, 104c at said plurality of time instances that correspond to the discrete signals 10b, 30b, 50b grouped based on parameter type of insulation fluid degradation; a step 105 of algorithmically converting the plurality of sets of primary first order chromatic parameters 104a, 104b, 104c at said plurality of time instances that are grouped according to parameters based on parameter type of insulation fluid degradation into corresponding sets of secondary first order chromatic parameters 105a, 105b, 105c that provide better interpretability of information through partitioning said primary first order chromatic parameters into components of distinct character; a step 106 of scaling each individual chromatic parameter of a given set of primary first order chromatic parameters 104a, 104b, 104c that correspond to a given parameter type with a reciprocal of an integer multiple of a secondary first order chromatic parameter of a corresponding set of secondary first order chromatic parameters 105a, 105b, 105c at said plurality of time instances to obtain a tertiary set of first order chromatic parameters 106a, 106b, 106c for each parameter type of insulation fluid degradation parameters at said plurality of time instances; and a step 107 of computing a rate of change with respect to time of each of said sets of tertiary 106a, 106b, 106c and secondary first order chromatic parameters 105a, 105b, 105c obtained for a plurality of time instances and which have been grouped according to parameter type of insulation fluid degradation. In accordance to a preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the number of Gaussian approximated non-orthogonal responses 10a, 30a, 50a utilized is numerically equal to three responses, which consequently result in three primary first order chromatic parameters, 'R' , 'G' , 'B' for each plurality of discrete signals 10b, 30b, 50b that are grouped according to a parameter type of dielectric insulation fluid degradation parameters (i.e. DGA , AWD and CI). Again as may have already been mentioned, in accordance to said preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the three primary first order chromatic parameters of any one of the plurality of sets of primary first order chromatic parameters 104a, 104b, 104c are converted into corresponding three secondary first order chromatic parameters namely 'Hue', 'Lightness' and 'Saturation' denoted as Ή', 'L' and 'S' that provide better interpretability of information through partitioning said primary first order chromatic parameters into components of distinct character. More particularly in accordance to said preferable embodiment, the conversion of said three primary first order chromatic parameters, 'R', 'G', 'B' for each set of primary first order chromatic parameters 104a, 104b, 104c for each plurality of discrete signals that are grouped according to a parameter type of insulation fluid degradation parameters (i.e. DGA, AWD and CI) into three corresponding secondary first order chromatic parameters, is executed utilizing the following transformation formula:

max(fl, G, B) - m R, G, B)

(3) max(R, G, B) + min(fl, G, B) In accordance to said preferable embodiment, the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the step of scaling each individual primary chromatic parameter of a given set of primary first order chromatic parameters 104a, 104b, 104c that correspond to a given parameter type comprises of multiplying each of said individual primary first order chromatic parameters with a reciprocal of "3L", in which "L" represents the "Lightness" secondary first order chromatic parameter of a set secondary first order chromatic parameters 105a, 105b, 105c that correspond to said set of primary first order chromatic parameters 104a, 104b, 104c. More particularly, in accordance to said preferable embodiment, the conversion of said three primary first order chromatic parameters, 'R', 'G', 'B' for each plurality of discrete signals that are grouped according to a parameter type of insulation fluid degradation parameters (i.e. DGA, AWD and CI) into three corresponding tertiary first order chromatic parameters ( i.e. R', G', B') to provide a plurality of sets of tertiary first order chromatic parameters 106a, 106b, 106c is executed utilizing the following transformation formulae:

R G B

R' = 3L ^ = 3L ^ = 3L >

In accordance to a preferable embodiment, the method for obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, further includes a step of mapping each of the sets 106a, 106b, 106c of tertiary first order chromatic parameters that respectively correspond to a given parameter type of insulation fluid degradation parameter type for a plurality of time instances, in Cartesian space 20, 40.

More particularly, with reference to figure 1 , in accordance to a preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment of the present invention, there is conceptually illustrated a process in which a plurality of discrete signals 10b (corresponding to discrete data) which are representative of a plurality of gas species present in an insulation fluid of an electrical power equipment, are subjected to the application of a plurality of non-orthogonal Gaussian approximated responses 10a (i.e. three non-orthogonal responses 10a in accordance to a preferable embodiment), which are responses of signal processor/filters (i.e. three signal processors/filters in accordance to a preferable embodiment) adapted to process/filter discrete signals indicative of the presence of particular gases, in parts per million (ppm) in a given insulation fluid sample of an electrical power equipment, which roughly correspond to steps 101 , 102 and 104 of said method 100, 200 of obtaining failure prognostic information in accordance to said preferable embodiment of the present invention. In accordance to said preferable embodiment, the gasses detected or sensed by said sensors/detectors include Methane (ChU), Hydrogen (H2), Ethane (C2H6), Acetylene (C2H2), Carbon dioxide (CO2) , Carbon monoxide (CO) and Ethylene (C2H4). In essence figure 1 is a diagram attempting to illustrate first order chromaticity processing of a plurality of discrete signals 10b representative of insulation fluid data that have been grouped according to the parameter type of "Dissolved Gas Analysis (DGA)" to consequently produce a set of primary first order chromatic parameters 104a, namely "RDGA, GDGA, BDGA". It should be noted that the figure 1 just represents the plurality of discrete signals which correspond to the parameter type of DGA at a given instance of time, and hence processing steps described with reference to figure 1 , will be repeated at a plurality of instances of time depending on the time duration a given insulation fluid of an electrical power equipment is being monitored to obtain failure prognostic information.

With reference to figure 2, aforementioned sets of tertiary first order chromatic parameters 160a are mapped into three dimensional Cartesian space, in which the dotted line in figure 2, represents a projection of the Y-axis onto the XZ plane. More particularly aforementioned sets of primary first order chromatic parameters 140a corresponding to the parameter type of Dissolved Gas Analysis (DGA) , namely "(RDGA, GDGA, BDGA)" in accordance to said preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment of the present invention, are subsequently transformed into corresponding sets of secondary first order chromatic parameters 150a, namely "(HDGA, LDGA, SDGA)" by way of the transformation of equations (1) to (3). This corresponds to step 105 of the method 100, 200 in accordance to a preferable embodiment of the present invention. Subsequently aforementioned sets of primary first order chromatic parameters 104a which correspond to said parameter type of DGA, are each scaled by multiplying each individual primary first order chromatic parameter of said sets of primary first order chromatic parameters 104a with a reciprocal of "3I_DGA", in which "LDGA" represents the "Lightness" secondary first order chromatic parameter of the sets of secondary chromatic parameters 105a that corresponds to said sets of primary first order chromatic parameters 104a which results in tertiary sets of first order chromatic parameters 106a, namely "(R'DGA, G'DGA, B'DGA)". The following represents the transformation of said sets of primary first order chromatic parameters 104a to corresponding sets of tertiary first order chromatic parameters 106a.

^{K 3} J

The above sequence of events detailed in the preceding paragraph, correspond to steps 105 and 106 in accordance to a preferable embodiment of the method 100, 200 of the present invention.

With reference to figure 2, each of the plotted coordinates points 'x', y, 'z' in three dimensional Cartesian space which represents plots of the sets of tertiary first order chromatic parameters 106a, i.e. (R' DGA, G'DGA, B'DGA) that correspond to the parameter type DGA at a plurality of time instances, are inherently normalized by a factor of 3I_DGA, which results in the following important features of the resulting chromatic map as shown in figure 2 for a given plurality of sets of tertiary chromatic parameters (R' DGA, G'DGA, B'DGA), 106a that correspond to the parameter type DGA at a given plurality of time instances;

• the sum of the 'x-coordinate' +'y-coordinate' + 'z-coordinate' = 1 (since R'DGA + G'DGA + B'DGA = 1 due to scaling with the reciprocal of 3I_DGA;

• the locus of the 'x-coordinate' = 'z-coordinate' for various values of y, is represented by the dotted lines in figure 2 (i.e. the projection of the y- axis on the XZ plane);

• at the origin of the XZ plane of the chromatic map of figure 2, i.e. 'x- coordinate' =0 , 'z-coordinate' =0, the 'y-coordinate' is a maximum, i.e. 'y' = . A plot of said tertiary first order chromatic parameters in different regions of said Cartesian space 20 illustrated in figure 2 correlates with various causes of insulation fluid degradation indicated by the presence of the previously mentioned gasses. Each of the dominant tertiary first order chromatic parameters, R'DGA, G'DGA and B'DGA of a given set of tertiary first order chromatic parameters 106a corresponding to the insulation fluid degradation parameter type DGA at given instance of time is defined to be a dominant contributor/cause to insulation fluid degradation when either the x-coordinate = R'DGA > 0.5, the y- coordinate = G' DGA ^> 0.5 or the z-coordinate B'DGA ^> 0.5. Hence, it is observed that analysis of the DGA parameter type which comprises of seven (7) parameters in accordance to a preferable embodiment (i.e. the presence of Methane (ChU), Hydrogen (H2), Ethane (C2H6), Acetylene (C2H2), Carbon dioxide (CO2), Carbon monoxide (CO) and Ethylene (C2H4) in parts per million) is simplified by chromaticity processing in which 7 parameters are reduced to a set 104a of three parameters, i.e. the three first order primary chromatic parameters "RDGA, GDGA, BDGA" which are mapped into Cartesian space to facilitate operator interpretability. The various regions correlated to various causes of insulation fluid degradation in relation to the DGA parameter type are identified with the aid of empirical data.

With reference to figure 3, in accordance to a preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment of the present invention, there is conceptually illustrated a process in which a plurality of discrete signals 30b (corresponding to discrete data) which are representative of the data categorized under the insulation fluid degradation parameter type of acidity of insulation fluid, water content of insulation fluid and dielectric strength of insulation fluid abbreviated as AWD, are first normalized and then subjected to the application of a plurality of non-orthogonal Gaussian approximated responses 30a (i.e. three non-orthogonal responses 30a in accordance to a preferable embodiment), which are responses of signal processors/filters (i.e. three signal processors/filters in accordance to a preferable embodiment) adapted to process/filter discrete signals respectively corresponding to the acidity of the insulation fluid, the water content of the insulation fluid and the dielectric strength of the insulation fluid. In essence figure 3 is a diagram attempting to illustrate first order chromaticity processing of a plurality of discrete signals 30b representative of insulation fluid data that have been grouped according to the parameter type "Acidity, Water Content and Dielectric Strength (AWD)" to consequently produce a set of primary first order chromatic parameters 104b, namely "RAWD, GAWD, BAWD".

The above sequence of events detailed in the preceding paragraph, similar to the case depicted in figure 1 , corresponds to steps 101 , 102, 103 and 104 of the method for obtaining failure prognostic information of electrical power equipment in accordance to a preferable embodiment of the present invention. With reference to figure 4, each of the plotted coordinate points 'x', y, '^ζ' in three dimensional Cartesian space 40, represent plots of the sets of tertiary first order chromatic parameters 106b, i.e. (R AWD, G'AWD, B'AWD) that correspond to the parameter type AWD at a plurality of time instances, in which the dotted line in figure 4 represents a projection of the Y-axis onto the XZ plane.

More particularly, aforementioned sets of primary first order chromatic parameters 140b, namely " RAWD, GAWD, BAWD" are subsequently algorithmically converted to sets of secondary first order chromatic parameters 105b namely "(HAWD, LAWD, SAWD)" that correspond to said parameter type AWD of insulation fluid degradation by way of the transformation of equations (1 ) to (3). Again, this corresponds to step 105 of the method 100, 200 of obtaining failure prognostic information in accordance to a preferable embodiment of the present invention. Subsequently, said sets of primary first order chromatic parameters 104b at a plurality of time instances that correspond to the insulation fluid degradation parameter type of AWD are each scaled by multiplying each individual primary first order chromatic parameter of said sets of primary first order chromatic parameters 104b with a reciprocal of "3I_AWD", in which " LAWD" represents the "Lightness" secondary first order chromatic parameter of the sets of secondary first order chromatic parameters 105b that correspond to said sets of primary first order chromatic parameters 104b which results in sets of tertiary first order chromatic parameters 106b, namely "(R' DGA, G'DGA, B'DGA)". The following represents the transformation of each primary first order chromatic parameter of said sets of primary first order chromatic parameters 104b to a corresponding tertiary first order chromatic parameter of corresponding sets of tertiary first order chromatic parameters 106b. p, _ RAWD _R , _ GAWD _r, _ BAWD

" AWD — · ^' AWD — ' " AWD ~

Jtiwn

Again, the above sequence of events detailed in the preceding paragraph, correspond to steps 105 and 106 in accordance to a preferable embodiment of the method 100, 200 of the present invention.

In essence, figure 4 represents plots of the sets of tertiary first order chromatic parameters 106a, i.e. (R AWD, G'AWD, B'AWD) that correspond to the parameter type AWD at a plurality of time instances, and which are inherently normalized by a factor of 3I_AWD, which results in the following important features of the resulting chromatic map as shown in figure 4 for a given plurality of sets of tertiary chromatic parameters (R' AWD, G'AWD, B'AWD), 106a that correspond to the parameter type AWD at a given plurality of time instances;

• the sum of the 'x-coordinate' +'y-coordinate' + 'z-coordinate' = 1 (since R'AWD + G'AWD + B'AWD = 1 due to scaling with the reciprocal of 3I_DGA;

• the locus of the 'x-coordinate' = 'z-coordinate' for various values of 'y', is represented by the dotted lines in figure 4 (i.e. the projection of the y- axis on the XZ plane);

• at the origin of the XZ plane of the chromatic map of figure 4, i.e. 'x- coordinate' =0 , 'z-coordinate' =0; the 'y-coordinate is a maximum, i.e. y = . With reference to figure 4 yet again, a plot of said tertiary first order chromatic parameters in different regions of said Cartesian space 40 illustrated in figure 4 correlates with various causes of insulation fluid degradation indicated by the presence of water/moisture content, reduction in dielectric strength of said insulation fluid being monitored and an increase in acidity of the insulation fluid being monitored. Each of the dominant tertiary first order chromatic parameters, R'AWD, G'AWD and B'AWD of a given set of tertiary first order chromatic parameters 106b corresponding to the insulation fluid degradation parameter type AWD at a given instance of time is defined to be a dominant contributor/cause to insulation fluid degradation when either the x-coordinate = R'AWD ^> 0.5, the y-coordinate— G'AWD > 0.5 or the z-coordinate = B'AWD > 0.5.

Generally, with reference to aforementioned figure 4, considering a preferable embodiment of the method 100, 200 of the present invention, in which the 'x- coordinate' = R'AWD, the 'y-coordinate' G'AWD and the 'z-coordinate' B'AWD, and with reference to figure 3, the tertiary first order chromatic parameter G'AWD, corresponds to a discrete signal representing Water/Moisture content (labelled W) , the tertiary first order chromatic parameter B' AWD corresponds to a discrete signal representing Acidity (labelled A) and the tertiary first order chromatic parameter R'AWD corresponds a discrete signal representing dielectric strength reduction (labelled D), then a plot of said tertiary first order chromatic parameters inclined towards the 'z'=0 and 'x'= 0 point of said Cartesian space 40 illustrated in figure 4, correlates to strong dominance of reduction in dielectric strength as a contributing factor to insulation fluid degradation..

With reference to figure 5, in accordance to a preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment of the present invention, there is conceptually illustrated a process in which one or more discrete signals 50b (corresponding to discrete data) which are representative of the data categorized under the insulation fluid degradation parameter type of insulation fluid colour index abbreviated as "CI" subjected to the application of a plurality of non-orthogonal Gaussian approximated responses 50a (i.e. three non-orthogonal responses 50a in accordance to a preferable embodiment). In essence figure 5 is a diagram attempting to illustrate first order chromaticity processing of one or more discrete signals 50b representative of insulation fluid data that have been grouped according to the parameter type "Colour Index (CI)" to consequently produce a set of primary first order chromatic parameters 104c, namely "RCI. GCI, Be ". Similar to the sets 104a, 104b of primary first order chromatic parameters corresponding to the insulation fluid degradation parameter types DGA and AWD, said set 104c of primary first order chromatic parameters, namely "Rci, Gci, Be " can be algorithmically converted to a corresponding set of tertiary first order chromatic parameters "R'ci, G'ci, B'ci" and subsequently plotted for a plurality of time instances in Cartesian space to provide an operator with knowledge of how insulation fluid colour changes over time. Said change in insulation fluid colour may be indicative of a severity of insulation fluid degradation.

With reference to figures 6 and 7, in accordance to a preferable embodiment, of the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the method 100, 200 further includes a step 201 of assigning scores to said plurality of sets of tertiary 106a, 106b, 106c and secondary 105a, 105b, 105c first order chromatic parameters at said plurality of time instances and the rate of change with respect to time of said sets of tertiary 106a, 106b, 106c and secondary 105a, 105b, 105c first order chromatic parameters, said sets of tertiary 106a, 106b, 106c and secondary 105a, 105b, 105c first order chromatic parameters and corresponding rate of change with respect to time of said sets of tertiary 106a, 106b, 106c and secondary 105a, 105b, 105c first order chromatic parameters categorized according to parameter type of insulation fluid degradation and are assigned scores based on empirical data of parameters categorized according to said parameter type of insulation fluid degradation.

In accordance to said preferable embodiment of the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, the method 100, 200 further includes a step 202 of combining scores assigned to said plurality of sets of tertiary 106a, 106b, 106c and secondary 105a, 105b, 105c first order chromatic parameters as well as the rate of change with respect to time of said plurality of sets of tertiary 106a, 106b, 106c and secondary 105c, 105b, 105c first order chromatic parameters at a plurality of time instances that fall under a given group of insulation fluid parameter type (DGA, AWD, CI) to obtain a combined score of insulation fluid degradation for each parameter type and computing the result of multiplication of the weighted integral of a plurality of non-orthogonal Gaussian approximated responses and said combined score for each parameter type to provide a set of second order chromatic parameters 202a, 202b, 202c that in turn provide an overall indication of insulation fluid degradation and hence health of components such as electrical windings and tap changers which are immersed in the insulation fluid under test, of a given electrical power equipment. In accordance to a preferable embodiment, the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention includes a step of mapping the set of second order chromatic parameters 202a in Cartesian space.

The advantage of providing for second order chromaticity processing (i.e. computing the result of multiplication of the weighted integral of a plurality of non-orthogonal Gaussian approximated responses and said combined score for each parameter type to provide a set of second order chromatic parameters 202a, 202b, 202c), is that an overall visualization of the progression of the insulation fluid degradation at a plurality of time instances in which data across all parameter types (i.e. in accordance to a preferable embodiment, DGA, AWD and CI) can be tracked and monitored easily by plotting said second order chromatic parameters into Cartesian space.

With reference to figure 1 1 , in accordance to a preferable embodiment of the method 100, 200 of the present invention, the sets of tertiary 106a, 106b, 106c and/or secondary 105a, 105b, 105c first order chromatic parameters as well as rate of change with respect to time of said sets of tertiary 106a, 106b, 106c and/or secondary 105a, 105b, 105c chromatic parameters corresponding to a given parameter type of insulation fluid degradation, are grouped for said given parameter type to form weighting factors pertaining to said given parameter type of insulation fluid degradation , in which individual scores that have been assigned to said sets of tertiary 106a, 106b, 106c and/or secondary 105a, 105b, 105c chromatic parameters as well as rate of change with respect to time of said sets of tertiary 106a, 106b, 106c and/or secondary 105a, 105b, 105c chromatic parameters are summed across weighting factors to provide a total score indicative of insulation fluid degradation attributed by a given parameter type of insulation fluid degradation.

In accordance to a preferable embodiment, the method 100, 200 of obtaining failure prognostic information of electrical power equipment based on insulation fluid degradation of the present invention, empirical data is utilized to determine assignment of scores to said plurality of sets of tertiary 105a, 105b, 105c and secondary 105a, 105b, 105c first order chromatic parameters at said plurality of time instances and the rate of change with respect to time of said sets of tertiary 106a, 106b, 106c and secondary 105a, 105b, 105c first order chromatic parameters which have been categorized according to parameter type of insulation fluid degradation. Said assignment of scores is more particularly based on the CIGRE (International Council On Large Electric Systems) levels of said insulation fluid degradation parameter types. CIGRE levels of said insulation fluid degradation parameter types are further referenced to provide upper limits of said insulation fluid data that corresponds to healthy components such as electrical windings and tap changers of an electrical power equipment that is immersed in said insulation fluid under test.

Again, as may have been mentioned in a preceding paragraph of this detailed description, in accordance to a preferable embodiment of the method 100, 200 of obtaining prognostic information of electrical power equipment of the present invention, said method 100, 200 is embodied as a software application residing in a server and is accessed by an operator/technician/engineer from a remote computer terminal with the computer terminal accessing the software application residing in a server. The software application being in communication with a plurality of sets of sensors/detectors configured to obtain real-time discrete signals corresponding to a given parameter type of insulation fluid degradation of a given electrical power equipment via an interface card and a wireless communication link between said plurality of sets of sensors/detectors and the server. The software application residing in the server being configured to sample discrete signals from said plurality of sets of sensors/detectors at a plurality of instances of time, and process the sampled discrete signals and provide chromatic maps and/or trend-line plots of tertiary and/or secondary first order chromatic parameters and/or rate of change with respect to time of tertiary and/or secondary first order chromatic parameters organized according to parameter type of insulation fluid degradation, at a plurality of time instances to enable an operator to easily interpret said plots to determine a failure prognosis of a given electrical power equipment based on the insulation fluid being monitored. With reference to figure 9, in accordance to yet another preferable embodiment, the method 100, 200 of obtaining prognostic information of electrical power equipment of the present invention, can be embodied as a software application residing in a server 503 and is accessed by an operator/technician/engineer from a remote computer terminal 502 with the computer terminal accessing the software application residing in the server 503 through a wired network comprising of IEEE 802.3 links and a hub. The server 503 including a database 504 containing data corresponding to insulation fluid degradation that are organized according to parameter type of insulation fluid degradation such as DGA, AWD and CI data. The software application residing in the server 503 executes the method 100, 200 in accordance to a preferable embodiments of the present invention detailed in this detailed description, in which said data in the data base can be thought of as discrete signals that have been grouped according to parameter type of insulation fluid degradation and the sensor/detector responses are simulated by the software application to consequently effect first order and second order chromaticity processing.

Discussion of Experimental Results: A discussion of results of prognosis of an electrical power equipment having components such as electrical windings and tap-changers immersed in insulation fluid, based on the method in accordance with a preferable embodiment of the present invention in which only DGA and AWD parameter types are considered and the conventional method of Duval analysis based on the results of Dissolved Gas Analysis (DGA) will now be detailed with reference to tables 1 to 3, below:

Table. 1

Diagnostic outcomes in terms of true negatives and false positive cases

Healthy

Item No. False True

cases Uncertainties positives negatives

submitted

i.) Duval, DGA 28 2

8

ii.) Method of the present

invention utilizing DGA 38

and AWD parameter 1 36 1 types of insulation fluid

data Table. 2

Prognostic outcomes in terms of true positives and false negative cases

*Please note that the Duval DGA results are based on a 3-year monitoring period.

Table. 3

Note: Sensitivity = (TP)/ (TP + FN);

Specificity = (TN)/ (TN + FP)

PPV Positive Predictive Value = (TP)/ (TP + FP)

NPV -> Negative Predictive Value = (TN)/(TN + FN)

Where TP, FP True, False Positive results;

TN, FN-> True, False Negative results

It is observed from the tables 1 to 3, that diagnosis and prognosis of electrical power equipment utilizing the method 100, 200 in accordance to a preferable embodiment of the present invention is more accurate and precise as compared to diagnosis and prognosis of electrical power equipment utilizing Duval DGA analysis. Moreover with reference to figure 3, the diagnosis and prognosis of electrical power equipment utilizing the method 100, 200 in accordance to a preferable embodiment of the present invention, provides a higher sensitivity of (86.7% vs 26%) and a higher specificity (97.3% vs 77.8%) as compared to the Duval DGA analysis method of obtaining prognostic and diagnostic information. The invention has been described with reference to various embodiments. Obviously, modifications and alterations will occur to others upon the reading and understanding of the specification. It is intended to include all such modifications and alterations insofar as they come within the scope of the appended claims or equivalents thereof.

Claims

1. A method (100, 200) for obtaining failure prognostic information of electrical power equipment based on degradation of dielectric insulation fluid that immerses components of said electrical power equipment that include electrical windings and tap-changers comprising of: a first stage (100) in which a plurality of discrete signals (10b, 30b, 50b) are grouped according to parameter type across one or more parameter types of insulation fluid degradation and chromatically processed to obtain a plurality of sets of primary (104a, 104b, 104c), secondary (105a, 105b, 105c) and tertiary (106a, 106b, 106c) first order chromatic parameters that correspond to each of said one or more parameter types of insulation fluid degradation at a plurality of time instances and in which scores are assigned to each chromatic parameter and a rate of change with respect to time of each chromatic parameter of said plurality of sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters at a plurality of time instances; and a second stage (200) in which the scores of at least one chromatic parameter and the rate of change of at least one chromatic parameter with respect to time of said plurality of sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters at a plurality of time instances are combined across a given parameter type of insulation fluid degradation and said combined score (201 a, 201 b, 201c) for each parameter type are collectively subjected to second order chromaticity processing (202) to provide a set of second order chromatic parameters (202a) at a plurality of time instances, that are utilized to provide an overall indication of the insulation fluid degradation and the health of components of an electrical power equipment immersed in said insulation fluid.

2. A method (100, 200) according to claim 1 wherein, the first stage (100) comprises of the following steps: a first step (101) of obtaining a plurality of discrete signals (10b, 30b, 50b) obtained from a plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters that include discrete signals corresponding to "Dissolved Gas Analysis" parameters, acidity of said insulation fluid, dielectric strength of said insulation fluid, moisture content of said insulation fluid as well as colour of said insulation fluid, at a plurality of time instances; a step (102) of grouping the plurality of discrete signals (10b, 30b, 50b) obtained from said plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters based on parameter types of Dissolved Gas Analysis (DGA), A WD (Acidity, Water Content and Dielectric Strength), and insulation fluid colour index (CI); a step (103) of normalizing the plurality of discrete signals (10b, 30b, 50b) obtained from said plurality of detectors/sensors that correspond to electrical power equipment insulation fluid degradation parameters that correspond to the AWD (Acidity, Water Content and Dielectric Strength) parameter type; a step (104) of applying a plurality of non-orthogonal Gaussian approximated responses (10a, 30a, 50a) and computing the result of multiplication of the weighted integral of said plurality of non-orthogonal Gaussian approximated responses (10a, 20a, 30a) and said discrete signals (10b, 30b, 50b) grouped according to parameter type obtained in the second (102) and third (103) steps for said plurality of time instances to hence provide a plurality of sets of primary first order chromatic parameters (104a, 104b, 104c) at said plurality of time instances that correspond to the discrete signals (10b, 30b, 50b) grouped based on parameter type of insulation fluid degradation; a step (105) of algorithmically converting the plurality of sets of primary first order chromatic parameters (104a, 104b, 104c) at said plurality of time instances that are grouped according to parameters based on parameter type of insulation fluid degradation into corresponding sets of secondary first order chromatic parameters (105a, 105b, 105c) that provide better interpretability of information through partitioning said primary first order chromatic parameters into components of distinct character; a step (106) of scaling each individual chromatic parameter of a given set of primary first order chromatic parameters (104a, 104b, 104c) that correspond to a given parameter type with a reciprocal of an integer multiple of a secondary first order chromatic parameter of a corresponding set of secondary first order chromatic parameters (105a, 105b, 105c) at said plurality of time instances to obtain a tertiary set of first order chromatic parameters (106a, 106b, 106c) for each parameter type of insulation fluid degradation parameters at said plurality of time instances; and a step (107) of computing a rate of change with respect to time of each of said sets of tertiary (106a, 106b, 106c) and secondary first order chromatic parameters (105a, 105b, 105c) obtained for a plurality of time instances and which have been grouped according to parameter type of insulation fluid degradation.

3. A method (100, 200) according to claim 2 wherein, the first stage (100) further includes a step of assigning scores based on empirical data of parameters categorized according to parameter type of insulation fluid degradation to said plurality of sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters at said plurality of time instances and the rate of change with respect to time of said sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters, said sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters and corresponding rate of change with respect to time of said sets of tertiary (106a, 106b, 106b) and secondary (105a, 105b, 105c) first order chromatic parameters categorized according to parameter type of insulation fluid degradation .

4. A method (100, 200) according to claim 2 wherein, the second stage (200) comprises: a step of combining scores assigned to said plurality of sets (106a, 106b, 106c) of tertiary and secondary (105a, 105b, 105c) first order chromatic parameters as well as the rate of change with respect to time of said plurality of sets of tertiary (106a, 106b, 106c) and secondary (105a, 105b, 105c) first order chromatic parameters at a plurality of time instances that fall under a given group of insulation fluid parameter type (DGA, AWD, CI) to obtain a combined score of insulation fluid degradation for each parameter type at each instance of time; and a step of computing the result of multiplication of the weighted integral of a plurality of non-orthogonal Gaussian approximated responses and said combined score for each parameter type to provide a set of second order chromatic parameters (202a) that in turn provides an overall indication of insulation fluid degradation and hence health of components such as electrical windings and tap changers which are immersed in the insulation fluid of a given electrical power equipment.

5. A method (100, 200) according to claim 1 or 2 wherein, the number of Gaussian approximated non-orthogonal responses (10a, 30a, 50a) utilized is numerically equal to three responses (10a, 30a, 50a), which consequently lead to the production of a set of three primary first order chromatic parameters, 'R', 'G', 'B' (104a, 104b, 104c) for each plurality of discrete signals (10b, 30b, 50b) that are grouped according to a parameter type of dielectric insulation fluid degradation parameters.

6. A method (100, 200) according to claim 5 wherein, the set three primary first order chromatic parameters (104a, 104b, 104c) are converted into a set of three secondary first order chromatic parameters of 'Hue (H)', 'Lightness (L)' and 'Saturation (S)' (105a, 105b, 105c) in cylindrical space.

7. A method (100, 200) according to claim 5 or 6, wherein each primary first order chromatic parameter of a set of primary first order chromatic parameters (104a, 104b, 104c) that correspond to a given parameter type at a given instance of time is scaled with the reciprocal of an integer multiple of a secondary chromatic parameter of "Lightness (L)" of a corresponding set of secondary chromatic parameters (105a, 105b, 105c).

8. A method (100, 200) according to claim 1 or 2, wherein the first stage (100) further includes a step of mapping each of the sets of tertiary chromatic parameters (106a, 106b, 106c) that respectively correspond to a given parameter type of insulation fluid degradation parameter type for a plurality of time instances, in Cartesian space (20, 40).

9. A method (100, 200) according to claim 1 or 2, wherein the first stage (100) further includes a step of plotting trend-lines of each secondary first order chromatic parameter of the sets of secondary first order chromatic parameters (105a, 105b, 105c) that respectively correspond to a given parameter type of insulation fluid degradation parameter type for a plurality of time instances.

10. A method (100, 200) according to claim 1 or 4 wherein the second stage (200) further includes a step of mapping the set (201 a) of second order chromatic parameters in Cartesian space.