JPH0454908B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a failure detection system for determining normality or abnormality in marine equipment such as a main engine and related auxiliary equipment, or a failure diagnosis device used in a system equipped with a data logger. .

Conventionally, as a failure detection system for detecting a failure of a power engine such as a main engine or an auxiliary engine in this kind of marine equipment, the present inventors proposed in Japanese Patent Application No. 57-71287 a sensor such as a microphone installed near the power engine. By arranging multiple sensors and analyzing the acoustic spectrum obtained through the sensors, we have developed a failure detection system that detects abnormal conditions in power engines. In this failure detection system, if there is no abnormality in the sensor, it is possible to non-contactly monitor the operating state of a part that operates at high temperature and high pressure, such as a power engine, and to detect and determine whether it is normal or abnormal.

However, in the above-mentioned failure detection system, if the characteristics of sensors such as microphones change due to changes over time, there is a risk that malfunctions such as alarms occurring will occur frequently even though the driving engine is operating normally. There is sex.

An object of the present invention is to provide a failure diagnosis device for marine equipment sensors that can eliminate malfunctions in the failure detection system as described above.

Another object of the present invention is to provide a failure diagnosis device for marine equipment sensors that can detect changes in sensor characteristics and prevent malfunctions due to deterioration of sensor characteristics.

According to the present invention, the first memory section stores reference data on the sensors, the second memory section regarding the current operating state of each sensor, and the corresponding data is read out from the first and second memory sections. an arithmetic processing unit that processes the corresponding data read from both memory units and determines whether the sensor is normal or abnormal based on the processing results; A marine equipment sensor failure diagnosis device is obtained, which is equipped with a display unit that displays .

Here, the reference data includes not only data individually assigned to each sensor, but also data determined in consideration of the correlation between each sensor and other sensors.

This will be explained below with reference to the drawings.

Referring to FIG. 1, a failure diagnosis device 10 according to an embodiment of the present invention includes a microphone, a temperature sensor, A failure detection system or data that detects whether a power engine is operating normally by monitoring a plurality of sensors 11 formed by a sensor, a contact switch, etc., and the output signal of each sensor 11. It is used in combination with a logger (hereinafter simply referred to as a failure detection system) 12. The failure detection system 12 itself is described in detail in the specification of Japanese Patent Application No. 71287/1983, so the explanation will be omitted here.

The sensor 11 installed adjacent to the prime mover is normally used in a bad environment, and
The characteristics also change depending on the type or manufacturer of the sensor 11. Therefore, if the failure detection system 12 only monitors the output signal of the sensor 11, there is a risk that the failure detection system 12 will issue a false alarm due to malfunction of the sensor 11. Therefore, the ship's crew not only monitors the status of the fault detection system 12, but also monitors the sensor 11.
They also have to carry out patrol inspections, placing a heavy burden on the crew.

The illustrated failure diagnosis device 10 is a device for improving the reliability of the failure detection system 12 by monitoring changes in characteristics in the sensor 11 and preventing generation of an alarm due to malfunction of the sensor 11.

In the figure, the failure diagnosis device 10 includes each sensor 11.
input interface 1 for receiving output signals from
4, and a memory 15 for storing output signals from each sensor 11 applied via this input interface 14. The input interface 14 includes, for example, a spectrum analyzer that analyzes the frequency of an acoustic signal from a microphone and detects the power at each frequency, and an A/D converter that converts the analog signal of the analyzer into a digital signal.
A D converter or the like may also be included. On the other hand, memory 15
has a first memory section 16 that stores reference data, which will be described later, and a second memory section 17 that stores current output data of each sensor 11. Further, the memory 15 and the input/output interface 14 are connected to an arithmetic processing unit (hereinafter abbreviated as CPU) 20 via a bus 18. The CPU 20 performs arithmetic processing in a format to be described later, and sends the processing results onto the bus 18. The processing results are displayed on the display interface 2 connected to the bus 18.
1 and displayed on the CRT display device 22 via this display interface 21. Also,
A keyboard 23 is connected to the bus 18,
Using this keyboard 23, data to be accessed can be arbitrarily specified and displayed on the CRT display device 22. Furthermore, an output interface 24 is connected to the bus 18 and the processing results are sent to the fault detection system 12 via this output interface 24 . In this configuration, CPU2
If an abnormality is found in the sensor 11 as a result of the processing in step 0, the CPU 20 automatically displays the abnormal state on the CRT display device 22 and sends it to the failure detection system 12 through the output interface 24. The failure detection system 12 also refers to the output signal from the failure diagnosis device 10 to detect each sensor 1.
The calculation of the measured value No. 1 is carried out by the method described in Japanese Patent Application No. 71287/1987.

Next, the operation of the above-described failure diagnosis device 10 will be explained in relation to the sensor 11. First, when diagnosing a malfunction in the sensor 11 attached to marine equipment, the measured values of the sensors that measure the main engine's rotational speed, the temperature of the engine room, etc. (hereinafter referred to as the first sensor group) are determined to be normal. Even when it's working,
It must be taken into account that the stern tube bearing temperature and the main engine exhaust gas temperature vary depending on the main engine rotational speed and the engine room temperature, respectively, while the stern tube bearing temperature and the main engine exhaust gas temperature vary significantly depending on the speed of the ship. This means that the measured values of the sensors that detect the stern tube bearing temperature and the main engine exhaust gas temperature (hereinafter referred to as the second sensor group) are the main engine rotational speed,
This means that there is a correlation with the engine room temperature. Therefore, when diagnosing a failure in the second sensor group, it is insufficient to simply monitor the measured values of the second sensor group, and also refer to the measured values of the first sensor group to determine whether there is a failure or not. It is necessary to determine whether Note that, since the status of the first sensor group is constantly monitored by the crew, the failure diagnosis device will only diagnose whether or not a disconnection or short circuit has occurred for this first sensor group. . Examples of the above-mentioned mutually correlated data include the lubricant clear tank temperature, lubricant inlet temperature, crosshead lubricant inlet pressure, and lubricant inlet pressure.

Referring also to FIG. 2, a case is shown in which the sensor that detects the engine room temperature is a sensor of the first sensor group, and the sensor that detects the main engine exhaust gas temperature is a sensor of the second sensor group. In this example,
The measured values of the machine room temperature sensor and the main engine exhaust gas temperature sensor are indicated by x ₂ and x ₁ , respectively, and each measured value is sequentially stored in the second memory section 17 as current data. Therefore, each measurement value
x ₂ and x ₁ are expressed as x ₂ (i) and x ₁ (i) (where i indicates each time point 1 to n). The CPU 20 calculates the measured values x ₂ and x ₁ according to a predetermined program.
(subscript omitted) is read from the second memory section 17,
From each measurement value x ₂ and x ₁ , it is determined whether there is a disconnection or short circuit in the sensor. If a disconnection or short circuit occurs, the sensor is disconnected or short-circuited, as shown by .
A short circuit is displayed on the display 22. Once it is confirmed that there are no disconnections or short circuits in each sensor,
The CPU 20 calculates a deviation value using n pieces of data obtained from the main engine exhaust gas temperature sensor (hereinafter referred to as CHECK mode). In this CHECK mode, as shown on the left side of Figure 2,
The measured value x ₁ (n+1) at time (n+1) is
Detect whether it is within the range (μ±kσ) determined by the average value μ and deviation σ calculated from n data x ₁ (1), ...x ₁ (n) before this point. . Measurement value x ₁
If (n+1) falls within the above range, there is a possibility that a failure has occurred in the main engine exhaust gas temperature sensor (k is a variable integer). For this reason, CPU20
The operation moves to stage _S10 , which will be described later.
It can be seen that in CHECK mode, (μ±kσ) is used as reference data.

Next, in CHECK mode, time n and (n+
A differential value dx 1 between the measured values x ₁ (n) and x ₁ (n+1) in _{1) is calculated, and this differential value dx 1 is} compared with a preset expected maximum differential value dx _nax . If the differential value dx ₁ exceeds the expected maximum differential value dx _nax ,
Move to stage S ₁₀ behavior, otherwise,
Calculate the next differential value. Here, the expected maximum differential value dx _nax is used as reference data. This maximum differential value dx _nax is calculated during a test run, etc., and is stored in the first memory section 16.

Furthermore, in CHECK mode, time n and (n−
_The difference Δx _{1 in} the measured value between 1 ₎ and (n)
+Δx ₁ ±α) is calculated as reference data. Next, the measured value x ₁ (n+1) at time (n+1) and the predicted value x^ ₁ (n+1) are compared, and when the measured value exceeds the predicted value r times, the process moves to stage _S10 .

In the CHECK and modes described above, the performance of the sensor is checked independently, but the measured value _x1 of the main engine exhaust gas temperature sensor changes depending on the engine room temperature, as described above. Therefore, in CHECK mode, the measured value
A correlation r ₁₂ with x ₁ and x ₂ is calculated, and it is determined whether this correlation coefficient is lower than the reference data calculated during the trial run. When the correlation coefficient r ₁₂ becomes lower than the pre-calculated reference data, the operation of the CPU 20 moves to stage S ₁₀ . Here, the correlation coefficient r ₁₂ is given by r ₁₂ =σ ₁₂ /(σ ₁ σ ₂ ). However, σ ₁ and σ ₂ are x ₁ and
It is the variance of x ₂ , and σ ₁₂ is the covariance of x ₁ and x ₂ .

In the CHECK mode, a time delay in transmission between two signals is detected, and an abnormality is determined based on whether or not this time delay is within the expected time delay width. Let τ ₀ be the delay time until the signal x ₁ representing the measurement value is transmitted to the signal x ₂ , and calculate this delay time τ ₀ using correlation ₁₂ . Note that in order for the correlation function to be defined, the measured values must have stationarity, so before calculating the correlation function ₁₂ , it is checked whether or not stationarity is established. When stationarity is established, the correlation function ₁₂ is calculated, and the time delay that gives its maximum value is calculated as τ ₀ . Next, it is determined whether this time delay τ ₀ deviates from the value set during the test run, and if it deviates, the CPU
The operation at 20 moves to stage _S10 .

The above-mentioned CHECK and mode are calculations that take into account the relationship between two sensors.

In stage _S10 , if an abnormality in the main engine exhaust gas temperature sensor is detected in three or more of the CHECK~modes, the abnormality in the sensor is displayed on the display device 22. If the number of signals indicating an abnormality is two or less, the calculation operation of the CPU 20 shifts to step 1, and the above-described operation is executed again. As is clear from the above, majority logic operation is performed in stage _S10 .

In the embodiment described above, correlations and correlation coefficients are calculated during calculations that take into account the relationship between sensors, but peripheral related elements that check the logical consistency between two measured values Check and regression model checks may be performed instead of or in addition to CHECK and mode. In the regression model check, the measured values x ₂ and x ₁
Set up a regression model between the regression value and the measured value x ₁
Checks whether or not is within the threshold.

Furthermore, in the present invention, the main engine rotational speed sensor and the crosshead lubricating oil inlet pressure sensor can be used as the sensors in the first sensor group, while the stern tube bearing temperature sensor and the lubrication oil inlet pressure sensor can be used as the second sensor group. Oil inlet pressure sensors can be used respectively.

As is clear from the above description, if the fault diagnosis device according to the present invention is used, crew members can
An abnormality in the sensor can be detected simply by monitoring the display on the display device, and therefore the crew's patrol maintenance work for inspecting the sensor 11 can be reduced. Furthermore,
Since the reliability of each sensor 11 is improved, the reliability of the processing results in the failure detection system 12 is also improved.

As described above, according to the present invention, the maintenance work of the driving engine section can be simplified, and the operation of the engine section can also be simplified.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a failure diagnosis apparatus for marine equipment sensors according to an embodiment of the present invention, and FIG. 2 is a flowchart for explaining the operation of FIG. 1. Explanation of symbols, 10: Failure diagnosis device, 11: Sensor, 12: Failure detection system, 14: Input interface, 15: Memory, 16: First memory area, 17: Second memory area, 18: Bus, 2
0: CPU, 21: Display interface, 2
2: CRT display, 23: keyboard, 24: output interface.

Claims (1)

[Claims] 1. A failure diagnosis device for marine equipment sensors, which is coupled to a system including a plurality of sensors that respectively measure a plurality of factors in marine equipment, and performs failure diagnosis by monitoring changes in the characteristics of the sensors. is divided into an independent factor that changes independently and a dependent factor that changes depending on the independent factor, and the sensors are divided into a first sensor group corresponding to the independent factor and a first sensor group corresponding to the dependent factor. and a second sensor group, the failure diagnosis device includes a first memory section that stores reference data for the first and second sensor groups, and a first memory section that stores reference data for the first and second sensor groups; a second memory section for sequentially storing data;
and a calculation processing unit that performs calculation processing on the reference data and measurement data read from the second memory unit, and the calculation processing unit calculates the reference data and measurement data corresponding to the first sensor group from the reference data and measurement data read from the second memory unit. While determining the normality of one sensor group, on the condition that the normality of the first sensor group is determined,
The quality of the second sensor group is determined from the reference data and measurement data corresponding to the second sensor group, and the measurement data of the first sensor group that has a correlation with the second sensor group, and the result of the determination is A failure diagnosis device for a marine equipment sensor, characterized in that the information is displayed on a display unit.