Work machine oil property diagnostic system and work machine oil property diagnostic method

Abstract

In the present invention, an oil property sensor (1 (1a, 1b)) detects, as physical property values of oil, at least the dielectric constant and the resistance value. A controller (50) estimates, on the basis of the relationship between the physical property values of the oil detected by the oil property sensor (1 (1a, 1b)) and the metal concentration or the iron concentration in the oil, the metal concentration or the iron concentration in the oil from the physical property values.

Description

Oil property diagnosis system for working machine and oil property diagnosis method for working machine

[0001] The present disclosure relates to an oil property diagnosis system for a work machine and an oil property diagnosis method for a work machine.

[0002] By detecting and analyzing the properties of the oil used to operate a work machine, it is possible to identify signs of a malfunction in the work machine. For example, in Japanese Patent Application Laid-Open No. 2016-113819 (Patent Document 1), it is determined whether oil analysis, which involves oil sampling, is necessary based on the oil properties detected by a sensor. If it is determined that oil analysis is necessary, oil sampling is carried out as quickly as possible, and the oil is analyzed by an oil analysis company. Then, an oil abnormality is determined based on past oil analysis information, and the cause of the oil abnormality is identified.

[0003] JP 2016-113819 A

[0004] In work machines, there is a demand for constant monitoring of the oil condition to prevent sudden breakdowns due to abnormal wear of internal parts, etc. However, in Patent Document 1, oil analysis by an oil analysis company is required, so the oil condition cannot be constantly monitored.

[0005] An object of the present disclosure is to provide an oil property diagnosis system and an oil property diagnosis method for a work machine that are capable of constantly monitoring the metal concentration and iron concentration in oil.

[0006] An oil property diagnosis system for a work machine according to the present disclosure includes an oil property sensor and a controller. The oil property sensor detects at least the dielectric constant and resistance value as physical property values ​​of the oil. The controller estimates the metal concentration in the oil from the physical property values ​​detected by the oil property sensor based on the relationship between the physical property values ​​of the oil and the metal concentration in the oil.

[0007] Another oil property diagnosis system for a work machine disclosed herein includes an oil property sensor and a controller. The oil property sensor detects at least the dielectric constant and resistance value as physical property values ​​of the oil. The controller estimates the iron concentration in the oil from the physical property values ​​detected by the oil property sensor based on the relationship between the physical property values ​​of the oil and the iron concentration in the oil.

[0008] One oil property diagnosis method for a work machine disclosed herein includes the following steps: acquiring at least the permittivity and resistivity as physical property values ​​of the oil; and estimating the metal concentration in the oil from the acquired physical property values ​​of the oil based on the relationship between the physical property values ​​of the oil and the metal concentration in the oil.

[0009] Another oil property diagnosis method for a work machine disclosed herein includes the following steps: At least the permittivity and resistivity are acquired as physical property values ​​of the oil, and the iron concentration in the oil is estimated from the acquired physical property values ​​based on the relationship between the physical property values ​​of the oil and the iron concentration in the oil.

[0010] According to the present disclosure, it is possible to realize an oil property diagnosis system and an oil property diagnosis method for a work machine that are capable of constantly monitoring the metal concentration and iron concentration in oil.

[0011] 4 is a diagram showing the configuration of a work machine according to an embodiment of the present disclosure. FIG. 5 is a diagram showing an example of an oil circuit that supplies and discharges hydraulic oil to a hydraulic actuator (e.g., a hydraulic cylinder). FIG. 6 is a diagram showing another example of an oil circuit that supplies lubricating oil to a drive unit (e.g., an engine). FIG. 7 is a diagram showing the configuration of an oil property diagnosis system for a work machine according to an embodiment of the present disclosure. FIG. 8 is a functional block diagram of a controller employed in the system of FIG. 4. FIG. 9 is a flow diagram showing an oil property diagnosis method for a work machine according to an embodiment of the present disclosure. FIG. 10 is a diagram for explaining the coefficient of determination and 80% error, which are evaluation indexes for the accuracy of an estimation model (arithmetic formula). FIG. 11 is a diagram for explaining the correlation coefficient with soot concentration, which is an evaluation index for the accuracy of an estimation model (arithmetic formula). FIG. 12 is a diagram showing the relationship between oil physical property values ​​and an evaluation index for iron concentration in the oil. FIG. 13 is a diagram showing the relationship between oil physical property values ​​and an evaluation index for metal (iron + copper + chromium + aluminum + silicon + lead) concentrations in the oil.

[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the drawings. In the specification and drawings, identical or corresponding components are designated by the same reference numerals, and redundant descriptions will not be repeated. In addition, in the drawings, configurations may be omitted or simplified for the sake of convenience. Furthermore, at least some of the embodiments and modified examples may be combined with each other in any desired manner.

[0013] In the following description, the fore-aft direction refers to the direction in which the boom 16 extends from the base end to the tip end in a plan view. The left-right direction refers to the direction perpendicular to the fore-aft direction in a plan view. The up-down direction refers to the direction perpendicular to a plane that includes the fore-aft direction and the left-right direction, which are perpendicular to each other.

[0014] <Configuration of work machine 10> Figure 1 is a diagram that shows a schematic configuration of a work machine in one embodiment of the present disclosure. As shown in Figure 1, the work machine 10 in this embodiment is, for example, a hydraulic excavator. However, the work machine 10 is not limited to hydraulic excavators, and may be any work machine 10 that has a work implement and uses oil during operation, such as a wheel loader, bulldozer, motor grader, dump truck, etc.

[0015] The oil used to operate the work machine 10 is hydraulic oil used to operate hydraulic actuators (for example, hydraulic cylinders, hydraulic motors, etc.) and lubricating oil used to lubricate lubricated parts (for example, engines). Lubricated parts are, for example, lubricated parts of drive parts, and drive parts include drive sources such as engines.

[0016] The oil used to operate the work machine 10 includes, for example, engine oil, hydraulic oil, swing machine oil, final drive oil, axle oil, transmission oil, etc. The oil to be collected varies depending on the model of the work machine 10.

[0017] The hydraulic excavator 10, which is an example of a work machine 10, has a main body 11 and a hydraulically operated work implement 12. The main body 11 has a revolving body 13 and a traveling body 15.

[0018] The running body 15 has a pair of crawler tracks 15Cr and a traveling motor 15M. The hydraulic excavator 10 can travel by rotating the crawler tracks 15Cr. The traveling motor 15M is provided as a drive source for the running body 15.

[0019] The rotating body 13 is disposed on and supported by the running body 15. The rotating body 13 can be rotated relative to the running body 15 about a rotation axis RX by a rotation motor (not shown). The rotation axis RX is an imaginary straight line that serves as the rotation center of the rotating body 13.

[0020] The rotating body 13 has a driver's cab 14. A driver's seat 14S for an operator is provided in the driver's cab 14. The operator sits in the driver's cab 14 and can operate the work implement 12, rotate the rotating body 13 relative to the traveling body 15, and travel the hydraulic excavator 10 using the traveling body 15.

[0021] The revolving body 13 has a drive source (for example, an engine) disposed behind the operator's cab 14. The drive source is covered by an exterior panel 13a.

[0022] The work implement 12 is supported by the revolving unit 13. The work implement 12 has a boom 16, an arm 17, and a bucket 18. The work implement 12 further has a boom cylinder 19a, an arm cylinder 19b, and a bucket cylinder 19c.

[0023] The boom 16 is rotatably connected to the main body 11. Specifically, the base end of the boom 16 is rotatably connected to the revolving unit 13 with a boom foot pin BF as a fulcrum. The base end of the boom 16 is disposed to the left and right of the operator's cab 14. The arm 17 is rotatably connected to the boom 16. Specifically, the base end of the arm 17 is rotatably connected to the tip of the boom 16 with a boom top pin BT as a fulcrum. The bucket 18 is rotatably connected to the arm 17. Specifically, the base end of the bucket 18 is rotatably connected to the tip of the arm 17 with an arm top pin AT as a fulcrum.

[0024] The boom 16 can be driven by a boom cylinder 19a relative to the main body 11. By this driving, the boom 16 can be rotated in the vertical direction relative to the revolving body 13 with a boom foot pin BF as a fulcrum.

[0025] The arm 17 can be driven by an arm cylinder 19b relative to the boom 16. By this drive, the arm 17 can rotate up and down or back and forth relative to the boom 16 with the boom top pin BT as a fulcrum.

[0026] The bucket 18 can be driven by a bucket cylinder 19c relative to the arm 17. By this drive, the bucket 18 can rotate up and down or back and forth relative to the arm 17 with the arm top pin AT as a fulcrum.

[0027] The boom cylinder 19a, arm cylinder 19b, bucket cylinder 19c, swing motor, etc. are hydraulic actuators that operate using hydraulic oil, for example. Furthermore, drive units such as a drive source (e.g., an engine) are configured to be lubricated with lubricating oil. In this embodiment, the physical properties of the oil supplied to the hydraulic actuators or drive units are detected. The detected physical properties are at least the dielectric constant and resistivity of the oil.

[0028] Furthermore, in this embodiment, as will be described later, at least one of the metal concentration and iron concentration in the oil is estimated with high accuracy from the detected oil property values. The estimated metal concentration or iron concentration in the oil makes it possible to determine the degree of accelerated wear of the main components of the work machine 10. For example, if engine oil monitoring determines that each of the metal concentration and iron concentration in the oil is equal to or greater than the abnormality determination value, it is estimated that the piston rings, cylinder liner, etc. are worn.

[0029] In addition, in this embodiment, at least one of the metal concentration and iron concentration in the oil is estimated from the detected physical property values ​​of the oil, making it possible to constantly monitor the metal concentration and iron concentration in the oil.

[0030] Furthermore, in this embodiment, if it is determined that the metal concentration and iron concentration in the oil are each equal to or greater than the abnormality determination value, a warning is issued to those in the vicinity by means of a screen display, sound, etc. This prevents sudden breakdowns due to abnormal wear of parts of the work machine 10, etc.

[0031] <Oil Circuit> Next, an example of an oil circuit for supplying and discharging hydraulic oil to and from a hydraulic cylinder serving as a hydraulic actuator, and an example of the arrangement of an oil property sensor will be described with reference to FIG.

[0032] 2 is a diagram showing an example of an oil circuit that supplies and discharges hydraulic oil to and from a hydraulic actuator (e.g., a hydraulic cylinder). As shown in Fig. 2, the example of the oil circuit includes a drive source 2, a main pump 3a, a main valve 4, a hydraulic actuator 5, an oil cooler 6a, an oil filter 7a, and an oil tank 8.

[0033] The drive source 2 is, for example, an engine. The main pump 3a is operated by the drive force from the drive source 2. When the main pump 3a operates, it pumps up hydraulic oil (oil) stored in an oil tank 8. The hydraulic oil pumped up by the main pump 3a is supplied to the hydraulic actuator 5 through the main valve 4. The hydraulic oil supplied to the hydraulic actuator 5 is discharged through the main valve 4.

[0034] The hydraulic actuator 5 is, for example, a hydraulic cylinder, but may also be another actuator such as a hydraulic motor. The hydraulic actuator 5 operates by supplying or discharging hydraulic oil. If the hydraulic actuator 5 is, for example, a hydraulic cylinder, the hydraulic cylinder 5 expands or contracts by supplying or discharging hydraulic oil.

[0035] The main valve 4 controls the amount of hydraulic oil supplied to and discharged from the hydraulic actuator 5. The hydraulic oil discharged from the hydraulic actuator 5 is cooled in an oil cooler 6a, filtered by an oil filter 7a, and then returned to an oil tank 8.

[0036] One example of the oil circuit further includes an oil property sensor 1a. The oil property sensor detects the properties of the oil passing through the oil passage. Examples of oil properties include temperature, viscosity, density, dielectric constant, and resistance value. These oil properties may be detected by multiple oil property sensors, or by a single oil property sensor. The oil property sensor 1a is disposed, for example, in the oil passage between the main pump 3a and the oil tank 8. The oil property sensor 1a may also be disposed in another oil passage of the oil circuit.

[0037] Next, another example of an oil circuit for supplying lubricating oil to an engine as a driving unit and an example of the arrangement of an oil property sensor will be described with reference to FIG.

[0038] 3 is a diagram showing another example of an oil circuit that supplies lubricating oil to a drive unit (e.g., an engine). As shown in FIG. 3, the other example of the oil circuit includes a lubrication pump 3b, an oil cooler 6b, an oil filter 7b, an oil tank 8, and a lubrication unit 9 for the drive unit.

[0039] The lubrication pump 3b is operated by a driving force from a drive source (not shown). When the lubrication pump 3b operates, it pumps up the lubrication oil (oil) stored in the oil tank 8. The lubrication oil pumped up by the lubrication pump 3b is cooled in the oil cooler 6b, filtered in the oil filter 7b, and then supplied to the lubrication unit 9 of the drive unit. After lubricating the lubrication unit 9 of the drive unit, the lubrication oil is returned to the oil tank 8.

[0040] Another example of the oil circuit further includes an oil property sensor 1b. The oil property sensor 1b is disposed, for example, in an oil passage connecting the oil passage between the oil filter 7b and the lubrication unit 9 and the oil tank 8. The oil property sensor 1b may also be disposed in another oil passage of the oil circuit.

[0041] 2 and 3, each of the oil property sensors 1a and 1b detects at least the dielectric constant and resistance value of the oil as physical property values ​​of the oil. In addition to the dielectric constant and resistance value of the oil, each of the oil property sensors 1a and 1b may also detect the viscosity and density of the oil.

[0042] Each of the oil property sensors 1a and 1b is electrically connected to the controller 50. Each of the oil property sensors 1a and 1b outputs a signal representing the detected physical property value of the oil to the controller 50.

[0043] In this embodiment, the metal and iron concentrations in the oil can be estimated with high accuracy by estimating the metal and iron concentrations in the oil from at least the dielectric constant and resistance value of the oil detected by each of the oil property sensors 1a and 1b.

[0044] <Oil Property Diagnostic System for Work Machine 10> Next, an oil property diagnostic system for the work machine 10 according to one embodiment of the present disclosure will be described with reference to FIG. 4 .

[0045] Figure 4 is a diagram showing the configuration of an oil property diagnosis system for a work machine 10 according to one embodiment of the present disclosure. As shown in Figure 4, the oil property diagnosis system for a work machine 10 includes, for example, the work machine 10, a server 70, and a service computer 80.

[0046] The work machine 10 has an oil property sensor 1, a controller 50, and a monitor 60. The oil property sensor 1 is either one of the oil property sensors 1a and 1b described above, or both of the oil property sensors 1a and 1b. The oil property sensor 1 and the monitor 60 are electrically connected to the controller 50.

[0047] A signal indicating the oil property values ​​detected by the oil property sensor 1 is output to the controller 50 of the work machine 10. The controller 50 estimates at least one of the metal concentration and the iron concentration in the oil based on the acquired oil property values. The controller 50 determines whether each of the estimated metal concentration and iron concentration is equal to or greater than an abnormality determination value.

[0048] The controller 50 outputs a signal indicating the determination result to the monitor 60. The monitor 60 displays an image based on the determination result obtained from the controller 50. The monitor 60 displays and issues a warning when at least one of the estimated metal concentration and iron concentration is equal to or greater than the abnormality determination value. A speaker may be used instead of the monitor 60. In this case, the warning may be issued by sound.

[0049] A signal indicating the oil property values ​​detected by the oil property sensor 1 is output to a server 70 provided outside the work machine 10. The signal indicating the oil property values ​​detected by the oil property sensor 1 may be transmitted wirelessly to the server 70. The server 70 is, for example, a server owned by a construction machine manufacturer.

[0050] The server 70 has the same functions as the controller 50. That is, the server 70 estimates at least one of the metal concentration and the iron concentration in the oil based on the acquired oil property values, and determines whether the estimated metal concentration and / or the iron concentration is abnormal. The server 70 outputs a signal indicating the determination result to the service computer 80.

[0051] The server 70 may acquire the determination result from the controller 50 of the work machine 10. In this case, the server 70 may output the acquired determination result to the service computer 80 as is.

[0052] The service computer 80 is, for example, a computer owned by an agency of a construction machine manufacturer, etc. The server 70 and the service computer 80 are connected via, for example, an intranet or the Internet.

[0053] The service computer 80 may wirelessly transmit a signal indicating the determination result acquired from the server 70 to the work machine 10. Furthermore, an agent of the construction machine manufacturer or the like who has confirmed the determination result acquired by the service computer 80 may repair or inspect the work machine 10 based on the determination result.

[0054] Furthermore, the user who owns the work machine 10 may request repairs and inspections from an agency of the construction machine manufacturer or the like based on the results displayed on the monitor 60. In this case, the agency of the construction machine manufacturer or the like may repair and inspect the work machine 10 based on the request.

[0055] <Functional Blocks of Controller 50> Next, functional blocks of the controller 50 will be described with reference to FIG.

[0056] Fig. 5 is a functional block diagram of a controller employed in the system of Fig. 4. As shown in Fig. 5, the controller 50 has an oil physical property value acquisition unit 51, a concentration estimation unit 52, a concentration determination unit 53, an output control unit 54, and a memory 55. The oil physical property value acquisition unit 51 acquires the physical property values ​​of the oil from the oil property sensor 1.

[0057] The concentration estimation unit 52 estimates at least one of the metal concentration and the iron concentration in the oil from the acquired physical property values ​​of the oil based on the relationship between the physical property values ​​of the oil and the metal concentration or iron concentration in the oil (physical property-concentration relationship). When making the above estimation, the concentration estimation unit 52 refers to the physical property-concentration relationship stored in the memory 55. The method of estimating the iron concentration and the metal concentration will be described below in <Method of Estimating Iron Concentration and Metal Concentration>.

[0058] The concentration determination unit 53 determines whether or not at least one of the metal concentration and iron concentration in the oil is equal to or greater than an abnormality determination value. When making the determination, the concentration determination unit 53 refers to the abnormality determination value for iron concentration or the abnormality determination value for metal concentration stored in the memory 55.

[0059] The output control unit 54 acquires the determination result of the concentration determination unit 53. The output control unit 54 outputs a control signal to the monitor 60 based on the determination result of the concentration determination unit 53 to control the display content of the monitor 60. When the output control unit 54 determines that at least one of the metal concentration and the iron concentration in the oil is equal to or greater than the abnormality determination value, the output control unit 54 issues a warning by displaying that fact on the monitor 60. Furthermore, when the metal concentration or the iron concentration in the oil is less than the abnormality determination value, the output control unit 54 displays on the monitor 60 that there is no abnormality.

[0060] The physical property value-concentration relationship and abnormality determination value may be stored in advance (at the time of shipment of the work machine 10) in memory 55. Furthermore, the physical property value-concentration relationship and abnormality determination value may be stored in memory 55 from outside the work machine 10 after shipment of the work machine 10. The physical property value-concentration relationship and abnormality determination value may be stored in memory 55 by operating an input device such as a touch panel mounted on the work machine 10. Furthermore, the device that issues the warning is not limited to the monitor 60, and may be a device that issues the warning by sound, such as a speaker.

[0061] The controller 50 includes a processor, a main memory, and a storage. The processor is, for example, a central processing unit (CPU). The main memory includes, for example, a nonvolatile memory such as a read-only memory (ROM) and a volatile memory such as a random access memory (RAM).

[0062] The controller 50 may be mounted on the work machine 10, or may be located remotely outside the work machine 10. If the controller 50 is located remotely outside the work machine 10, the controller 50 may be wirelessly connected to the oil property sensor 1, monitor 60, etc. The controller 50 may be stored in a server 70 located remotely from the work machine 10.

[0063] The controller 50 reads the program stored in the storage, loads it into the main memory, and executes predetermined processing in accordance with the program. The program may also be distributed to the controller 50 via a network.

[0064] <Method of Estimating Iron Concentration and Metal Concentration> Next, a method of estimating the iron concentration of the iron concentration and the metal concentration in the oil in the concentration estimating unit 52 in FIG. 5 will be described using the iron concentration as an example.

[0065] The iron concentration in oil is estimated by machine learning. In machine learning, a set of measured values ​​of the oil's physical properties (dielectric constant, viscosity, density, and resistivity) and a measured value of the iron concentration in the oil corresponding to the measured values ​​of the physical properties are used as training data. The measured values ​​of the oil's physical properties are detected by an oil property sensor 1. The oil property sensor 1 can detect, for example, four physical properties of oil: viscosity, density, dielectric constant, and resistivity. The measured value of the iron concentration is detected, for example, by an ICP (Inductively Coupled Plasma) analyzer. The measured values ​​of the oil's physical properties and the measured value of the iron concentration in the oil corresponding to the measured values ​​of the physical properties are each detected from used oil.

[0066] A large amount of the training data is input to a machine learning algorithm. The machine learning algorithm receives the large amount of training data as input and constructs an estimation model. The estimation model is created as an arithmetic formula that calculates the iron concentration using the least squares method from the relationship between physical property values ​​and iron concentration.

[0067] For example, when two variables, dielectric constant and resistance, are used as the physical properties of oil, the calculation formula is expressed as follows: iron concentration [ppm] = a (dielectric constant ε) + b (resistance value R) + c. In this calculation formula, the coefficients a, b, and c are determined by a machine learning algorithm. The above calculation formula shows the relationship between the physical properties of oil and the iron concentration in oil.

[0068] The iron concentration in the oil is calculated by inputting the physical property values ​​of the oil detected by the oil property sensor 1 shown in Figure 5 into the calculation formula created as described above. In this way, the iron concentration in the oil is estimated.

[0069] The metal concentration in oil is estimated using a similar method to the iron concentration. The metal concentration in oil is estimated as the total concentration of one or more metals selected from the group consisting of iron (Fe), copper (Cu), chromium (Cr), aluminum (Al), silicon (Si), and lead (Pb). For example, when two variables, dielectric constant and resistance, are used as the oil's physical properties, the equation showing the relationship between the oil's physical property values ​​and the metal concentration in the oil is expressed as follows: Metal Concentration [ppm] = a1 (Dielectric Constant ε) + b1 (Resistance Value R) + c1. In this equation, the coefficients a1, b1, and c1 are determined by a machine learning algorithm.

[0070] <Oil property diagnosis method for a work machine> Next, an oil property diagnosis method for a work machine according to one embodiment of the present disclosure will be described with reference to Figs. 5 and 6 .

[0071] Fig. 6 is a flow diagram showing an oil property diagnosis method for a work machine according to one embodiment of the present disclosure. As shown in Figs. 5 and 6, the oil property sensor 1 detects the physical property values ​​of the oil. The oil physical property value acquisition unit 51 of the controller 50 acquires the physical property values ​​of the oil from the oil property sensor 1 (step S1: Fig. 6). As described above, the physical property values ​​of the oil include at least the dielectric constant and resistivity of the oil, and may also include the viscosity and density of the oil in addition to the dielectric constant and resistivity of the oil.

[0072] The oil physical property value acquisition unit 51 outputs the acquired physical property values ​​to the concentration estimation unit 52. The concentration estimation unit 52 estimates at least one of the metal concentration and the iron concentration in the oil from the acquired physical property values ​​of the oil based on the physical property value-concentration relationship (step S2: FIG. 6). When estimating at least one of the metal concentration and the iron concentration in the oil, the concentration estimation unit 52 refers to the physical property value-concentration relationship stored in the memory 55.

[0073] The physical property value-concentration relationship is, for example, an arithmetic expression obtained by the machine learning. The concentration estimation unit 52 inputs the oil physical property values ​​acquired from the oil physical property value acquisition unit 51 into the arithmetic expression obtained by the machine learning, thereby estimating at least one of the metal concentration and the iron concentration in the oil.

[0074] The concentration estimation unit 52 outputs the estimated metal concentration and / or iron concentration in the oil to the concentration determination unit 53. The concentration determination unit 53 determines whether or not the acquired metal concentration and / or iron concentration in the oil is equal to or greater than an abnormality determination value (step S3: FIG. 6 ).

[0075] The abnormality determination values ​​for each of the metal concentration and the iron concentration are stored in advance in the memory 55 of the controller 50. When making the above determination, the concentration determination unit 53 refers to the abnormality determination value for the metal concentration or the abnormality determination value for the iron concentration stored in the memory 55.

[0076] If the concentration determination unit 53 determines that the metal concentration and iron concentration in the oil are less than the abnormality determination value, steps S1, S2, and S3 are repeated. Note that even if the concentration determination unit 53 determines that the metal concentration and iron concentration in the oil are less than the abnormality determination value, the concentration determination unit 53 may output the determination result to the monitor 60, and steps S1, S2, and S3 may be repeated. On the other hand, if the concentration determination unit 53 determines that at least one of the metal concentration and the iron concentration in the oil is equal to or greater than the abnormality determination value, the concentration determination unit 53 outputs a signal indicating the determination result to the output control unit 54.

[0077] The output control unit 54 outputs a control signal based on the acquired determination result to the monitor 60. If the output control unit 54 determines that at least one of the metal concentration and the iron concentration in the oil is equal to or greater than the abnormality determination value, it issues a warning by displaying that fact on the monitor 60 (step S4: FIG. 6).

[0078] The oil property diagnosis method for a work machine according to this embodiment is carried out as described above. <Examples> Next, the inventors of the present invention will explain the investigation into the relationship between the physical property values ​​of oil and the estimation accuracy with reference to FIGS.

[0079] Fig. 7 is a diagram illustrating the coefficient of determination and 80% error, which are evaluation indices of the accuracy of the estimation model (calculation formula). Fig. 8 is a diagram illustrating the correlation coefficient with soot concentration, which is an evaluation indices of the accuracy of the estimation model (calculation formula). Fig. 9 is a diagram illustrating the relationship between the physical property values ​​of oil and the evaluation indices for the iron concentration in oil. Fig. 10 is a diagram illustrating the relationship between the physical property values ​​of oil and the evaluation indices for the metal (iron + copper + chromium + aluminum + silicon + lead) concentrations in oil.

[0080] The inventors used an oil property sensor to detect the physical properties (dielectric constant, viscosity, density, and resistivity) of used engine oil (oil collected from the market) with different iron concentrations. They then varied the combinations of the detected physical properties and estimated the iron concentration in the oil for each combination. Specifically, they estimated the iron concentration in the oil from only one of the four physical properties (one variable), a combination of two properties (two variables), and a combination of three properties (three variables).

[0081] The estimated iron concentration thus obtained was compared with the actual iron concentration in the oil measured using an ICP analyzer to evaluate the estimation accuracy for each combination of physical properties. The evaluation indices used in this evaluation were (1) the coefficient of determination, (2) the 80% error, and (3) the correlation coefficient with soot concentration. Each evaluation indices is explained below.

[0082] (1) Coefficient of determination (= R 2 ) Coefficient of determination R 2 The coefficient of determination R is calculated by the following formula (1), where yi is the true value (actually measured data) of the i-th data set N(n), yei is the value estimated by the regression equation, and yai is the average value of all the actually measured data. 2 The closer the value is to 1, the higher the accuracy of the iron concentration estimation. In other words, as shown in FIG. 7, the closer the data is to the line SL1, which indicates a one-to-one relationship between the actual measured value of iron concentration and the estimated value of iron concentration, the higher the accuracy of the estimation. 2 is evaluated in absolute value.

[0083]

[0084] (2) 80% Error (= RMSE × 1.25) The RMSE (Root Mean Squared Error) calculated by the following formula (2) represents the magnitude of the discrepancy between the measured and estimated data values. Similar to the standard deviation, it is considered that approximately 68% of the data falls within the RMSE error range. Therefore, since it is considered that approximately 80% of the data falls within the range of (RMSE × 1.25), (RMSE × 1.25) was defined as the 80% error and used as one of the evaluation indices. As shown in FIG. 7 , the 80% error refers to the size of the error range ER within which 80% of the data showing the relationship between the measured and estimated iron concentrations falls, relative to the line SL1. The smaller the 80% error, the higher the estimation accuracy.

[0085]

[0086] (3) Correlation coefficient (= r) between measured soot concentration value x and estimated iron concentration value y Let the measured soot concentration value and estimated iron concentration value be (x, y), their respective average values ​​be (xa, ya), and the i-th data out of n data points be (xi, yi). Also, let sxy be the covariance of x and y, sx be the standard deviation of x, and sy be the standard deviation of y. The correlation coefficient r in this case is expressed by the following equation (3):

[0087]

[0088] The correlation coefficient r is a calculation of the degree of correlation between the estimated iron concentration obtained from an oil sample using the above method and the measured soot concentration for the same oil sample. The closer the correlation coefficient r is to 1, the more difficult it is to distinguish the estimated iron concentration from the measured soot concentration, meaning that the accuracy of the iron concentration estimation is lower. In other words, as shown in Figure 8, the closer the data is to the line SL2, which shows a one-to-one relationship between the measured soot concentration and the estimated iron concentration, the lower the estimation accuracy. The correlation coefficient r is evaluated as an absolute value.

[0089] Soot is a mixture of soot and oil sludge contained in exhaust gas generated by fuel combustion, and is contaminated, for example, in engine oil. Soot concentration is detected from the intensity of transmitted light obtained by infrared spectroscopy and serves as an indicator of contamination in engine oil, for example. (4) Evaluation Results of Estimated Iron Concentration in Oil. Figure 9 shows the evaluation results of iron concentration in oil using the above evaluation index. As shown in Figure 9, samples (1) to (4) are evaluated using one variable, samples (5) to (10) are evaluated using two variables, and samples (11) to (14) are evaluated using three variables. Sample (2) was evaluated using only one variable, viscosity, and sample (3) was evaluated using only one variable, density. When iron concentration was estimated using at least the oil's dielectric constant and resistivity as physical properties, as in samples (7), (12), and (13), the coefficient of determination was 0.3 or greater, the 80% error was less than 27, and the correlation coefficient was less than 0.6. Thus, it was found that samples (7), (12), and (13) had higher accuracy in estimating iron concentration than the other samples (1) to (6), (8) to (11), and (14).

[0090] From this, it was found that by estimating the iron concentration using at least the dielectric constant and resistance value of the oil, the accuracy of the iron concentration estimation is higher than when the iron concentration is estimated using physical property values ​​that do not include the combination of the dielectric constant and resistance value of the oil.

[0091] Furthermore, it was found that the correlation coefficients for samples (7) and (12) were 0.4 or less, indicating that the estimation accuracy of the iron concentration was higher than that of the other samples.

[0092] Furthermore, when three variables (dielectric constant, resistivity, and viscosity) of the oil were used as the oil's physical properties, as in sample (12), the coefficient of determination approached 1 and the 80% error was smaller than when two variables (dielectric constant and resistivity) of the oil were used, indicating a higher accuracy in estimating the iron concentration. (5) Evaluation Results of Estimated Values ​​of Metal Concentrations in Oil The concentrations of metals (iron + copper + chromium + aluminum + silicon + lead) in the oil were estimated using the same method as for the iron concentration, and the accuracy of the estimated values ​​was evaluated. The evaluation results are shown in Figure 10. As shown in Figure 10, when the oil's dielectric constant and resistivity were used as the oil's physical properties, as in sample (26), the coefficient of determination was 0.07 or higher, the 80% error was less than 95, and the correlation coefficient was less than 0.88, indicating a higher accuracy in estimating the metal concentration than the other samples (21) to (25) and (27).

[0093] From this, it was found that by estimating the metal concentration using at least the dielectric constant and resistance value of the oil, the accuracy of estimating the metal concentration is higher than when estimating the metal concentration using physical property values ​​that do not include the combination of the dielectric constant and resistance value of the oil.

[0094] The oil analyzed in this example contained at least iron among iron, copper, chromium, aluminum, silicon, and lead.

[0095] <Effects> The effects of this embodiment will be described below.

[0096] In a work machine, if a sudden failure occurs due to abnormal wear of an internal part, etc., a delay in detecting the failure can lead to malfunction of the internal part and a shutdown of the work machine. This malfunction and shutdown can lead to serious damage to the customer, such as delays in work plans due to downtime. Therefore, there is a demand for constant monitoring of the concentration of metals, including iron, in the oil in order to detect sudden failures due to abnormal wear of internal parts, etc., as early as possible and prevent downtime. However, in Patent Document 1, when it is determined that oil analysis is necessary based on the oil properties detected by the sensor, oil analysis is performed on oil samples taken from the work machine using precision analytical equipment owned by an oil analysis company, etc., to identify the cause of the oil abnormality, so it is not possible to constantly monitor changes in the concentration of metals, including iron, in the oil.

[0097] 9 and 10, at least the dielectric constant and resistance value are detected as the oil's physical property values, and the metal and iron concentrations in the oil are estimated from the detected oil's physical property values. This makes it possible to estimate the metal and iron concentrations in the oil with higher accuracy than when the metal and iron concentrations in the oil are estimated using oil's physical property values ​​that do not include the combination of the oil's dielectric constant and resistance value.

[0098] Furthermore, in estimating both the metal concentration and the iron concentration, the dielectric constant and resistance value of the oil can be detected with the simple oil property sensor 1. Therefore, there is no need for a large-scale device such as an ICP analyzer to detect the metal concentration or iron concentration in the oil.

[0099] Furthermore, unlike an ICP analyzer or the like, the simple oil property sensor 1 can be easily mounted on the work machine 10. By mounting the simple oil property sensor 1 on the work machine 10, it becomes possible to constantly monitor the metal and iron concentrations in the oil using the simple oil property sensor 1 mounted on the work machine 10. Furthermore, it becomes possible to detect suddenly occurring failures (failures due to wear of internal parts, etc.) at times other than when oil is being removed at an early stage, enabling timely response (oil change, part replacement, etc.) before the work machine 10 is severely damaged.

[0100] Furthermore, according to this embodiment, the metal concentration is the total concentration of one or more metals selected from the group consisting of iron, copper, chromium, aluminum, silicon, and lead, which makes it possible to accurately estimate the total concentration of the above metals contained in the oil.

[0101] According to this embodiment, the metal concentration is the iron concentration, which makes it possible to accurately estimate the iron concentration contained in the oil. The metal concentration includes at least the iron concentration.

[0102] Furthermore, according to this embodiment, the relationship between the physical property values ​​of the oil and the metal concentration or iron concentration in the oil is stored in advance in the controller 50. This makes it possible to estimate the metal concentration or iron concentration in the oil based on the physical property values ​​of the oil.

[0103] Furthermore, according to this embodiment, an abnormality determination value for the metal concentration or iron concentration in the oil that is determined for each work machine 10 is stored in advance in the controller 50, and an oil abnormality is determined based on this abnormality determination value and the estimated metal concentration or iron concentration. By detecting an oil abnormality in this way, it becomes possible to prevent sudden breakdowns.

[0104] Furthermore, according to this embodiment, the oil that is analyzed is the hydraulic oil or engine oil of the work machine 10. This makes it possible to prevent sudden failures in the engine or hydraulic equipment such as the hydraulic actuator, main pump, and main valve.

[0105] <Additional Notes> The above-described embodiment includes the following technical ideas.

[0106] (Supplementary Note 1) An oil property diagnosis system for a work machine, comprising: an oil property sensor that detects at least the dielectric constant and resistance value as physical property values ​​of the oil; and a controller that estimates the metal concentration in the oil from the physical property values ​​of the oil detected by the oil property sensor, based on the relationship between the physical property values ​​of the oil and the metal concentration in the oil.

[0107] (Supplementary Note 2) The oil property diagnosis system for a work machine according to Supplementary Note 1, wherein the metal concentration is a total concentration of one or more metals selected from the group consisting of iron, copper, chromium, aluminum, silicon, and lead.

[0108] (Supplementary Note 3) The oil property diagnosis system for a work machine according to Supplementary Note 1 or Supplementary Note 2, wherein the metal concentration is an iron concentration.

[0109] (Supplementary Note 4) The oil property diagnosis system for a work machine according to any one of Supplementary Note 1 to Supplementary Note 3, wherein the controller pre-stores a relationship between the physical property values ​​of the oil and the metal concentration in the oil.

[0110] (Supplementary Note 5) The oil property diagnosis system for a work machine according to any one of Supplementary Note 1 to Supplementary Note 4, wherein the controller stores in advance an abnormality determination value for the metal concentration in the oil that is determined for each of the work machines, and the controller determines an abnormality in the oil based on the estimated metal concentration and the abnormality determination value.

[0111] (Supplementary Note 6) The oil property diagnosis system for a work machine according to any one of Supplementary Note 1 to Supplementary Note 5, wherein the oil is hydraulic oil or engine oil of the work machine.

[0112] (Supplementary Note 7) An oil property diagnosis system for a work machine, comprising: an oil property sensor that detects at least the dielectric constant and resistance value as physical property values ​​of the oil; and a controller that estimates the iron concentration in the oil from the physical property values ​​of the oil detected by the oil property sensor, based on the relationship between the physical property values ​​of the oil and the iron concentration in the oil.

[0113] (Supplementary Note 8) An oil property diagnosis method for a work machine, comprising: a step of detecting at least a dielectric constant and a resistance value as physical property values ​​of the oil; and a step of estimating a metal concentration in the oil from the obtained physical property values ​​of the oil based on a relationship between the physical property values ​​of the oil and the metal concentration in the oil.

[0114] (Supplementary Note 9) An oil property diagnosis method for a work machine, comprising: a step of detecting a dielectric constant and a resistance value as physical property values ​​of the oil; and a step of estimating the iron concentration in the oil from the obtained physical property values ​​of the oil based on the relationship between the physical property values ​​of the oil and the iron concentration in the oil.

[0115] The embodiments disclosed herein should be considered to be illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims, not by the above description, and is intended to include all modifications within the meaning and scope of the claims.

[0116] DESCRIPTION OF SYMBOLS 1, 1a, 1b Oil property sensor, 2 Drive source, 3a Main pump, 3b Lubrication pump, 4 Main valve, 5 Hydraulic actuator, 6a, 6b Oil cooler, 7a, 7b Oil filter, 8 Oil tank, 9 Lubrication unit, 10 Work machine, 11 Main body, 12 Work equipment, 13 Swing body, 13a Exterior panel, 14 Operator's cab, 14S Operator's seat, 15 Travel body, 15Cr Track, 15M Travel motor, 16 Boom, 17 Arm, 18 Bucket, 19a Boom cylinder, 19b Arm cylinder, 19c Bucket cylinder, 50 Controller, 51 Oil property value acquisition unit, 52 Concentration estimation unit, 53 Concentration determination unit, 54 Output control unit, 55 Memory, 60 Monitor, 70 Server, 80 Service computer, AT Arm top pin, BF Boom foot pin, BT Boom top pin, RX rotation axis.

Claims

1. An oil property diagnosis system for a work machine comprising: an oil property sensor that detects at least the dielectric constant and resistance value as physical property values ​​of the oil; and a controller that estimates the metal concentration in the oil from the physical property values ​​of the oil detected by the oil property sensor, based on the relationship between the physical property values ​​of the oil and the metal concentration in the oil.

2. An oil property diagnostic system for a working machine as set forth in claim 1, wherein the metal concentration is a total concentration of one or more metals selected from the group consisting of iron, copper, chromium, aluminum, silicon and lead.

3. The oil property diagnosis system for a work machine according to claim 2, wherein the metal concentration is an iron concentration.

4. An oil property diagnosis system for a work machine as set forth in claim 1, wherein the controller pre-stores a relationship between the physical property values ​​of the oil and the metal concentration in the oil.

5. An oil property diagnosis system for a work machine as described in claim 1, wherein the controller pre-stores an abnormality determination value for the metal concentration in the oil that is determined for each work machine, and the controller determines whether the oil is abnormal based on the estimated metal concentration and the abnormality determination value.

6. The oil property diagnosis system for a work machine according to claim 1, wherein the oil is hydraulic oil or engine oil of the work machine.

7. An oil property diagnosis system for a work machine comprising: an oil property sensor that detects at least the dielectric constant and resistance value as physical property values ​​of the oil; and a controller that estimates the iron concentration in the oil from the physical property values ​​of the oil detected by the oil property sensor based on the relationship between the physical property values ​​of the oil and the iron concentration in the oil.

8. A method for diagnosing oil properties for a working machine, comprising the steps of: acquiring at least dielectric constant and resistance as physical property values ​​of the oil; and estimating a metal concentration in the oil from the acquired physical property values ​​of the oil based on a relationship between the physical property values ​​of the oil and the metal concentration in the oil.

9. A method for diagnosing oil properties for a working machine, comprising the steps of: acquiring at least dielectric constant and resistance as physical property values ​​of the oil; and estimating the iron concentration in the oil from the acquired physical property values ​​of the oil based on the relationship between the physical property values ​​of the oil and the iron concentration in the oil.

Images