CN105158347A - Variation trend-based oil chromatography comprehensive analysis method - Google Patents

Abstract

The invention belongs to the technical field of oil charge electrical equipment fault diagnosis and relates to a variation trend-based oil chromatography comprehensive analysis method. The method is mainly used in oil chromatography analysis utilizing electrical equipment adopting oilpaper as a main insulating material. The method utilizes a transformer, an electric reactor, a mutual inductor and a cannula. The method analyzes oil chromatography variation trend and combines a modified three-ratio method and a Duval triangle method to realize oil chromatography comprehensive analysis. The method has the advantages of accurate acquisition of oil chromatography variation trend and no code absence.

Description

A kind of oil chromatography comprehensive analysis method based on variation tendency

Technical field

The invention belongs to oil-filled electric equipment fault diagnosis technology field, spy relates to a kind of oil chromatography comprehensive analysis method based on variation tendency, be particularly applicable in oilpaper in the oil chromatogram analysis process of the electrical equipment being predominating insulation, comprise transformer, reactor, mutual inductor and sleeve pipe etc.

Background technology

Oil chromatogram analysis technology is considered to carry out fault diagnosis most convenient, effective method to oil-filled electric equipment, in recent years, Chinese scholars proposes a lot of oil chromatogram analysis method, as: improve three ratios, David's trigonometry, stereographic map method, German three-ratio method, improvement Rogers's method and electricity association and grind method etc.Improvement three-ratio method is applied to the most extensive a kind of effective method at present, it is when breaking down according to transformer, five kinds of gas (H2, CH4, C2H6, C2H4, C2H2) component contents extracted from transformer oil calculate corresponding ratio: C2H2/C2H4, CH4/H2, C2H4/C2H6, obtain a group coding by coding rule, then find out fault type corresponding thereto by three ratio fault diagnosis tables.The method is simple and practical, and accuracy rate is higher, but this diagnostic method exists following problem in actual applications: 1) when characteristic gas content rising tendency in oil exceeds standard, when single air content does not reach demand value, cannot diagnose in this way.2) there is coding entirely in the method, the problem that encoded boundary is too absolute.When the group coding obtained exceeds known coded table, then cannot find fault type, there is the phenomenon of " short in size "; 3) when various faults occurs simultaneously, improveing three ratio diagnosises cannot diagnose.

At present, the variation tendency of oil chromatography is all calculated by absolute factor of created gase or relative factor of created gase, as Chinese invention patent " a kind of transformer device of transformer substation on-line monitoring system " (application number: application number: 201310572278.1), the present invention, by the linear function of oil chromatography in least square fitting nearest a period of time, utilizes slope to judge the variation tendency of oil chromatography.

Summary of the invention

For improvement three-ratio method Problems existing, the present invention analyzes the variation tendency of oil chromatography, merges the method such as improvement three ratio and David's trigonometry, comprehensively analyzes oil chromatography.The present invention has accurately reflection oil chromatography variation tendency and the advantage without " short in size ".

Technical scheme of the present invention is:

Based on an oil chromatography comprehensive analysis method for variation tendency, it is characterized in that, comprise the following steps:

Step 1: the variation tendency being calculated oil chromatography by least square method.Check GB/T7252-2001 directive/guide, obtain the demand value of characteristic gas content in equipment oil.

Step 2: judge that in oil, whether characteristic gas content is greater than demand value, if be less than demand value, then goes to step 3; Otherwise, go to step 4.

Step 3: judge that in oil, whether characteristic gas increases obviously, if so, then goes to step 4, otherwise, go to step 5.

Step 4: whether judge three ratios " short in size ", if not, then adopts improvement three-ratio method to carry out oil chromatogram analysis; Otherwise, adopt David's trigonometry to analyze.

Step 5: the net result obtaining oil chromatogram analysis, investigation failure cause, carries out breakdown maintenance.

Oil chromatography comprehensive analysis method as above, is characterized in that, in described step 1, assuming that acquisition time sequence is { t in during certain section ₁, t ₂..., t _n, the oil chromatography sequence gathered is { y ₁, y ₂, y _n, then the linear regression function of matching is: y=a+bt.Utilize least square method calculating parameter, then wherein oil chromatography sequence is preferably the collection value of nearest a week, or gets the sampled value of nearest 10 times.The present invention defines oil chromatography variation tendency and increases obvious condition and be: | b| > 1.

Oil chromatography comprehensive analysis method as above, is characterized in that, in described step 4, because David's trigonometry includes the possible situation of all three ratios, when the situation of " short in size " appears in improvement three ratio, carries out oil chromatogram analysis by David's trigonometry.

Accompanying drawing explanation

Fig. 1 is based on David's triangular plot of trend;

Fig. 2 is based on the oil chromatography comprehensive analysis method process flow diagram of variation tendency.

Embodiment

Below in conjunction with the drawings and specific embodiments, the invention will be further described.

Improvement three-ratio method is on the basis of the three-ratio method of IEC60599 recommendation, has carried out refinement according to domestic practical experience to coded combination and fault type.The method utilizes five kinds of gas (CH ₄, C ₂h ₄, C ₂h ₆, C ₂h ₂, H ₂) three correlative value (C ₂h ₂/ C ₂h ₄cH ₄/ H ₂c ₂h ₄/ C ₂h ₆) the incompatible method of carrying out fault type judgement of code set, generally use after characteristic gas content exceedes demand value, its coding rule and fault type determination methods are in table 1 and table 2.

Table 1 coding rule

Table 2 fault type determination methods

GB/T7252-2001 directive/guide regulation is when device interior oil dissolved gas content or factor of created gase are greater than demand value, and adopt improvement three-ratio method to carry out oil chromatogram analysis, the demand value of Oil Dissolved Gases Concentration and factor of created gase is respectively in table 3 and table 4.

The demand value unit of table 3 Oil Dissolved Gases Concentration is μ L/L

Table 4 transformer and reactor absolute gas production rate demand value unit are mL/d

Gas composition	Enclosure-type	Open
			Total hydrocarbon	12	6
Acetylene	0.2	0.1
			Hydrogen	10	5
Carbon monoxide	100	50
			Carbon dioxide	200	100

In practical application, the total oil mass of equipment and oil density are difficult to obtain, so definitely factor of created gase is difficult to calculate.Relative factor of created gase is calculated by the measured value of nearest twice, and due to the reason such as equipment error and measuring error, relative factor of created gase exists very large deviation.The present invention utilizes the linear function of oil chromatography in least square fitting nearest a period of time, utilizes the value of slope to judge the variation tendency of oil chromatography.When variation tendency increases obvious, start oil chromatogram analysis.

Least square method (also known as least square method) is a kind of mathematical optimization techniques.It finds the optimal function coupling of data by the quadratic sum of minimum error.Utilize least square method can try to achieve unknown data easily, and between the data that these are tried to achieve and real data, the quadratic sum of error is minimum.Assuming that acquisition time sequence is { t in during certain section ₁, t ₂..., t _n, the desired value sequence gathered is { y ₁, y ₂, y _n, then linear regression function is:

y＝a+bt(1)

Utilize least square method calculating parameter a and b:

$\left\{\begin{array}{c}a=\stackrel{-}{y}-b\stackrel{-}{t}\\ b=\frac{{\mathrm{\Σ}}_{i=1}^n{t}_i{y}_i-n\stackrel{-}{\mathrm{ty}}}{{\mathrm{\Σ}}_{i=1}^n{t}_i^2-n\stackrel{-}{t}}\end{array}\right.---\left(2\right)$

Wherein parameter b can reflect the variation tendency of oil chromatography, and the present invention defines oil chromatography variation tendency and increases obvious condition and be:

|b|＞1(3)

David's triangular plot adding variation tendency as shown in Figure 1, can clear variation tendency of observing oil chromatography visually by Fig. 1 user, and when in data point set, variation tendency is steady; When data point is disperseed, variation tendency is obvious.When data point shifts in zones of different, user according to the trend of change, should take appropriate measures.

The trigon region limit value of David is in table 5.

Table 5 region limit

As shown in Table 5, David's trigonometry includes the possible situation of all three ratios, when the situation of " short in size " appears in improvement three ratio, carries out oil chromatogram analysis by David's trigonometry.Based on the oil chromatography comprehensive analysis method of variation tendency process flow diagram as shown in Figure 2, detailed step is as follows:

Step 1: calculated the oil chromatography variation tendency in nearest a period of time by least square method.Check GB/T7252-2001 directive/guide, obtain the demand value of characteristic gas content in equipment oil.Oil chromatography sequence is preferably the collection value of nearest a week, or gets the sampled value of nearest 10 times.Step 2: judge that in oil, whether characteristic gas content is greater than demand value, if be less than demand value, then goes to step 3; Otherwise, go to step 4.Define oil chromatography variation tendency in the present embodiment to increase obvious condition and be: .

Step 3: judge that in oil, whether characteristic gas increases obviously, if so, then goes to step 4, otherwise, go to step 5.

Step 4: whether judge three ratios " short in size ", if not, then adopts improvement three-ratio method to carry out oil chromatogram analysis; Otherwise, adopt David's trigonometry to analyze.

Step 5: the net result obtaining oil chromatogram analysis, according to the patrolling and examining of equipment, the status information such as on-line monitoring, live detection and power failure test, reference table 6 is further to diagnosing malfunction.

Table 6 fault diagnosis contents table

Claims (5)

1., based on an oil chromatography comprehensive analysis method for variation tendency, it is characterized in that, comprise the following steps:

Step 1: the variation tendency being calculated oil chromatography by least square method;

Step 2: judge that in oil, whether characteristic gas content is greater than demand value, if be less than demand value, then goes to step 3; Otherwise, go to step 4;

Step 3: judge that in oil, whether characteristic gas increases obviously, if so, then goes to step 4, otherwise, go to step 5;

Step 4: whether judge three ratios " short in size ", if not, then adopts improvement three-ratio method to carry out oil chromatogram analysis; Otherwise, adopt David's trigonometry to analyze;

Step 5: the net result obtaining oil chromatogram analysis, investigation failure cause, carries out breakdown maintenance.

2. oil chromatography comprehensive analysis method as claimed in claim 1, is characterized in that, in described step 1, assuming that acquisition time sequence is { t in during certain section ₁, t ₂..., t _n, the oil chromatography sequence gathered is { y ₁, y ₂, y _n, then the linear regression function of matching is: y=a+bt, utilizes least square method calculating parameter, then wherein

3. oil chromatography comprehensive analysis method as claimed in claim 1 or 2, it is characterized in that, described oil chromatography sequence is the collection value of nearest a week, or gets the sampled value of nearest 10 times.

4. oil chromatography comprehensive analysis method as claimed in claim 2, it is characterized in that, described oil chromatography variation tendency increases obvious condition and is: | b| > 1.

5. oil chromatography comprehensive analysis method as claimed in claim 1, it is characterized in that, in described step 4, because David's trigonometry includes the possible situation of all three ratios, when the situation of " short in size " appears in improvement three ratio, carry out oil chromatogram analysis by David's trigonometry.