Method for generating oil property estimation model, method for diagnosing oil property, and oil property diagnosis system

Abstract

To provide a technique of optically determining the amount of a compound which does not become discolored by oxidation degradation.SOLUTION: A method for diagnosing an oil property according to a first aspect of the present invention includes the steps of: acquiring an absorption spectrum in the range of 800 nm-3000 nm wavelength of oil including an additive in which the maximum value of a molar absorption coefficient in the range of 250 nm-800 nm ultraviolet visible wavelength is 50 at a maximum by using a light source and an optical sensor having a detector of light radiated from the light source; and inputting data based on the absorption spectrum into an estimation model equipped in an information processor and predicting the concentration of the additive.SELECTED DRAWING: Figure 19

Description

Technical Field

[0001] The present invention relates to oil diagnostic techniques. In particular, regarding the maintenance of large machinery using industrial oils such as lubricating oil, insulating oil, and processing oil, it measures the compositional changes associated with the use of oils such as lubricating oil, and performs remaining life diagnosis of the oil and prognostic diagnosis of the machine, thereby relating to a technology suitable for monitoring the machine.

Background Art

[0002] In performing the maintenance and repair of large rotating machinery, the diagnosis of the properties of lubricating oil used in rotating parts such as bearings and gears is an important technology. Examples of large rotating machinery include, for example, speed increasers of wind turbines, air compressors, ships, power generation turbines, construction machinery, agricultural machinery, cutting machines, pumps, and speed reducers of railway vehicles.

[0003] In addition to lubricating oil, insulating oil for electrical insulation is used in transformers, etc., and the diagnosis of the properties of insulating oil is also important. Also, processing oil, etc. is used during machining. There are various processing oils suitable for different applications, such as cutting processing oil, press processing oil, heat treatment oil, rust preventive oil, and cleaning oil. In this specification, etc., industrial oils such as lubricating oil, insulating oil, and processing oil may be collectively referred to as oil.

[0004] Lubricating oils include types such as engine oil, turbine oil, hydraulic oil, bearing oil, sliding surface oil, gear oil, compressor oil, cutting oil, etc. depending on the purpose of use. Various additives are blended into the base oil (the oil serving as the base material) so that various lubricating oils meet the required performance. Additives are also blended in other oils to obtain the properties required for each.

[0005] In recent years, the condition monitoring of machinery often adopts strategies to minimize the life cycle cost of the machinery. Large machinery such as power generation turbines uses a large amount of lubricating oil. Since lubricating oil replacement is carried out with the machinery stopped, there are negative aspects such as power generation loss and production stoppage. In addition, new oil purchase and delivery costs, oil replacement operation costs, waste oil treatment costs, etc. are required. Therefore, it is desirable to use the lubricating oil for as long as possible. Oil diagnosis is also carried out for the refrigerant liquid of electric vehicles and data centers, and replacement and hardware repairs are carried out. Similarly, the insulating oil of transformers is monitored for color and the like.

[0006] Also, recently, from the perspective of carbon neutrality, automobiles and the like that use a large amount of petroleum-derived fuels are being electrified. In the future, fuel demand will decrease, but industrial oils often have no alternative methods, and it is required to minimize the usage amount by extending the oil change cycle, etc. This is because reducing the oil consumption reduces the carbon dioxide emissions. However, overlooking the deterioration and contamination of the oil will lead to machinery failures.

[0007] Regarding the property diagnosis of lubricating oil, "deterioration" and "contamination" are defined and distinguished respectively. Generally speaking, it is necessary to diagnose two types: (1) the oxidative deterioration of lubricating oil over time, and (2) the contamination caused by external contaminants such as water, dust, and wear powder.

[0008] As for the oxidative deterioration of the lubricating oil in (1), there are deterioration due to the oxidation of the base oil and deterioration due to the consumption of additives. Due to the oxidative deterioration of the lubricating oil, a decrease in anti-wear performance, changes in viscosity and viscosity index, a decrease in rust prevention performance, a decrease in corrosion prevention performance, etc. occur. As a result, the wear and material fatigue of the speed increaser may be promoted. While it is desired to use the oil for as long as possible, it is necessary to promptly replace the oil and inspect the equipment in case of abnormal deterioration or contamination.

[0009] As a conventional lubricating oil diagnosis technique, Patent Document 1 discloses a system comprising a sensor for measuring the optical properties of a lubricating oil containing an additive having a diphenylamine skeleton, a storage device for storing correlation data, and a processing device for determining the remaining amount of the additive having a diphenylamine skeleton in the lubricating oil based on the data obtained by the sensor and the correlation data.

Prior Art Documents

Patent Documents

[0010]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0011] In Patent Document 1, a transmissive optical sensor including a visible light source and a light receiving element and measuring the chromaticity of a lubricating oil is used, and the remaining amount of an additive having a diphenylamine skeleton in the lubricating oil is quantified from the chromaticity of the lubricating oil obtained by the optical sensor. This method is excellent as a method capable of non-contact and remote measurement of the properties of oils such as lubricating oils.

[0012] However, the method of measuring the properties of a lubricating oil by chromaticity has difficulty in quantifying additives whose color does not change due to oxidative degradation.

[0013] Therefore, an object of the present invention is to provide a technique for optically quantifying a compound whose color does not change due to oxidative degradation.

Means for Solving the Problems

[0014] One aspect of the present invention uses an information processing system including an input device, an output device, a processing device, and a storage device. The information processing system includes an estimation model generation unit that generates a machine-learnable estimation model. The estimation model generation unit inputs an absorption spectrum obtained by transmitting light of at least a part of wavelengths from 800 nm to 3000 nm through an oil containing an additive, and a value reflecting the characteristics of the oil, uses a value based on the absorption spectrum as an explanatory variable, and uses a value reflecting the characteristics of the oil as an objective variable to generate an estimation model. This is a method for generating an estimation model of oil characteristics.

[0015] Another aspect of the present invention is a method for diagnosing oil characteristics using a diagnostic system including an information processing device implementing the above estimation model. The diagnostic system inputs an absorption spectrum obtained by transmitting light of at least a part of wavelengths from 800 nm to 3000 nm through a diagnostic target oil of the same type as the oil containing the additive, inputs a value based on the absorption spectrum into the estimation model, and obtains a value reflecting the characteristics of the diagnostic target oil from the estimation model. This is a method for diagnosing oil characteristics.

[0016] Another aspect of the present invention is a diagnostic system for oil characteristics including an information processing device implementing the above estimation model.

[0017] Another aspect of the present invention uses an optical sensor having a light source and a detector for detecting light emitted from the light source to obtain an absorption spectrum in the range of 800 nm to 3000 nm of an oil containing an additive having a maximum molar absorption coefficient of 50 or less in the ultraviolet-visible wavelength range of 250 nm to 800 nm, and inputs data based on the absorption spectrum into an estimation model implemented in an information processing device to predict the concentration of the additive. This is a method for diagnosing oil characteristics.

Advantages of the Invention

[0018] According to the present invention, a compound whose color does not change due to oxidative degradation can be optically quantified. Other problems, configurations, effects, etc. will become clear from the description of the following embodiments.

Brief Description of the Drawings

[0019]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26A

Figure 26B

Figure 26C

Figure 27

Mode for Carrying Out the Invention

[0020] The embodiments will be described in detail with reference to the drawings. However, the present invention is not construed as being limited to the description of the embodiments shown below. It will be readily understood by those skilled in the art that the specific configuration can be changed without departing from the spirit or gist of the present invention.

[0021] In the configurations of the examples described below, the same reference numerals are commonly used among different drawings for the same parts or parts having similar functions, and duplicate descriptions may be omitted.

[0022] When there are a plurality of elements having the same or similar functions, they may be described with different subscripts attached to the same reference numeral. However, when it is not necessary to distinguish the plurality of elements, the subscripts may be omitted in the description.

[0023] In this specification and the like, notations such as "first", "second", "third", etc. are attached to identify components, and do not necessarily limit the number, order, or content thereof. Also, numbers for identifying components are used for each context, and the numbers used in one context do not necessarily indicate the same configuration in other contexts. Further, it does not prevent a component identified by a certain number from also having the functions of a component identified by another number.

[0024] The positions, sizes, shapes, ranges, etc. of each configuration shown in the drawings and the like may not represent the actual positions, sizes, shapes, ranges, etc. in order to facilitate understanding of the invention. For this reason, the present invention is not necessarily limited to the positions, sizes, shapes, ranges, etc. disclosed in the drawings and the like.

[0025] Publications, patents, and patent applications cited in this specification constitute a part of the description of this specification as they are.

[0026] In this specification, components represented in the singular form shall include the plural form unless otherwise clearly indicated in the context.

[0027] In this specification and the like, the range from a wavelength of 250 nm to 800 nm is described as the ultraviolet-visible region, and the range from a wavelength of 800 nm to 2500 nm is described as the near-infrared region, respectively.

[0028] In this specification and the like, a large molar absorption coefficient at a certain wavelength indicates a large light absorption intensity at that wavelength, and a small molar absorption coefficient at a certain wavelength indicates a small light absorption intensity at that wavelength.

[0029] The monitoring method of a machine using the change in the light absorption intensity in the near-infrared region of oil containing additives such as lubricating oil according to the embodiment realizes the accurate quantification of the concentration of additives whose maximum value of the molar extinction coefficient in the ultraviolet-visible region is 50 or less, and further 20 or less, in the diagnosis of the deterioration and contamination of oil by an optical sensor in the near-infrared region. In principle, if (transmittance = 1 - absorbance), the transmittance and absorbance have the same meaning in the embodiment. Also, the transmittance and absorbance can be expressed as a percentage.

[0030] Lubricating oils include types such as engine oil, turbine oil, hydraulic operating oil, bearing oil, sliding surface oil, gear oil, compressor oil, cutting oil, etc. Lubricating oil is composed of a base oil and additives. Additives include antioxidants, rust inhibitors, defoamers, viscosity index improvers, oiliness improvers, extreme pressure additives, detergents, pour point depressants, emulsifiers, etc. Oils used for purposes other than lubrication include transformer oil, cleaning oil, cutting oil, etc. For many oils, the degree of additive consumption due to use serves as an indicator for determining the replacement time and mechanical abnormalities. The concentration of the additive can be quantified by measuring the change in the absorbance in the near-infrared region.

[0031] One example described in the embodiment is a diagnostic method for oil containing additives whose molar extinction coefficient in the range of wavelengths from 250 nm to 800 nm, that is, in the ultraviolet-visible region, is 50 or less. The method measures the intensity of the transmitted light of the oil with an optical sensor having a detector sensitive in the range of wavelengths from 800 nm to 2500 nm, that is, in the near-infrared region, to quantify the concentration of the additive and determine the state of the oil.

[0032] According to this configuration, when detecting the change in the light absorption intensity of the oil (measurement target) in the range of wavelengths from 250 nm to 800 nm by the optical sensor to detect the change in the properties of the oil, the state of the oil can be accurately determined by accurately quantifying the concentration of the additive that is difficult to change in color even when oxidized.

[0033] <Function of Additive> Lubricating oil is composed of a base oil and additives. Base oils include mineral oils made from petroleum, high-performance synthetic oils, bio-oils made from plants, biodegradable oils, etc. The deterioration of lubricating oil is an oxidation reaction involving oxygen. Antioxidants, which are typical additives, are added to prevent the oxidation of the base oil. Usually, when the antioxidant is depleted to a certain extent, the oxidation of the base oil begins. When the oxidation of the base oil starts, changes in viscosity occur, and the thickness of the lubricating film changes, etc., resulting in a decrease in lubricating performance. Therefore, generally, oil change is recommended. Therefore, by monitoring the depletion degree of the antioxidant, it becomes possible to estimate the remaining life of the lubricating oil.

[0034] Among the additives, extreme pressure additives and friction inhibitors have the function of preventing wear on the sliding surfaces of parts. The depletion of these extreme pressure additives and friction inhibitors leads to an acceleration of part wear. Therefore, the remaining life of the lubricating oil may be determined from the perspective of the concentration of extreme pressure additives and friction inhibitors. Therefore, by monitoring the depletion degree of extreme pressure additives and friction inhibitors, it becomes possible to estimate the remaining life of the lubricating oil. When any of the remaining amounts or concentrations of the antioxidant, extreme pressure additives, and friction inhibitors reaches a threshold value, it is determined as the life of the lubricating oil. Generally, at this time, generally, oil change is recommended.

[0035] Regarding other additives such as rust inhibitors, defoamers, viscosity index improvers, oiliness improvers, detergents, pour point depressants, and emulsifiers, it is known that they are consumed with the use of lubricating oil. Threshold values are set from the functional aspect of the lubricating oil, and they may be used as indicators for oil change determination.

[0036] <Deterioration and Coloring of Additives> The deterioration of organic compounds is generally oxidation by reaction with oxygen and active radicals generated from oxygen.

[0037] Compounds with an aromatic ring tend to form colored compounds when oxidized. Compounds without an aromatic ring, such as aliphatic hydrocarbons used as base oils, polyalkylene oxides (PAOs), carbamates used as additives, esters of alkyl alcohols and phosphoric acid, surfactants, dispersants, defoamers, etc., strongly tend not to form colored compounds when oxidized. This is because non-aromatic compounds have no electron conjugation system or, if any, a small conjugation system.

[0038] <Color of Organic Compounds> The relationship between the structure and color of organic compounds is explained. There are two types of how color appears: the color of luminescence and the color visible by light absorption. The color of organic compounds is manifested by absorbing light of specific wavelengths in the visible light range from 400 nm to 800 nm. Specifically, for example, in the color wheel, red and green, yellow and blue, etc. are in opposing positions, which is called a complementary color relationship. Absorbing light of short-wavelength blue makes it appear yellow, and absorbing light of long-wavelength red makes it appear green.

[0039] Light absorption means that a molecule can absorb the energy that the light has. There are multiple modes of light absorption by molecules depending on the wavelength. Light absorption in the visible light region is, energetically, due to the excitation of valence electrons of molecules, and it occurs when electrons move from an orbital with relatively low energy to another orbital with higher energy. At this time, if there is no overlap of the orbitals between the orbital where the electron is initially located and the orbital where the electron is located after excitation, the electron cannot move. The bond between atoms is formed by the overlap of electron orbitals.

[0040] Here, the σ bond and π bond of organic compounds are explained. For example, a carbon atom can use four hands to bond with other atoms or molecules. When using one hand to bond with another atom such as hydrogen, it is a σ bond. The σ bond has a very high bond energy and strong bonding force, so it is stable. Since the σ bond is connected by only one hand, it can rotate freely.

[0041] On the one hand, π bonds are formed by using two or three bonds. These are called double bonds and triple bonds. In a double bond, after extending a bond perpendicular to the bond axis between atoms, the state of struggling to form a bond is the π bond. In a double bond, there is one σ bond and one π bond, and the bond is weak. A triple bond is composed of one σ bond and two π bonds. A π bond is unstable and highly reactive compared to a σ bond. However, compounds having an aromatic ring such as benzene and substituted benzene, naphthalene, and anthracene have an electronic state called conjugation, and the electrons are spread over the entire aromatic ring, resulting in a very stable structure.

[0042] Next, the relationship between the bond between atoms and color will be explained. The energy difference of electron transition in a σ bond, that is, the energy difference between the bonding orbital and the antibonding orbital of the σ bond, is 500 kJ / mol or more. In terms of wavelength, it is 200 nm or less, so it only absorbs ultraviolet light that is not visible as color. In a conjugated π bond, due to the spread of electrons, there is an effect of reducing the transition energy, the wavelength of the absorbed light becomes longer, and light in the ultraviolet region from 200 nm to 400 nm is absorbed.

[0043] <Oxidation of Compounds with Aromatic Rings> Figure 1 is a diagram showing the molecular structure of benzene. Figure 2 is a diagram showing the molecular structure of naphthalene. Figure 3 is a diagram showing the molecular structure of phenol. Figure 4 is a diagram showing the molecular structure of aniline.

[0044] Among the compounds having an aromatic ring, compounds having a highly symmetric structure such as benzene (Figure 1) and naphthalene (Figure 2) have no electron bias, so they have very low reactivity and are chemically stable. That is, they are difficult to be oxidized.

[0045] On the other hand, in aromatic compounds having substituents such as phenol (Figure 3) and aniline (Figure 4), the symmetry of the molecular structure is reduced and there is an electron bias in the molecule. Therefore, chemical reactions easily occur at bonds with low energy such as the C-O bond in phenol.

[0046] As antioxidants added to most lubricating oils, phenolic compounds and phenylamine compounds are typical. The function of the antioxidant utilizes such reactivity. That is, by the antioxidant itself being oxidized, it prevents the base oil and other additives from being oxidized and losing their functions.

[0047] It is known that phenolic antioxidants and amine antioxidants produce quinone compounds that strongly absorb visible light when oxidized (see, for example, Patent Document 1).

[0048] Figure 5 is a diagram showing the reaction in which the typical phenolic antioxidant BHT is oxidized to produce a benzoquinone derivative. For example, when benzene and p-benzoquinone are compared, the absorption maximum of benzene is 255 nm and it is colorless, while p-benzoquinone has an absorption maximum at 440 nm and is a vivid yellow compound. For example, even when an alkyl group is bonded to the benzene ring, since the electron conjugation does not spread to the alkyl group, there is no significant change in the absorption wavelength for alkyl-substituted phenols such as BHT.

[0049] Figure 6 shows the structures of TPP and TPPT, which are phosphorus-based additives. Among the compounds widely used as phosphorus-based antioxidants and phosphorus-based extreme pressure agents, there are ester compounds of phosphoric acid with phenol and substituted phenols, such as TPP and TPPT. These function as antioxidants. Also, when heated on the sliding surface, they decompose to produce phenol (or substituted phenol), and further, the phenol is oxidized to produce quinone. Although compounds other than quinone are also produced from compounds having an aromatic ring, generally, colored compounds are likely to be produced.

[0050] <Compound that does not color even when oxidized> The deterioration of organic compounds is generally oxidation due to reaction with oxygen and active radicals generated from oxygen.

[0051] Compounds without an aromatic ring, such as aliphatic hydrocarbons used as base oils, polyalkylene oxides (PAO), carbamates used as additives, esters of alkyl alcohols and phosphoric acid, surfactants, dispersants, defoamers, etc., tend not to form colored compounds even when oxidized. This is because non-aromatic compounds have no electron conjugation system or, if any, a small conjugation system.

[0052] <Molar absorption coefficient> Aromatic compounds tend to form colored compounds when oxidized. However, from the perspective of the ultraviolet-visible absorption spectrum, they tend to have stronger absorption in the ultraviolet region compared to compounds without an aromatic ring.

[0053] The molar absorption coefficient ε is defined as the optical density of a 1-mol solution per centimeter of light path, that is, the amount representing the intensity of absorption per centimeter of a 1-mol solution of a certain substance for light of a certain wavelength, and is expressed by Equation (1). A = -logT = -log(I / I₀) = ε × C × l... Equation (1) A: Absorbance, T: Transmittance, I: Intensity of transmitted light, I₀: Intensity of incident light, C: Concentration (mol / L), ε: Molar absorption coefficient (L / (mol·cm)), l: Light path length (cm)

[0054] The larger the molar absorption coefficient ε, the stronger the color development, the more sensitive it becomes, and the sensitivity of quantification increases. Aromatic compounds generally have a molar absorption coefficient ε of 100 or more at the maximum absorption wavelength in the ultraviolet to visible region from 250 nm to 800 nm in wavelength. For example, the ε of benzene is 180. When substituents are added to the benzene ring, the absorption peak shifts to a longer wavelength and ε tends to increase. That is, aromatic compounds can be measured sensitively in the ultraviolet to visible region.

[0055] On the one hand, the molar extinction coefficient ε at the maximum absorption wavelength in the ultraviolet to visible region from 250 nm to 800 nm of a compound without an aromatic ring is 50 or less. That is, for a compound without an aromatic ring, the sensitivity in the ultraviolet to visible region is low.

[0056] <Additives involved in and not involved in the coloring due to the oxidation of lubricating oil> As described above, the additives involved in the coloring due to the oxidation of lubricating oil are additives having an aromatic ring. The additives not involved in the coloring are compounds without an aromatic ring. There is a need to monitor the concentration of additives not involved in the coloring.

[0057] <Quantification of additives using near-infrared absorption spectra> As a result of the inventors' studies, it has been found that it is possible to quantify additives that are difficult to color due to oxidative degradation in lubricating oil, that is, compounds without an aromatic ring, by measuring the near-infrared light absorption spectrum from 800 nm to 2500 nm using an optical sensor.

[0058] Figure 7 is a diagram showing one form of an optical sensor. The optical sensor 700 incorporates a light source 701 that generates at least a part of the wavelengths of near-infrared light from 800 nm to 3 μm (3000 nm), for example, and a detector 703 capable of detecting the light generated from the light source 701. The light radiated from the light source 701 (indicated by the arrow in the figure) passes through the lubricating oil 702 that is the object to be measured, and the transmitted light is measured by the detector 703. The optical sensor 700 only needs to be able to obtain the absorption spectrum of at least a part of the wavelengths of near-infrared light from at least 800 nm to 3 μm (3000 nm).

[0059] In near-infrared spectroscopy, the vibrational energy of atomic bonds in organic compounds is measured. The near-infrared wavelength region corresponds only to overtones and combination tones of the fundamental tone and does not detect the fundamental tone, so it is characterized by low absorption intensity. The reason for the low absorption is that overtones and combination tones are forbidden transitions with a low probability of occurrence.

[0060] The advantages of near-infrared spectroscopy for measuring low-probability forbidden transitions are excellent permeability, and since the concentration saturation problem that is problematic in ultraviolet-visible spectroscopy and mid-infrared spectroscopy is less likely to occur, it is not necessary to extremely shorten the optical path length in liquid measurements such as lubricating oil, and an optical path length of about 3 mm to 20 mm can be selected.

[0061] For spectral analysis for the quantification of additives in lubricating oil, spectral preprocessing is performed as in the following representative examples. In the first step, smoothing for removing spectral noise is performed. For example, a method is used in which the method of moving average centered is applied to a selected section of about 3 to 20 adjacent points. In the second step, by applying a second derivative filter to the spectrum, wavelengths with large changes in the amount of change in the spectrum can be extracted, and peaks can be extracted from spectra that are characteristically complex and overlapping in the near-infrared absorption spectrum and have unclear peak positions. These two steps may be executed in either order.

[0062] Next, multivariate analysis, which is a type of machine learning, is performed on the preprocessed spectrum. For the quantification of additive concentration, the prediction method of PLS (Partial Least Square) regression analysis can be used. To create a calibration curve, a data set is created with the spectrum as the explanatory variable and the additive concentration obtained by a quantitative analysis method such as HPLC (High Performance Liquid Chromatography) as the target variable, and PLS regression analysis is performed. Cross-validation can be used for the validation of PLS regression analysis using the data set. Using the created additive concentration estimation model, the additive concentration is estimated for lubricating oil with an unknown additive concentration.

[0063] <Form of measurement by sensor> The measurement of the near-infrared absorption spectrum of lubricating oil is performed by an optical sensor 700 having a light source that generates light with wavelengths in the near-infrared region and a detector that detects wavelengths in the near-infrared region. For the lubricating oil to be measured, the detector 703 detects the light intensity after the light emitted from the light source 701 passes through the lubricating oil 702.

[0064] Although Fig. 7 shown above illustrates one form of the sensor, the light emitted from the light source 701 can also be incident on the detector 703 after passing through the lubricating oil 702 by means such as reflection by a mirror, light collection by a lens, and passing through a prism.

[0065] Fig. 8 is a diagram showing another form of the sensor. The optical sensor 700A has the light source 701 and the detector 703 arranged on the same plane, brought into contact with the lubricating oil 702, and a reflector 704 is installed on the opposite side. There is a form in which the light emitted from the light source 701 (indicated by an arrow in the figure) is reflected by the reflector 704 and received by the detector 703. Even in this form, a lens, a mirror, or a prism may be installed in the middle of the optical path.

[0066] <Additive to be targeted> In this embodiment, the properties of an additive that does not contain an aromatic ring in its molecular structure can be optically measured. Examples of additives composed of organic compounds that do not contain an aromatic ring are shown below.

[0067] Fig. 9 is a diagram showing the structure of a sulfur-containing additive. Fig. 10 is a diagram showing the molecular structure of an alkyl ester-based additive of phosphoric acid. Here, R is an alkyl group. Fig. 11 is a diagram showing the molecular structure of a polyacrylate-type dispersion viscosity modifier.

[0068] <Scope of application> The types of oils containing additives include engine oil, turbine oil, hydraulic oil, bearing oil, sliding surface oil, gear oil, compressor oil, cutting oil, insulating oil, cutting processing oil, press processing oil, heat treatment oil, rust preventive oil, cleaning oil, etc. The types of additives include antioxidants, rust inhibitors, defoamers, viscosity index improvers, oiliness improvers, extreme pressure additives, detergents, pour point depressants, emulsifiers, etc.

Example

[0069] Figure 12 shows an example of a system used for diagnosing engine oil in an embodiment. The diagnostic system 1200 of the embodiment is configured using a general computer such as a server. The diagnostic system 1200 includes an input device 1210, an output device 1220, and a processing device 1230, and the storage device 1240 stores programs and data for processing. The storage device 1240 may be configured by combining known storage devices such as semiconductor memories and magnetic disk devices.

[0070] The storage device 1240 includes an estimation model generation unit 1250 and an oil property estimation unit 1260. In this embodiment, the estimation model generation unit 1250 and the oil property estimation unit 1260 are included in the same device, but they can also be configured by separate devices.

[0071] The estimation model generation unit 1250 includes an actual measurement database 1251, a preprocessing unit 1252A, a learning database 1253, and a multivariate analysis unit 1254.

[0072] The oil property estimation unit 1260 includes a preprocessing unit 1252B, an estimation model 1261, and a characteristic conversion table 1262.

[0073] Figure 13 shows an example of diagnosing engine oil using the diagnostic system 1200. First, actual measurement data is acquired from actual engine oil to generate the actual measurement database 1251. This is performed by collecting samples of engine oil or analyzing engine oil by known methods (S1310).

[0074] Using the preprocessing unit 1252A of the estimation model generation unit 1250, explanatory variables and objective variables for model generation are prepared from the actual measurement data and stored in the learning database 1253. The multivariate analysis unit 1254 generates an estimation model 1261 of the oil properties using the explanatory variables and the objective variables (S1320).

[0075] Using the obtained estimation model 1261, the oil property estimation unit 1260 estimates the oil properties (S1330).

[0076] In this example, the viscosity of the new engine oil is 20 cP, and it is recommended to replace the oil when the viscosity reaches 25 cP due to use. This engine oil experiences an increase in viscosity due to the consumption of the antioxidant, but does not contain an antioxidant with an aromatic ring.

[0077] An antioxidant composed of a sulfur-containing additive shown in FIG. 9 was used in the engine oil, and the concentration in the new oil was 3 wt%. The molar extinction coefficient of this antioxidant at the maximum absorption wavelength between 250 nm and 400 nm in wavelength was 42.

[0078] FIG. 14 is a flowchart of the measured data collection process S1310. The above engine oil was applied to an automobile engine and continuously operated. Every 10 hours (S1311), 10 ml of engine oil was sampled (S1312), and the absorption spectrum in the near-infrared region was measured (S1313) using the optical sensor shown in FIG. 7 or FIG. 8. The wavelength resolution of the near-infrared absorption spectrum was 1 nm. Also, for the sampled engine oil, the concentration of the antioxidant was quantified using LC / MS, which is a type of high-performance liquid chromatography (HPLC). Further, the viscosity and total acid value of the engine oil were measured by any known method (S1314). Also, the RGB color coordinates were measured by the method described in Patent Document 1 etc. (S1315). The measured near-infrared region absorption spectrum, antioxidant concentration, viscosity, total acid value, and RGB color coordinates were associated with time information and recorded in the measured database 1251 (S1316). Note that the sample acquisition amount and acquisition time interval are examples, and the analysis method for the oil characteristics may be arbitrarily selected from known ones.

[0079] FIG. 15 shows, as an example of the data representation of the measured database 1251, a graph of the relationship between the usage time and the viscosity, total acid value, and antioxidant concentration of the oil. By preparing such a relationship as reference data, the total acid value can be obtained from the antioxidant concentration. Also, the viscosity can be obtained from the antioxidant concentration.

[0080] Furthermore, the near-infrared absorption spectrum of the collected engine oil was measured by the optical sensor shown in FIG. 7 or FIG. 8 (S1313), and the RGB color coordinates were separately measured in the visible light region by the method described in Patent Document 1 (S1315). These data were also stored in the measured database 1251 in association with time. Note that the measurement of the RGB color coordinates is for confirming the measurability by visible light and may be omitted in actual operation.

[0081] FIG. 16 shows an example of the near-infrared absorption spectrum of engine oil. In this example, the absorbance with respect to wavelengths from 1550 nm to 1950 nm is shown. The plurality of curves show the spectra at points in time after different usage times.

[0082] Next, explanatory variables are generated using the measured near-infrared absorption spectrum, and an oil property estimation model is generated with the antioxidant concentration measured by LC / MS as the target variable (S1320).

[0083] In generating the model, model generation using known machine learning with the explanatory variable as the input and the target variable as the output can be used. In this embodiment, PLS regression analysis was performed.

[0084] FIG. 17 shows a flowchart of the oil property estimation model generation process S1320. This process is to be performed by the estimation model generation unit 1250 of a general computer by the process executed by the processing device 1230 for the software.

[0085] First, the preprocessing unit 1252A reads the data of the near-infrared absorption spectrum at a certain point in time from the measured database 1251 (S1321). For the preprocessing of the near-infrared absorption spectrum in the preprocessing unit 1252A, smoothing (S1322) by the Savitzky-Golay method (SG method) using the data of adjacent 9 points and second differentiation of the spectrum (S1323) were performed.

[0086] FIG. 18 shows a graph of the result of differentiating the spectrum twice. The plurality of curves show the differentiation of the spectrum at points in time after different usage times. By differentiating, the differences in the near-infrared absorption spectrum over time can be emphasized. There may be cases where differentiation is not necessary. Also, differentiation may be performed only once. Alternatively, differentiation may be performed three or more times.

[0087] The near-infrared absorption spectrum differentiated twice is stored in the learning database 1253 with the explanatory variable x and the antioxidant concentration at the same time as the target variable y.

[0088] The multivariate analysis unit 1254 performs PLS regression analysis using the explanatory variable x and the target variable y (S1324) and generates an oil property estimation model 1261. After the PLS regression analysis, cross-validation was performed (S1325). The generated estimation model 1261 is implemented in the oil property estimation unit 1260.

[0089] FIG. 19 is a flowchart of the oil property estimation process S1330 using the generated estimation model 1261. This process is to be performed by the oil property estimation unit 1260 of a general computer by software processing.

[0090] First, the near-infrared absorption spectrum of the engine oil to be judged for its properties is obtained and input from the input device 1210 of the diagnostic system 1200 (S1331). The near-infrared absorption spectrum is obtained by the optical sensor shown in FIG. 7 or FIG. 8. As one of the features of the embodiment, such a near-infrared absorption spectrum can be collected non-contact and remotely.

[0091] The near-infrared absorption spectrum is pre-processed by the pre-processing unit 1252B, such as smoothing and double differentiation, in the same manner as when the estimation model was generated in FIG. 17 (S1332). The pre-processed near-infrared absorption spectrum is input to the oil property estimation model, and the antioxidant concentration is estimated (S1333).

[0092] The oil property estimation unit 1260 can have a characteristic conversion table 1262 that converts the antioxidant concentration, viscosity, and total acid value mutually using all or part of the measured database 1251 (not necessarily in a table format, and may be data showing the relationship between the antioxidant concentration and viscosity as shown in FIG. 15).

[0093] Estimate the viscosity of the engine oil based on the estimated antioxidant concentration and the characteristic conversion table 1262 (S1334). Using the result of this analysis, it was found that when the antioxidant decreased to 40% of the initial concentration, the viscosity of the engine oil reached 25 cP.

[0094] FIG. 20 is a graph showing the consistency between the predicted value and the measured value of the antioxidant concentration using the estimation model 1261. The reference value is the measured value of the sample, and as a result of analysis using the PLS model, the obtained one is the PLS regression equation. When the result predicted from the measured value using this regression equation (estimation model) is the predicted value and the reference value and the predicted value match well, a good PLS model can be constructed. Using FIG. 20 as a calibration curve, the concentration of a sample with an unknown additive concentration can be quantified.

[0095] On the other hand, the correlation coefficient between the RGB color coordinates of the collected oil and the antioxidant concentration was as small as 0.5, and it was found that it was difficult to predict the viscosity using the color coordinates for organic compounds having no benzene ring.

[0096] In the above example, finally, the viscosity of the engine oil was obtained and output from the output device 1220. However, as will be described later, the total acid value may be output, or the concentration or the usage time may be output. Also, the generation of the estimation model used multivariate analysis by software that can execute PLS without using special hardware, but other known machine learning using a GPU (Graphic Processor Unit) or the like may be adopted.

[0097] In the above example, the near-infrared absorption spectrum was used as the explanatory variable x, and the antioxidant concentration was used as the objective variable y. However, it is also possible to use values that reflect other properties of the oil (such as the concentrations of other additives, viscosity, total acid number, usage time, etc.) as the objective variable.

Example

[0098] A diagnostic example of the hydraulic oil for large ships (hereinafter referred to as hydraulic oil) is shown. This hydraulic oil contained 2% of an alkyl ester-based extreme pressure agent shown in Fig. 10 in the new oil. The acid number of this new hydraulic oil was 1.6, and it was specified that it should be replaced when the acid number reached 2. From past experience, it was known that it would reach the state requiring replacement after an average of 3000 hours of use.

[0099] Under the same conditions as those for use in large ships, the hydraulic oil was continuously used, and 15 ml was sampled every 10 hours. Using the optical sensor shown in Fig. 7, the absorption spectrum in the near-infrared region of the sampled oil was obtained. Also, using the sampled oil, the acid number, viscosity, and contamination level (mass method) were determined respectively. When the correlation coefficients between the total acid number, viscosity, and contamination level obtained in the sampled oil and the antioxidant concentration were calculated, the correlation coefficient between the acid number and the antioxidant concentration was the largest, being 0.98. Therefore, it was found that the acid number could be quantified by quantifying the antioxidant concentration.

[0100] The obtained near-infrared absorption spectrum was used as the explanatory variable for PLS regression analysis, and the antioxidant concentration was used as the objective variable. After performing 11-point smoothing and second derivative on the near-infrared absorption spectrum, PLS regression analysis was carried out. Cross-validation was performed on this result.

[0101] From the results of this analysis, it was found that it was possible to predict the antioxidant concentration from the near-infrared absorption spectrum. Also, it was found that it was possible to predict the acid number from the antioxidant concentration.

Example

[0102] Shows a diagnostic example of gas engine oil. This new gas engine oil contains 1% of a polyacrylate type dispersant viscosity modifier with the structure shown in Fig. 11. The viscosity of the new oil is 10 cP, and it is recommended to replace the oil when the viscosity reaches 13 cP.

[0103] This gas engine was continuously operated, and an optical sensor capable of measuring the near-infrared absorption spectrum was installed in the sight glass (a component made of a transparent material and installed so that the color of the oil can be visually observed) provided in the engine oil piping.

[0104] During the operation of the engine, the near-infrared absorption spectrum was measured every hour and continued until 1000 hours. Also, a small amount of engine oil was sampled every 20 hours, and viscosity measurement and viscosity modifier concentration measurement were carried out. The results of the viscosity measurement and viscosity modifier concentration measurement can be shown in a graph like Fig. 15, similar to Example 2.

[0105] PLS regression analysis was performed using the obtained near-infrared absorption spectrum, separately measured viscosity data, and viscosity modifier concentration data to generate an estimation model for the viscosity modifier concentration. It was confirmed that the viscosity and viscosity modifier concentration can be predicted from the near-infrared absorption spectrum of the engine oil using the estimation model.

Example

[0106] This example applies the configurations of Examples 1 to 3 to a monitoring system and method for the lubricating oil of a wind turbine generator. This example is a monitoring system for the lubricating oil supplied to the mechanical drive part of a wind turbine generator. This system includes the diagnostic system 1200 shown in Fig. 12.

[0107] The storage device in the monitoring system stores, as a reference, additive concentration data that stores the concentration of the lubricating oil additive in time series. The diagnostic system 1200 estimates the time when the additive concentration in the lubricating oil obtained from the near-infrared absorption spectrum of the lubricating oil reaches a predetermined threshold value.

[0108] (1. Overall system configuration) Fig. 21 shows a schematic diagram of a lubricating oil monitoring system for a wind turbine having a lubricating oil supply system. Inside the nacelle 3 of the wind turbine 1, there are a main shaft 31, a speed increaser 33, a generator 34, and bearings such as yaw and pitch (not shown), and lubricating oil is supplied to these components from an oil tank 37. In addition, the wind turbine also has a general configuration including a hub 4, a nacelle partition 30, a shrink disk 32, a main frame 35, a radiator 36, a coupling 38, etc.

[0109] As shown in Fig. 21, usually a plurality of wind turbines 1 are installed within the same site, and these are collectively called a farm 200a or the like. For each wind turbine 1, various sensors (not shown) are installed in the lubricating oil supply system, and the sensor signals reflecting the state of the lubricating oil are aggregated in a server 210 inside the nacelle 3.

[0110] Also, the sensor signals obtained from the server 210 of each wind turbine 1 are sent to an aggregation server 220 arranged for each farm. The data from the aggregation server 220 is sent to a central server 240 via a network 230. Data from other farms 200b and 200c is also sent to the central server 240. In addition, the central server 240 can send instructions to each wind turbine 1 via the aggregation server 220 and the server 210. The central server 240 basically has the functions of the diagnostic system 1200 shown in Fig. 12, and the system of the embodiment enables remote monitoring of oil without a benzene ring.

[0111] (2. Sensor Arrangement) Fig. 22 is a conceptual diagram of a rotating component equipped with a lubricating oil sensor. Lubricating oil is supplied from a lubricating oil supply device 301 such as a pump to a rotating component 302. The lubricating oil supply device 301 is connected to an oil tank 37 and receives the supply of lubricating oil. The rotating component 302 is, for example, a general part where mechanical contact occurs in a speed increaser 33 or other components, and is not particularly limited.

[0112] An optical sensor 304 is arranged in a lubricating oil flow path or the like to detect the state of the lubricating oil. Specific examples of the optical sensor 304 are shown in Fig. 7 or Fig. 8.

[0113] In this embodiment, a transparent measurement unit 303 is provided in a flow path (near the end of the lubricating oil path) branched from the flow path of the lubricating oil connected to the oil drain port of the rotating part 302, and a part of the lubricating oil is introduced into this measurement unit 303. And an optical sensor 304 is installed in the measurement unit 303. The reason for not providing the measurement unit 303 in the main flow path of the lubricating oil is to adjust the flow rate of the lubricating oil in the measurement unit 303 to a flow rate suitable for detecting the state of the lubricating oil. The lubricating oil discharged from the rotating part 302 returns to the oil tank 37 via the filter 305. Note that the filter 305 is not essential. The optical sensor 304 measures the near-infrared absorption spectrum of the oil. The state of the lubricating oil can be evaluated based on the temporal change of the near-infrared absorption spectrum of the lubricating oil.

[0114] And in this embodiment, the optical sensor 304 includes an optical sensor including a near-infrared light source and a light receiving element. The near-infrared absorption spectrum of the lubricating oil is acquired by the optical sensor. The optical sensor 304 transmits the near-infrared light from the near-infrared light source into the oil, and acquires the near-infrared absorption spectrum by detecting the near-infrared light transmitted through the oil with a light receiving element having sensitivity according to the wavelength. The acquired near-infrared absorption spectrum reflects the absorption rate or transmittance of the oil for the light of each wavelength. Using the acquired near-infrared absorption spectrum and the estimation model, the amount of residual additive in the lubricating oil is obtained, and the degradation degree diagnosis and the remaining life diagnosis are performed.

[0115] The quality of the lubricating oil deteriorates during use and it can no longer perform its initial functions. Therefore, it is necessary to perform maintenance such as replacement according to the deterioration status of the quality. In order to know the timing of such maintenance, it is useful for the efficiency of maintenance management to be able to monitor the data that can be collected by the optical sensor 304 installed on site at a remote location. The data collected by the optical sensor 304 is collected, for example, by the server 210 in the nacelle 3, and then sent to the central server 240 that aggregates the data of multiple farms via the aggregation server 220 that aggregates the data within the farm 200.

[0116] However, for analyses that require equipment for measurement, such as LC (liquid chromatography) measurement, FT-IR (Fourier transform infrared spectroscopy) measurement, and NMR (nuclear magnetic resonance) measurement, it is necessary to appropriately collect samples of the lubricating oil and perform the analysis using separately provided equipment. The results measured by these LC measurement, FT-IR measurement, and NMR measurement should also be stored as data in a separately provided central server 240, the data should be aggregated, and it is desirable to grasp the properties of the lubricating oil in consideration of these data.

[0117] In addition, the data to be aggregated may include not only data related to the lubricating oil but also data indicating the operating status of the wind turbine. For example, the windmill output value (the greater the value, the greater the deterioration rate of the lubricating oil), the actual operating time (the longer the time, the greater the deterioration rate of the lubricating oil), the machine temperature (the higher the temperature, the greater the deterioration rate of the lubricating oil), the rotational speed of the shaft (the faster the speed, the greater the deterioration rate of the lubricating oil), and so on. These can be collected from sensors with a known configuration installed at various locations of the wind turbine and from the control signals of the device.

[0118] (3. Flow of Lubricating Oil Diagnosis) FIG. 23 is a flowchart showing the lubricating oil diagnosis process according to this embodiment. The process shown in FIG. 23 is performed under the control of any one of the servers 210, aggregation server 220, and central server 240 in FIG. 20. In the following example, it is assumed that the central server 240 performs the process. Functions such as calculation and control are realized by software stored in the storage device of the server being executed by the processor in cooperation with other hardware to perform the defined process. Note that functions equivalent to those configured by software can also be realized by hardware such as FPGA (Field Programmable Gate Array) and ASIC (Application Specific Integrated Circuit).

[0119] When the central server 240 performs control, since it has a plurality of wind turbines 1 under its jurisdiction, the following processes shall be performed for each wind turbine. This process is basically a repetitive process, and the start timing is set by a timer or the like. For example, the process starts at 0:00 every day (S601). Also, the central server 240 can be performed at an arbitrary timing according to the operator's instruction.

[0120] In process S602, the central server 240 checks the replacement time of the lubricating oil. The initial value of the replacement time can be calculated physically using the Arrhenius reaction rate on the premise that the lubricating oil is operating at the design temperature, for example, and the remaining life can be initially set. This replacement time can be updated later in process S610 based on the measured data.

[0121] If it is the replacement time of the lubricating oil, the lubricating oil is replaced in process S603. Since the replacement of the lubricating oil is usually a task performed by an operator, the central server 240 performs a display and notification for instructing the operator of the time and target for replacement.

[0122] If it is not the replacement time of the lubricating oil, in process S604, the central server 240 performs a diagnosis using sensor data. As the sensor data, in addition to the near-infrared absorption spectrum of the lubricating oil obtained by an optical sensor, the oil temperature, oil pressure, concentration of particles contained in the lubricating oil, etc. that can be measured by the prior art can be used. The data collected by the optical sensor 304 is sent to the central server 240. For example, the central server compares the oil temperature, oil pressure, and concentration of particles contained in the lubricating oil obtained from the sensor with the previously determined threshold values to evaluate the characteristics of the lubricating oil.

[0123] If the result of the diagnosis in process S605 is abnormal, the lubricating oil is replaced in process S603. If there is no abnormality, process S606 is performed.

[0124] Note that in this embodiment, it is also possible to use the near-infrared absorption spectrum of the lubricating oil and obtain and utilize the B / R value of the oil as disclosed in the conventional Patent Document 1. For example, in process S605, for example, when the B / R value changes from a decrease to an increase based on the R, G, and B values of the optical sensor, it is determined that there is a contamination abnormality. Thus, in process S605, it is also possible to make a determination using the G / R value, B value, G value, and ΔE value. Details of such evaluation of the characteristics of the oil using color information and calibration curves are described in Patent Document 1.

[0125] In S605, using the correlation between the viscosity and the remaining additive amount, when the viscosity corresponding to the remaining additive amount estimated from the near-infrared absorption spectrum measured by the optical sensor exceeds a predetermined threshold value, it is determined that there is a viscosity abnormality. Note that it is also possible to determine that there is an abnormality when the remaining additive amount becomes smaller than a predetermined threshold value without obtaining the viscosity. The method for estimating the remaining additive amount from the near-infrared absorption spectrum is as described in Examples 1 to 3.

[0126] In process S606, the near-infrared absorption spectrum, chromaticity measurement data, etc. are input to the central server 240, and the data is stored in time series.

[0127] From the viewpoints of preventive maintenance and planned maintenance of the wind turbine generator, it is desirable to perform a prognostic diagnosis of the deterioration of the lubricating oil based on the transition of the concentration of the additive contained in the lubricating oil before it is determined that there is an abnormality.

[0128] FIG. 24 is a graph showing the concept of the antioxidant concentration of the lubricating oil stored in time series. The horizontal axis represents time (months), and the vertical axis represents the antioxidant concentration. For example, it is assumed that the antioxidant concentration is observed at fixed points, and the antioxidant concentrations up to the passage of 60 months are plotted. A significant relationship is recognized between the elapsed time and the antioxidant concentration. For example, the antioxidant concentration decreases linearly with time.

[0129] In this example, in process S607, the viscosity threshold is set to 200, and the time point when the viscosity estimated from the additive concentration measurement results stored in time series reaches 200 is estimated as the replacement time. As the estimation method, various known methods may be adopted. If actual measurement values are obtained, a known method of extrapolating data can be used on the premise that the viscosity increases monotonically. Further, when the viscosity changes more complexly, a known method such as function fitting (curve fitting) can be used.

[0130] Note that in this embodiment, the time-series near-infrared absorption spectra measured by the optical sensor are stored, and the degree of deterioration of the lubricating oil is estimated based on them.

[0131] The replacement time estimation result by process S607 can be displayed as the lubricating oil diagnosis result (process S608).

[0132] FIG. 25 shows an example of the result display by process S610. After 50 months, since the viscosity is predicted to reach 200, the time before that (for example, half a month before) can be set as the new replacement time. One cycle of processing in process S610 is completed, and in the next cycle of process S602, the determination process is performed according to the new replacement time.

[0133] Note that, for example, after S608, the near-infrared absorption spectrum and chromaticity data measured by the optical sensor can be displayed on the display screen of the lubricating oil diagnosis result. By thus displaying the deterioration state of the lubricating oil in color on the display screen, the operator can visually recognize the deterioration state of the lubricating oil. This helps, for example, the operator to roughly grasp the deterioration state of the lubricating oil when visually observing the state of the lubricating oil on site.

Example

[0134] FIGS. 26A to 26C show examples of organic compounds that can be quantified using the near-infrared absorption spectra measured by the optical sensor described in Examples 1 to 4. The structural formulas (a) to (w) are all antioxidants. The names of the structural formulas after (g) are as follows. (g) Triphenyl phosphite (h) Tris(2,4-ditert-butylphenyl) phosphite (i) Isodecyl diphenyl phosphite (j) 2,2'-Methylenebis(4,6-di-tert-butylphenyl) 2-ethylhexyl phosphite (k) 3,9-Bis(2,6-di-tert-butyl-4-methylphenoxy)-2,4,8,10-tetraoxa-3,9-diphosphaspiro [5.5] undecane (l) 3,9-Bis(octadecyloxy)-2,4,8,10-tetraoxa-3,9-diphosphaspiro[5.5]undecane (m) Tris(nonylphenyl) phosphite (n) Diphenylamine (o) Substituted diphenylamine (p) Substituted diphenylamine (q) 4,4'-Bis(α,α-dimethylbenzyl)diphenylamine (r) N,N'-Di-sec-butyl-1,4-phenylenediamine (s) N-(1,3-Dimethylbutyl)-N'-phenyl-1,4-phenylenediamine (t) N-Isopropyl-N'-phenyl-1,4-phenylenediamine (u) N,N'-Diphenyl-1,4-phenylenediamine (v) 1-Anilinonaphthalene (w) 6-Ethoxy-2,2,4-trimethyl-1,2-dihydroquinoline

[0135] Note that R shown by (o) in the figure represents a linear or branched alkyl group having 2 to 20 carbon atoms, or an alkyl-substituted phenyl group.

[0136] Fig. 27 shows examples of organic compounds that can be quantified using the near-infrared absorption spectra measured by the optical sensors described in Examples 1 to 4. The structural formula in Fig. 26 is used as a detergent dispersant.

[0137] In the near-infrared region, in principle, it is possible to quantify organic compounds regardless of the presence or absence of a benzene ring, and the method of the examples is applicable to all organic compounds. This is because the wavelengths at which interatomic vibrations such as C-H and O-H occur are determined by the molecular structure. However, compounds containing a benzene ring, particularly those containing the structures of phenol and phenylamine, are easily colored due to oxidative degradation, and RGB color diagnosis as described in Patent Document 1 is possible. For example, diphenylamine-based antioxidants that could be quantified by ΔE and B values in visible light can also be quantified by near-infrared absorption spectra.

[0138] However, visible light sensors for RGB color diagnosis are inexpensive, while near-infrared sensors are expensive. For organic compounds having a benzene ring and capable of color diagnosis, RGB color diagnosis is prioritized, and for those difficult to diagnose by color, the near-infrared absorption spectrum is used. By combining the two according to the application and the additive to be targeted, it is possible to achieve excellent cost performance in quantification.

[0139] As described above, according to this example, by using the near-infrared absorption spectrum by an optical sensor, it is possible to detect the remaining amounts of many types of oil additives such as antioxidants, rust preventives, antifoaming agents, viscosity index improvers, oiliness improvers, extreme pressure additives, detergent dispersants, pour point depressants, and emulsifiers. Therefore, by performing appropriate maintenance such as lubricating oil replacement, it is possible to prevent abnormalities in wind turbines in advance. It is also possible to optimize the lubricating oil replacement cycle. In addition, the viscosity can be measured by a simple method, and if an optical sensor is installed in the nacelle, it is also possible to perform online remote monitoring of the degradation of lubricating oil.

[0140] In this embodiment, a method and a system for monitoring by installing an optical sensor in the lubricating oil of a rotating part are described. However, the lubricating oil inside the rotating part can be sampled during inspection or the like, and measurement can be performed by an optical sensor outside the rotating part, and the same diagnosis can be performed. According to this embodiment, when measuring the change in the absorption spectrum in the near-infrared wavelength range of the oil (measurement target) by an optical sensor and detecting the change in the composition of the oil, it was conventionally difficult to quantify with color measurement or the like. It is possible to accurately determine the state of the oil by quantifying the concentration of an additive that is difficult to change in color even when oxidized and deteriorated. According to the above embodiment, since efficient oil maintenance management can be realized, it consumes less energy, reduces carbon emissions, contributes to preventing global warming, and realizing a sustainable society.

[0141] According to this embodiment, when measuring the change in the absorption spectrum in the near-infrared wavelength range of the oil (measurement target) by an optical sensor and detecting the change in the composition of the oil, it was conventionally difficult to quantify with color measurement or the like. It is possible to accurately determine the state of the oil by quantifying the concentration of an additive that is difficult to change in color even when oxidized and deteriorated.

[0142] According to the above embodiment, since efficient oil maintenance management can be realized, it consumes less energy, reduces carbon emissions, contributes to preventing global warming, and realizing a sustainable society.

Description of Reference Numerals

[0143] Diagnosis system 1200, Estimation model generation unit 1250, Oil property estimation unit 1260, Measured data collection process S1310, Property estimation model generation process S1320, Oil property estimation process S1330

Claims

1. Using an information processing system including an input device, an output device, a processing device, and a storage device, the information processing system includes an estimation model generation unit that generates an estimable model capable of machine learning, inputting into the estimation model generation unit an absorption spectrum obtained by transmitting light of at least a part of wavelengths from 800 nm to 3000 nm through oil containing an additive, and a value reflecting the characteristics of the oil, generating an estimation model using, as an explanatory variable, a value based on the absorption spectrum and, as an objective variable, a value reflecting the characteristics of the oil. A method for generating an estimation model of oil characteristics.

2. The wavelength of the light is in the range of 800 nm to 2500 nm, The method for generating an estimation model of oil characteristics according to Claim 1.

3. The oil contains, as the additive, an organic compound having no aromatic ring, and the value reflecting the characteristics of the oil has a correlation with the amount of the organic compound. The method for generating an estimation model of oil characteristics according to Claim 1.

4. The oil contains, as the additive, an organic compound having a maximum molar absorption coefficient in the ultraviolet-visible region of 250 nm to 800 nm of 50 or less, and the value reflecting the characteristics of the oil has a correlation with the amount of the organic compound. The method for generating an estimation model of oil characteristics according to Claim 1.

5. The explanatory variable is a value obtained by differentiating the absorption spectrum one or more times. The method for generating an estimation model of oil characteristics according to Claim 1.

6. The value reflecting the characteristics of the oil is the concentration of the additive. The method for generating an estimation model of oil characteristics according to Claim 1.

7. The optical path length of the light transmitted through the oil is 1 mm to 20 mm. Method for generating an estimation model of oil properties according to claim 1.

8. The estimation model generation unit performs multivariate analysis using PLS regression analysis. Method for generating an estimation model of oil properties according to claim 1.

9. A diagnostic method for oil properties using a diagnostic system comprising an information processing apparatus implementing the estimation model according to claim 1, inputting into the diagnostic system an absorption spectrum obtained by transmitting light of at least a part of wavelengths from 800 nm to 3000 nm to a diagnostic target oil of the same type as the oil containing the additive, inputting a value based on the absorption spectrum into the estimation model, obtaining a value reflecting the properties of the diagnostic target oil from the estimation model. Diagnostic method for oil properties.

10. Inputting a value obtained by differentiating the absorption spectrum one or more times into the estimation model. Diagnostic method for oil properties according to claim 9.

11. The value reflecting the properties of the oil is the concentration of the additive. Diagnostic method for oil properties according to claim 9.

12. The diagnostic system can utilize data for converting a value reflecting the properties of the oil into at least one of the viscosity, total acid value, and usage time of the oil, outputting at least one of the viscosity, total acid value, and usage time of the oil. Diagnostic method for oil properties according to claim 9.

13. Furthermore, color information obtained by transmitting light of wavelengths in the ultraviolet-visible region in the range of 250 nm to 800 nm to a diagnostic target oil of the same type as the oil containing the additive is input into the diagnostic system, estimating the properties of the oil using the color information and a calibration curve. The method for diagnosing oil characteristics according to claim 9.

14. An oil characteristic diagnosis system comprising an information processing apparatus that implements the estimation model according to claim 1.

15. Using an optical sensor having a light source and a detector that detects light emitted from the light source, Obtaining an absorption spectrum in the range of 800 nm to 3000 nm of an oil containing an additive having a maximum molar absorption coefficient of 50 or less in the ultraviolet-visible wavelength range of 250 nm to 800 nm, A method for diagnosing oil characteristics, characterized by inputting data based on the absorption spectrum into an estimation model implemented in an information processing apparatus to predict the concentration of the additive.

Images