JP6697980B2 - Equipment inspection order setting device and equipment inspection order setting method - Google Patents

Description

  The present invention relates to an equipment inspection order setting device and an equipment inspection order setting method, and in particular, an equipment inspection order setting device for setting a priority order of equipment for maintenance and inspection based on failure history collected from equipment such as cooling and heating equipment, and Regarding equipment inspection order setting method.

  Conventionally, there is a remote monitoring service for monitoring equipment such as a plurality of cooling and heating equipment from a monitoring center. In this remote monitoring service, equipment status data indicating the operating status of each equipment such as temperature and pressure of the equipment being operated is collected from each equipment. When an abnormality of the equipment is detected by analyzing the collected equipment state data, the maintenance staff performs the inspection of the equipment of which the abnormality is detected according to the created inspection schedule.

  In recent years, an inspection schedule may be created so as to prevent the occurrence of an abnormality in advance, by also inspecting equipment in which a sign of failure has occurred even if no abnormality is detected. In order to prevent abnormalities from occurring more reliably, an inspection schedule is created to prioritize inspections of equipment that is expected to fail, in other words, equipment that has a high probability of failure However, it is preferable to reduce the occurrence of failures.

  For example, in Patent Document 1, the failure history is clustered according to a predetermined condition, a failure probability distribution is created for each group of the clustered failure history (hereinafter referred to as “failure group”), and one of the failure groups is created. There is proposed a technology for calculating the priority of maintenance and inspection of the equipment, which belongs to the above-mentioned category, based on the failure probability distribution created for the failure group. Further, in Patent Document 1, a limit time (dispatch limit time) at which equipment that requires maintenance personnel to be further broke down is calculated, and the maintenance staff is dispatched based on this dispatch time limit, thereby causing equipment failure. Techniques for further reduction have also been proposed.

  In Patent Document 2, a state of a facility after a predetermined period is simulated based on a DOMDD model (partially observable Markov decision process) by using information such as facility maintenance and inspection information, facility-related alarms, and various measured values. Then, the technique of deciding the necessity of maintenance inspection of equipment is proposed.

  Patent Document 3 proposes a technique of automatically creating, adding, and updating an abnormality diagnosis model for diagnosing and responding to an abnormality of equipment by using a graph network structure. In particular, the graph network structure is used to represent the causal relationship between information such as maintenance work data, alarms, and operation events.

  In Patent Document 4, there is a failure risk table showing a correspondence relationship between the elapsed time from the time when the failure sign state is detected and the magnitude of the failure risk, and within a predetermined period from the time when the failure sign state is detected. There has been proposed a technique for determining the magnitude of failure risk of a device failure based on a failure risk table.

JP, 2014-167667, A JP, 2014-229001, A JP, 2010-122847, A JP, 2010-091840, A

  When equipment breaks down, users cannot use the equipment, so it is important to prevent failures. Therefore, it is necessary to preferentially inspect equipment that has a high probability of failure with respect to the equipment in which a sign has occurred and to reduce the failure.

  Therefore, in Patent Documents 1-4, the probability of failure is calculated based on the occurrence of a sign, but when a plurality of signs occur in a time series, which of these signs is related to the failure. So far it has not been considered. Since multiple signs include signs that are highly related to failures and signs that are less related to failures, even if the probability of failure is calculated based on the signs that are not related to failures, the accuracy of the failure probability is high. It will be low.

  Therefore, in the present invention, when a plurality of signs have occurred, a sign highly associated with the failure is extracted, and the priority of the equipment for maintenance and inspection is set with high accuracy based on the extracted sign. To aim.

The equipment inspection order setting device of the present invention includes a failure history storage unit that stores information on signs collected from a plurality of pieces of equipment and information on failures that have occurred after the sign, and the failure history storage unit that has stored the failure history storage unit. Based on the sign and the failure, extracting the sign associated with the failure from the plurality of signs that occurred before the failure, a sign pattern extraction unit that creates a sign pattern based on the extracted sign. A transition probability accumulating unit that calculates and accumulates transition probabilities between predictive patterns and predictive failures in the predictive pattern, and when the first predictive sign in the predictive pattern occurs, the failure from the predictive sign based on the transition probability. To a cumulative failure probability calculating unit for calculating a cumulative failure probability up to, and a priority calculation for calculating the priority of the equipment for performing maintenance and inspection on the basis of the cumulative failure probability for a plurality of the equipment in which the sign has occurred The predictive pattern extraction unit is configured to compare the rate of change of the failure probability between the predictors with a threshold value for determining the relevance to the failure, thereby relating the failure to the failure from a plurality of the predictors. It is characterized in that the above-mentioned sign to be extracted is extracted .

  Further, the predictive pattern extraction unit, when the predictor of the same type continuously exists in the predictive pattern, creates the predictive pattern with the same predictor of the same type as one of the predictors. To do.

  Further, the cumulative failure probability calculation unit is characterized in that the cumulative failure probability is calculated by Monte Carlo simulation, and an upper limit number of times of the Monte Carlo simulation is set according to the number of signs in the sign pattern.

  Further, based on the priority calculated by the priority calculation unit, a dispatch limit time calculation unit for calculating a dispatch limit time which is a limit time for the equipment to fail unless the maintenance personnel are dispatched is provided. Is characterized by.

Furthermore, the equipment inspection order setting method of the present invention is an equipment inspection order setting method for setting the priority order of the equipment to be inspected and maintained based on the failure history collected from each of the plurality of installations. and information sign collected respectively from the step of storing the information of the failure that occurred after sign, based on the sign and the fault associated with the failure of a plurality of said sign that occurred prior to the failure extracting the sign, and creating a sign pattern based on the extracted the sign, a step of storing and calculating the transition probabilities between sign and between sign fault in the sign pattern, the first in the predictor pattern When the sign occurs, based on the transition probability, a step of calculating a cumulative failure probability from the sign to the failure, and for the plurality of the equipment in which the sign occurred, based on the cumulative failure probability Comprising a step of calculating the priority of the equipment to perform maintenance, the step of creating the predictive pattern, the change rate of the failure probability between each predictor, and a threshold for determining the relationship with the failure. By comparing, the predictor associated with the failure is extracted from a plurality of the predictors .

  According to the present invention, when a plurality of signs have occurred, a sign highly associated with a failure is extracted, and based on the extracted signs, the priority of the facility for maintenance and inspection is set with high accuracy. be able to.

It is a block configuration diagram of a cooling and heating facility system including a facility inspection order setting device. It is the figure which showed the data structural example of the failure history information which shows the failure history accumulate|stored in the failure history database. It is a hardware block diagram of the computer which comprises an equipment inspection order setting device. It is a schematic diagram of a sign leading to a failure. It is a flowchart which shows the process of an equipment inspection order setting device. It is a schematic diagram which shows the some sign pattern regarding the sign leading to a failure. It is a schematic diagram of the predictive pattern which each shows the some predictive pattern shown in FIG. It is a mimetic diagram of a sign explaining a case where the same type of sign is thinned. It is a figure explaining the transition state from a sign occurrence to a failure, (a) shows a continuous sign relationship when going from a sign occurrence to a failure, and (b) calculates the transition probability over time in a continuous sign relationship. The table is shown. It is a figure which shows the path|route to a failure when a sign occurs, (a) shows the path|route from the sign Y2 to the failure K2, (b) shows the path|route from the sign Y3 to the failure K2, (c) is. The route from the sign Y4 to the failure K2 is shown. 7 is a flowchart showing a process of calculating a cumulative failure probability by Monte Carlo simulation. 7 is a flowchart showing a process of calculating a cumulative failure probability by Monte Carlo simulation.

  FIG. 1 is a block diagram showing an embodiment of an equipment inspection order setting device according to the present invention. FIG. 1 shows a cooling/heating facility 2, an operation history receiving unit 3, a failure history database (DB) 4, and a facility inspection order setting device 10. As shown in FIG. 1, the equipment inspection order setting device 10 is connected to a plurality of cooling and heating equipments 2 installed in a plurality of contractor facilities via an operation history receiving unit 3 and a failure history database 4. In the present embodiment, the cooling/heating equipment 2 is shown as the equipment to be inspected, but other equipment that requires inspection may be used.

  The operation history receiving unit 3, the failure history database 4, and the equipment inspection order setting device 10 are installed in, for example, a monitoring center. The operation history receiving unit 3 receives the equipment state data indicating the equipment state from the cooling/heating equipment (hereinafter, simply referred to as “equipment”) 2, the equipment state data when a sign and an abnormality occur, and the received equipment state data. Are accumulated as necessary to generate failure history information and accumulate it in the failure history database 4. The operation history receiving unit 3 is composed of a computer having a communication function, and the failure history database 4 is composed of a database server. As a method of receiving the equipment state data and accumulating it in the failure history database 4, a method of transmitting it online via a telephone line and an internet line by communication software installed in a control device (not shown) of the equipment 2 is used. There is.

  FIG. 2 is a diagram showing a data configuration example of failure history information showing failure history accumulated in the failure history database 4 in the present embodiment. The failure history information is configured by associating a failure time interval, installation environment information, and occurrence status information with a sign occurrence date and time that indicates a date and time when a sign of failure has occurred. The sign occurrence date and time is the date and time when the sign of occurrence is detected by the operation history receiving unit 3 referring to the equipment state data transmitted from the equipment 2. The failure time interval is a time interval from the occurrence of a sign to the failure of the equipment 2, and is a value indicating the difference between the date and time of the abnormality and the date and time of the sign. In this embodiment, the number of days is shown, but it may be hour. Here, “abnormal” means a case where one piece of equipment state data deviates from a normal state indicating a value within a normal range. “Failure” means that the equipment 2 stops due to the occurrence of an abnormality in the equipment state data. Further, the “sign” means that the equipment state data exceeds a threshold value indicating a state close to an abnormal state even within the normal range.

  The installation environment information is information about the specifications and installation location of the equipment 2, and includes an equipment number that identifies the equipment 2, a building number that identifies the building in which the equipment 2 is installed, an installation location that indicates an address where the equipment 2 is installed, It includes various information regarding the installation start indicating the year of installation, the model name of the equipment 2, and the installation direction indicating the direction in which the equipment 2 is installed in the building.

  The occurrence status information is information related to the occurrence status when a sign or an abnormality occurs. The sign content indicating the contents of the sign that has occurred, the temperature and humidity at the time when the sign occurred, the temperature and the set temperature at the time when the sign occurred. And the temperature difference, the number of times the thermostat of the equipment 2 is switched ON/OFF from the start of counting to the occurrence of a sign, the number of times of starting/stopping, the equipment from the start of counting to the occurrence of a sign 2 is the operating time when the time is running, the thermo ON time when the thermostat of the equipment 2 is ON from the start of counting to the occurrence of the sign Start-stop frequency ratio indicating the number of start-stop operations, thermo-ON time ratio indicating the thermo-ON time per operating time, the average that the indoor temperature of the equipment 2 reaches the set temperature between the start of counting and the occurrence of a sign It includes a target temperature arrival time indicating time, an abnormality occurrence date and time indicating a date and time when the facility 2 is stopped due to the occurrence of an abnormality, and various kinds of information regarding an abnormality content indicating details of the abnormality that has occurred. Note that FIG. 2 illustrates a part of the occurrence status information described above.

  FIG. 3 is a hardware configuration diagram of a computer forming the equipment inspection order setting device 10 according to the present embodiment. The computer forming the equipment inspection order setting device 10 in the present embodiment can be realized by a hardware configuration of a general-purpose personal computer (PC) that has existed in the past. That is, as shown in FIG. 3, the computer is provided with a CPU 21, a ROM 22, a RAM 23, an HDD controller 25 to which a hard disk drive (HDD) 24 is connected, a mouse 26 and a keyboard 27 provided as input means, and a display device. An input/output controller 29 for connecting each of the displays 28 and a network controller 30 provided as a communication means are connected to an internal bus 31.

  FIG. 4 shows the signs Y1, Y2, Y3, Y4 in the time series that occurred between the failure K1 and the failure K2. That is, the signs Y1, Y2, Y3 and Y4 leading to the failure K2 are shown. In FIG. 4, the outline content of setting the priority order of the equipment inspection by the equipment inspection order setting device 10 will be described, and then each configuration of the equipment inspection order setting device 10 will be described. In FIG. 4, signs Y1, Y2, Y3, and Y4 occur after the occurrence of the failure K1, and then the failure K2 occurs. Further, among the signs Y1, Y2, Y3, and Y4, only the sign type of the sign Y3 is the B sign, and the sign types of the other signs Y1, Y2, and Y4 are the A sign.

  The equipment inspection order setting device 10 extracts a sign highly related to the failure K2 from the signs Y1, Y2, Y3, and Y4 leading to the failure K2 based on the failure history data, and the sign based on the extracted sign. The pattern D is extracted. Then, the equipment inspection order setting device 10 calculates a cumulative failure probability up to the failure K2 when the first sign Y2 in the sign pattern D occurs in the equipment operating state, and a plurality of cumulative failure probabilities are calculated based on the cumulative failure probability. The priority order of the equipment 2 for performing maintenance inspection on the equipment 2 is set.

  Returning to FIG. 1, the equipment inspection order setting device 10 includes a predictive pattern extraction unit 11, a transition probability accumulation unit 12, a cumulative failure probability calculation unit 13, a priority calculation unit 14, and a dispatch limit time calculation unit 15. The predictive pattern extraction unit 11 extracts a predictive sign that is highly related to the failure from the plurality of predictive signs leading to the failure. As shown in FIG. 4, for example, a failure K1 occurs, and after the failure K1 is recovered, a plurality of Y1, Y2, Y3, and Y4 occur while the next failure K2 occurs. The failure K2 may occur due to, for example, one sign Y4, but usually, it often occurs due to a plurality of consecutive signs Y1, Y2, Y3, and Y4. Even if a plurality of signs Y1, Y2, Y3, and Y4 are consecutive, among these signs Y1, Y2, Y3, and Y4, there is a time series from the occurrence of the sign Y1 or the failure K2, which has low relevance to the failure occurrence. A warning sign (not shown) is also included. Therefore, the predictive pattern extraction unit 11 closely relates to, for example, predictors Y2, Y3, and Y4 that are highly related to the failure K2 among the plurality of predictors Y1, Y2, Y3, and Y4 that lead to the failure K2, that is, failure occurrence. Extract relevant signs Y2, Y3, Y4. The predictive pattern extraction unit 11 extracts the predictors Y2, Y3, and Y4 that are closely related to the occurrence of the failure, excludes the predictor Y1 that has a low relevance to the failure occurrence, and predicts the signs Y2, Y3, and Y4 that lead to the failure K2. Predictive pattern D consisting of The predictive pattern D is a summary of one or a plurality of predictors (predictors Y2, Y3, Y4 in FIG. 4) immediately before the failure, and when a plurality of predictors are involved, a predictive sign Y2 that easily leads to the failure K2. The order of occurrence of Y3 and Y4 is also defined. In FIG. 4, the predictive pattern D is defined in the order of the predictive signs Y2, Y3, and Y4.

  The transition probability accumulating unit 12 calculates the transition probability between the signs in the extracted sign pattern D, and accumulates the transition probability of the sign pattern D. The transition probability is the probability of transition from one sign to the next sign or failure. The transition probability accumulating unit 12 calculates transition probabilities in a lapse of time between predictors and predictive failures and accumulates the calculation results. For example, as shown in FIG. 9A, in the predictive pattern D3 including predictors Y2, Y3, and Y4, when the predictor Y2 reaches the failure K2, the predictive Y2 and the failure K2 are continuous (continuous portion L1), A part where the sign Y2 and the sign Y3 are continuous (continuous part L2), a part where the sign Y3 and the failure K2 are continuous (continuous part L3), a part where the sign Y3 and the sign Y4 are continuous (continuous part L4), a sign Y4 And a failure K2 are continuous in the part (continuous part L5) of the predictive interval and the predictive failure. The transition probability accumulating unit 12 calculates transition probabilities for the continuous portions L1 to L5 and accumulates the transition probabilities. That is, in the cumulative failure probability calculation unit 13, in order to calculate the cumulative failure probability up to the failure K2, the transition probability of the predictive pattern D is calculated in advance, accumulated, and stored as a table.

  When the first sign of the sign pattern D occurs, the cumulative failure probability calculator 13 calculates the cumulative failure probability from the first sign to the failure based on the transition probability calculated by the transition probability accumulator 12. For example, as shown in FIG. 9A, when the first sign Y2 in the sign pattern D3 occurs, the cumulative failure probability leading to the failure K2 is calculated. Since the signs Y2, Y3, and Y4 do not always occur in the order of the signs of the sign pattern D3 during the actual operation of the facility, the cumulative failure probability of reaching the failure K2 is calculated for this sign pattern D3.

  The priority calculation unit 14 calculates the priority of maintenance and inspection of each facility in which a sign has occurred, based on the cumulative failure probability calculated by the cumulative failure probability calculation unit 13. The dispatch time limit calculation unit 15 calculates the dispatch time limit based on the priority calculated by the priority calculation unit 14. The maintenance staff is on standby at the branch office or the like, and heads to the facility 2 which is the maintenance/inspection destination, if necessary. Basically, when a sign of a failure occurs, it is preferable to perform maintenance and inspection before the time when a failure is predicted to occur in order to prevent the occurrence of the failure in the facility 2 in which the sign has occurred. .. Here, the dispatch limit time is a limit time when the equipment 2 fails unless the maintenance personnel are dispatched.

  The specific processes in the predictive pattern extraction unit 11, the transition probability accumulation unit 12, the cumulative failure probability calculation unit 13, the priority calculation unit 14, and the dispatch limit time calculation unit 15 will be described later in detail in the equipment inspection order setting device 10. The operation will be described in detail.

  The program for executing the facility inspection order setting method according to the present embodiment can be provided not only by communication means but also by being stored in a computer-readable recording medium such as a CD-ROM or a DVD-ROM. Is. The program provided from the communication means or the recording medium is installed in the computer, and the CPU of the computer sequentially executes the program to realize the equipment inspection order setting method.

  Next, the operation of the equipment inspection order setting device 10 will be described. While the equipment 2 is operating, the equipment history data including the information necessary for generating the failure history information is transmitted from the equipment 2 having the sign or the failure to the operation history receiving unit 3. When the operation history reception unit 3 detects the occurrence of a sign by analyzing the received equipment state data, the sign of occurrence date/time of the sign, installation environment information, sign content included in the occurrence status information, temperature, humidity, The failure history information is generated by setting the temperature difference, the number of starts and stops, the operating time, the thermo ON time, the start and stop ratio, the thermo ON time ratio, and the target temperature arrival time. Further, when the operation history receiving unit 3 detects the occurrence of the abnormality by analyzing the received equipment state data, the operation history receiving unit 3 refers to the installation environment information or the like from the failure history information generated due to the occurrence of the sign and the abnormality occurs. The failure history information corresponding to the installed equipment 2 is specified, and the date and time and the content of the abnormality of the failure history information are set and registered based on the received equipment state data. Further, the operation history receiving unit 3 calculates the time interval from the occurrence of the sign to the occurrence of the abnormality, and sets and registers the failure time interval. In this way, the failure history information is generated by the operation history receiving unit 3 in accordance with the occurrence of the sign and the failure, and the failure history information shown in FIG. 2 is accumulated in the failure history database 4.

  Next, the processing of the equipment inspection order setting device 10 will be described with reference to the flowchart shown in FIG. In step S101, the predictive pattern extraction unit 11 extracts a predictive sign that is highly associated with a failure from a plurality of predictive signs that lead to a failure, and creates a predictive sign pattern based on the extracted predictive sign. Extraction of a sign and creation of a sign pattern will be described with reference to FIGS. In FIG. 6, similar to FIG. 4, the signs Y1, Y2, Y3, Y4 occur after the occurrence of the failure K1, and then the failure K2 is reached. In FIG. 6, the symbol T indicates the time axis. Further, among the signs Y1, Y2, Y3, and Y4, only the sign type of the sign Y3 is the B sign, and the sign types of the other signs Y1, Y2, and Y4 are the A sign.

  First, the creation of the predictive pattern D will be described. As shown in FIG. 6, it is determined which of the signs Y1, Y2, Y3, Y4 is closely related to the failure K2 by tracing back from the failure K2 by using the change rate of the failure probability when the number of signs is changed. To do. In other words, whether the cause of the failure K2 is only the sign Y4 (sign pattern D1), the signs Y3, Y4 (sign pattern D2), the signs Y2, Y3, Y4 (sign pattern D3), Alternatively, it is determined whether the sign is Y1, Y2, Y3, Y4 (sign pattern D4).

  Which of the predictive patterns D1 to D4 is selected is based on the following concept. For example, when the predictive pattern D1 and the predictive pattern D2 are compared, if the change rate of the failure probability is small, the change in the failure tendency due to the increase in the number of signs is small, that is, even if the number of signs is increased, the relevance to the failure is low. It is determined to be low, and the predictor pattern D1 having a small number of predictors is adopted. Further, when the predictive pattern D1 and the predictive pattern D2 are compared, if the change rate of the failure probability is large, the change in the failure tendency due to the increase in the number of predictors is large, that is, by increasing the number of predictors, the relevance to the failure is high. When it is determined that the number of signs is high, the sign pattern D2 having many signs is adopted.

  If the number of predictors in the predictive pattern is increased without limitation because it is determined that the number of predictive patterns is high, the probability of failure will decrease due to the increase in the number of predictive patterns, and the data amount of the predictive pattern will also increase. Therefore, it is preferable that the number of signs in the sign pattern is as small as possible. Also, in order to estimate the occurrence of a failure, it is preferable that there are many types of predictive patterns. Further, when there is a predetermined time interval between the signs, that is, when the signs are separated in time series, the sign that is far from the failure occurrence is judged to be unlikely to be related to the failure, and Exclude from the pattern.

Next, a specific method of determining the predictive pattern will be described. FIG. 7 shows the predictive patterns D1, D2, D3, D4 leading to the failure K2 shown in FIG. 6 separately. Here, as indicated by the symbol F1, the failure probability P ₁ when the sign Y4 occurs and reaches the failure K2 (symptom pattern D1) is 0.2, and as indicated by the sign F2, the sign Y4 is followed by the sign Y4. Occurs and the failure K2 occurs (symptom pattern D2), the failure probability P ₂ is 0.35, and as indicated by the symbol F3, when the signs Y3 and Y4 follow the sign Y2 and the failure K2 occurs ( The failure probability P ₃ of the predictive pattern D3) is 0.23, and as indicated by reference sign F4, the failure probability P of when the predictive patterns Y2, Y3, Y4 occur following the predictive Y1 to reach the failure K2 (the predictive pattern D4). Set ₄ to 0.2.

The change rate PC _x of the failure probability P _x can be calculated by the following mathematical formula 1.
The rate of change _{PC x = | P x -P x} + 1 | × 100 / P x ··· number 1
In addition, a threshold value for determining the relevance to a failure is set in advance based on the change rate PC _x of P _x of the predictive pattern D. When the rate of change PC _x of P _x of the predictive pattern D exceeds the threshold value, it is determined that the correlation with the failure is high. Therefore, when the threshold value is set small, the level of judging the relationship with the failure is lowered, the number of signs in the sign pattern D is increased, and conversely, when the threshold value is set large, the level of judging the relationship with the failure is increased, The number of signs in the sign pattern D decreases. Here, the threshold value is set to, for example, a threshold value of 30%. Then, the change rate PC _x of the failure probability _{P x} of predictor pattern D is, until the threshold value 30% or less, comparing the change rate PC _x of the failure probability _{P x} of predictor pattern D.

When the change rate PC _x of the failure probability P _x of the predictive patterns D1, D2, D3, D4 is calculated based on the mathematical formula 1, the predictive pattern D1 and the predictive pattern D2 are
Rate of change _{PC x (D1 · D2) =} | 0.2-0.35 | next × 100 / 0.2 = 75%, when compared to 30% threshold, since more than 30% threshold, the predictor pattern D2 adopt.

Similarly, between the predictive pattern D2 and the predictive pattern D3,
The rate of change PC _x (D2·D3)= | 0.35-0.23 | ×100/0.35=34%, which exceeds the threshold of 30% as compared with the threshold of 30%. The pattern D3 is adopted.

Similarly, in the predictive pattern D3 and the predictive pattern D4,
Rate of change _{PC x (D3 · D4) =} | 0.23-0.2 | next × 100 / 0.23 = 13%, when compared to 30% threshold, because it is 30% or less the threshold, in this case the number of sign The predictive pattern D3 with few In this way, the predictive pattern D3 is adopted as the predictive pattern D leading to the failure K2 by comparing the change rates PC _x of the failure probabilities P _x with respect to the predictive patterns D1, D2, D3, D4.

  Further, in FIG. 8, after the signs Y10, Y11, and Y12 of the same type (A sign) are successively generated, the sign Y13 of the sign B is generated, and then the sign Y14 of the sign A is generated and the failure K2 is reached. Indicate the case. As shown in FIG. 8, when signs Y10, Y11, and Y12 of the same type (A sign) occur consecutively, the signs of the same type are regarded as one sign, and the signs of the same type are thinned out to give a sign. Create a pattern D. That is, the same type of sign often occurs consecutively (for example, the sign A of FIG. 8 continuously occurs), and when the sign pattern D is created including all of the same type of sign, Since many similar predictor patterns D are created, when predictors of the same type occur consecutively, the predictor pattern D is created by regarding the predictors of the same type as one predictor.

  Returning to the flowchart of FIG. 5, after the sign pattern D3 is created in step S101, the process proceeds to step S102. In step S102, the transition probabilities between the respective predictors in the predictive pattern D3 and the elapsed time between the predictive failures are calculated, and the calculated transition probabilities are accumulated. As for the predictive pattern D3, as shown in FIG. 9A, in the path from the predictive Y2 to the failure K2, there are five predictive intervals and predictive failure, that is, continuous portions L1, L2, L3, L4, and L5. The transition probability accumulating unit 12 calculates transition probabilities in the passage of time between the predictors Y2, Y3, Y4 in the continuous portions L1, L2, L3, L4, L5 and between the predictors Y2, Y3, Y4 and the failure K2, The calculated transition probabilities are accumulated. FIG. 9B shows a table in which the transition probabilities of the continuous portions L1, L2, L3, L4, L5 over time are calculated. The calculated transition probability table is stored in the RAM 23 or the HDD 24. After calculating the transition probability, the process proceeds to step S103.

  In step S103, the cumulative failure probability of the predictive pattern D3 is calculated. A known Monte Carlo method is used to calculate the cumulative failure probability of the predictive pattern D3 in step S103. The Monte Carlo method is a method of performing a numerical calculation using random numbers to obtain an approximate solution of a required solution, and in the present embodiment, the Metropolis method is used among the Monte Carlo methods.

  The predictive pattern D3 in FIG. 10A will be described as an example. As shown in FIG. 10A, there are three routes G1, G2, and G3 when the sign Y2 occurs and the failure K2 occurs. That is, when the sign Y2 occurs, the route G1 from the sign Y2 directly to the failure K2, the route G2 from the sign Y2 to the failure Y3 to the failure K2, and the failure from the sign Y2 to the signs Y3 and Y4. There are three routes, G3, which leads to K2. By using the Monte Carlo method for these three routes G1, G2 and G3, one cumulative failure probability is calculated.

  Further, as shown in FIG. 10B, when the sign Y3 occurs, that is, the route G4 directly from the occurrence of the sign Y3 to the failure K2 and the route from the sign Y3 to the failure K2 via the sign Y4. Similarly, for two G5 types, one cumulative failure probability is calculated using the Monte Carlo method. Note that, as shown in FIG. 10C, when the sign Y4 occurs, there is one path G6 leading to the failure K2, and therefore it is not necessary to use the Monte Carlo method.

  11 and 12 show flowcharts of processing by Monte Carlo simulation. First, an outline of Monte Carlo simulation will be described. In the Monte Carlo simulation, (1) the sign that occurs at the beginning of the sign pattern D is set as an initial value x0, and (2) the current state is set as xt, and the next state candidate, that is, the sign or failure that occurs next is x. 'And select (3) the ratio r=P(x')/P(xt) of the probability of the next state candidate and the probability of the current state, and (4) generate the uniform random number R, (5) If R<r, the next state candidate (next sign or failure) is set to the next state, and if R<r, the current state is set as the next state, and (6) (2) to ( 5) is repeated until time t. Further, the processes (1) to (6) are repeated until the distribution of cumulative failure probabilities converges. If it does not converge, this simulation is repeated up to the preset upper limit number of times.

  In FIG. 11, in step S201, a sign that occurs at the beginning of the sign pattern D3 is set. That is, the sign Y2 of the sign pattern D3 is set as an initial value, and the process proceeds to step S202.

  In step S202, the initial value (symptom Y2) set in step S201 is selected as the current state, and the process proceeds to step S203.

  In step S203, a transition candidate that occurs next to the sign Y2 is selected. As shown in FIG. 10A, the transition candidate that occurs next to the sign Y2 may be the sign Y3 or the failure K2. Therefore, the sign Y3 or the failure K2 is selected as the transition candidate, and the process proceeds to step S204.

  In step S204, the transition probability at time t (here, 1 hour) is calculated for the transition candidate selected in step S203, and the process proceeds to step S205. Based on the above P(x′)/P(xt), the transition probability between the sign Y3 and the failure K2 at the time t is P(sign Y3)/P(sign Y2)=0.2, P(fail K2) )/P (symptom Y2)=0.4. The transition probability of the sign Y3 and the transition probability of the failure K2 may be calculated for each process of the flowchart, but the table in FIG. 9B is used here. With reference to the table in FIG. 9B, the transition probability of the sign Y3 and the transition probability of the failure K2 that occur within one hour from the occurrence of the sign Y2 are 0.2 for the sign Y3 and 0.4 for the failure K2. is there.

  In step S205, a uniform random number R is generated and the process proceeds to step S206. This uniform random number R is used to estimate a transition candidate that occurs next to the sign Y2. For example, the generated uniform random number R is 0.7.

  In step S206, the uniform random number R is compared with the transition probability of the sign Y3 and the transition probability of the failure K2 calculated in step S204 to determine the next state. As the next state, as shown in FIG. 10A, there is a state of maintaining the current state, transiting to the next sign, or transiting to a failure. If it is maintained, the process proceeds to step S207, if it transits to the next sign, it proceeds to step S209, and if it transits to a failure, it proceeds to step S211.

  This time, the state next to the sign Y2 is determined. When the uniform random number R is smaller than the transition probability of the sign Y3, the transition to the sign Y3 occurs, and when the uniform random number R is smaller than the sum of the transition probability of the sign Y3 and the transition probability of the failure K2, When the transition to the failure K2 occurs and the uniform random number R exceeds the total value, the current sign Y2 is maintained.

  For example, judging from the above numerical values, the uniform random number R(0.7), the transition probability of the sign Y3 (0.2), and the transition probability of the failure K2 (0.4), so the uniform random number R(0 .7) exceeds the total value (0.2+0.4), the process proceeds to step S207 and the current state of the sign Y2 is maintained.

  In step S207, neither the sign Y3 nor the failure K2 is transitioned, so the state of sign Y2 is maintained and the process proceeds to step S208. In step S208, after counting the time t, the time t is compared with the upper limit time N to determine whether the time t has reached the upper limit time N. That is, when the time t has not reached the upper limit time N (Yes), the process returns to step S202, and step S202 and subsequent steps are repeated. If the time t has reached the upper limit time N (No), the process proceeds to step S212 in FIG.

  Subsequently, a case where the process returns from step S208 to step S202 will be described. When step S202 and subsequent steps are repeated, the transition candidate from the sign Y2 becomes the sign Y3 or the failure K2 as in the previous case. Since the time t is counted up to 2 hours this time, the transition probability of the sign Y3 and the transition probability of the failure K2 are 0.3 for the sign Y3 and 0 for the failure K2 by referring to the table in FIG. 9B. .2. As a result of generating the uniform random number R in step S205, for example, when the uniform random number R is 0.25, the uniform random number R (0.25), the transition probability (0.3) of the sign Y3, The transition probability of the failure K2 is (0.2). Under this condition, in step S206, it corresponds to a case where the uniform random number R(0.25) is smaller than the transition probability (0.3) of the sign Y3, and the process proceeds to step S209 to transit to the next state of the sign Y3. ..

  In step S209, it is determined that the next sign (sign Y3) is entered, and the process proceeds to step S210. In step S210, the next sign (symptom Y3) is changed to the current state, that is, the current state is changed from sign Y2 to sign Y3, the process returns to step S202, and steps S202 and subsequent steps are repeated. When returning to step S202 and repeating step S202 and subsequent steps, the transition candidate from the sign Y3 becomes the sign Y4 or the failure K2, as shown in FIG. 10B. From the table of FIG. 9B, the transition probability of the sign Y4 and the transition probability of the failure K2 are 0.5 for the sign Y4 and 0.5 for the failure K2. Then, as a result of generating the uniform random number R in step S205, for example, when the uniform random number R is 0.6, the uniform random number R(0.6), the transition probability of the sign Y4 (0.5), The transition probability of the failure K2 is (0.5). Under this condition, in step S206, if the uniform random number R(0.6) is smaller than the total value of the transition probability (0.5) of the sign Y4 and the transition probability (0.5) of the failure K2, Correspondingly, the process proceeds to step S211, and transitions to the state of failure K2.

  In step S211, it is determined that the failure K2 is made, and the process proceeds to step S212. In step S212, one simulation is completed, the number of simulations is counted up, and the process proceeds to step S213.

  In step S213, it is determined whether the number of simulations has reached the preset upper limit number. When the upper limit number is reached (Yes), a series of processing is ended, and when the upper limit number is not reached (No), the process proceeds to step S214.

  In step S214, a series of processes is repeated until the distribution of cumulative failure probabilities converges. That is, in step S214, if the cumulative failure probability distribution has converged (Yes), a series of processing is terminated, and if the cumulative failure probability distribution has not converged (No), the process returns to step S201, and steps S201 and subsequent steps are performed. repeat.

  Here, the setting of the upper limit number of simulations will be described. The greater the number of predictors in the predictive pattern D, the greater the number of branches leading to a failure. Therefore, the upper limit number of simulations is set according to the number of signs in the sign pattern D. For example, when the number of signs in the sign pattern D is 1, the upper limit number of simulations is set to 10,000 times, and when the number of signs is two in the sign pattern D, the upper limit number of simulations is set to 20000 times. When the number of signs in the sign pattern D is 3, like the sign pattern D3, the upper limit number of simulations is set to 30,000.

  Returning to the flowchart of FIG. 5, in step S104, the priority of performing maintenance and inspection on each facility in which a sign has occurred is calculated based on the cumulative failure probability calculated in step S103. In other words, the equipment that requires the most maintenance/inspection, that is, the equipment that has the highest probability of failure is selected based on the cumulative failure probability from among the equipment that has generated the sign. In step S104, in the equipment operating state, when each of a plurality of facilities has a sign, by comparing the cumulative failure probabilities of the sign pattern D relating to the sign, the probability of the most failure from the plurality of facilities having the sign is determined. Select expensive equipment and proceed to step S105.

  As a specific example of the process of step S104, for example, when a sign occurs in a plurality of equipment at a certain timing, the cumulative failure probabilities of the sign patterns related to these signs are compared, and the cumulative failure probability is high among these equipment. Select equipment with a predictive pattern. Then, priority is given to maintenance and inspection of the selected equipment. In this way, the equipment that requires the most maintenance and inspection (equipment with a high probability of failure) can be grasped based on the cumulative failure probability.

  In step S105, the dispatch time limit is calculated for the equipment selected in step S104 that requires the most maintenance. The dispatch time limit is calculated by comparing the predicted failure time, which is predicted to lead to a failure after the sign has been generated, with the maintenance time required for maintenance personnel to reach the facility and complete maintenance and inspection. ..

  The failure prediction time is calculated based on a predictive pattern regarding a predictive sign that has occurred. That is, it is calculated based on the estimated time from the occurrence of the sign to the failure. Further, the maintenance time is calculated based on the time required for a maintenance worker who is waiting at a branch office or the like close to the facility where the sign has occurred to go to the facility and complete the maintenance inspection. Then, the dispatch time limit is calculated so that the maintenance time is shorter than the failure prediction time, and the maintenance staff performs the maintenance inspection based on the dispatch time limit.

  As described above, the equipment inspection order setting device 10 creates a predictive pattern D by extracting a predictor that is highly related to a failure from among a plurality of predictors leading to a failure based on the failure history data. When the first sign of the sign pattern D occurs, the cumulative failure probability leading to a failure is calculated. Then, the equipment inspection order setting device 10 calculates the priority of the equipment 2 to be inspected and maintained based on the cumulative failure probability. For this reason, when multiple signs are generated before the failure, signs that are not related to the failure are excluded from these signs and the failure is reached based on the signs that are highly related to the failure. The failure probability can be calculated, and the accuracy of the failure probability can be improved.

  As a result, the priority of the equipment 2 to be inspected can be calculated with high accuracy, and the priority of the equipment 2 to be inspected can be determined in consideration of the failure time and the probability of failure. Further, by setting the maintenance inspection order from the equipment 2 that is prone to failure, it is possible to efficiently perform the maintenance and inspection, reduce the failure, and use the equipment 2 for a long period of time.

  Further, it has a threshold value for judging the relationship between the sign and the failure, and by adjusting this threshold value, the number of signs related to the failure in the sign pattern D can be adjusted. Also, the data amount of the predictive pattern D can be adjusted.

  In addition, when the same type of sign continuously exists in the sign pattern D, since the sign pattern D is created by using the same type of sign that is continuous as one sign, the similar sign pattern D can be excluded. It is possible to secure the data amount of different types of predictive patterns D.

  Further, the cumulative failure probability is calculated by Monte Carlo simulation, and the upper limit of the number of times of Monte Carlo simulation is set according to the number of predictors in the predictor pattern D, so that the simulation adapted to the predictor pattern D can be performed. , The load of calculation by simulation can be reduced. In particular, when the amount of equipment and the amount of failure history data are large, the amount of calculation for Monte Carlo simulation is enormous, but by setting the upper limit of the number of simulations, you can obtain the necessary calculation result and the calculation load. Can be reduced.

  Further, since the dispatch limit time calculation unit 15 calculates the dispatch limit time based on the priority calculated by the priority calculation unit 14, the maintenance staff performs maintenance and inspection of the equipment 2 based on the calculated dispatch limit time. By doing so, it becomes possible to further reduce the failure of the equipment 2.

  2 equipment, 3 operation history receiving unit, 4 failure history database, 10 equipment inspection order setting device, 11 predictive pattern extraction unit, 12 transition probability storage unit, 13 cumulative failure probability calculation unit, 14 priority calculation unit, 15 dispatch limit time Calculator, D, D1, D2, D3, D4 predictive pattern, G1, G2, G3, G4, G5, G6 route, K1, K2 failure, L1, L2, L3, L4, L5 continuous part, Y1, Y2, Y3 , Y4, Y10, Y11, Y12, Y13Y, 14 Signs.

Claims (5)

A failure history storage unit that stores the sign information collected from each of the plurality of facilities and the information about the failure that occurred after the sign,
Based on the sign and the failure accumulated in the failure history storage unit, the sign associated with the failure is extracted from the plurality of signs that have occurred before the failure, and the sign based on the extracted sign. A predictive pattern extraction unit that creates a pattern,
A transition probability accumulating unit that calculates and accumulates transition probabilities between the predictive signs and the predictive failures in the predictive pattern,
When the first sign of the sign pattern occurs, based on the transition probability, a cumulative failure probability calculation unit that calculates a cumulative failure probability from the sign to the failure,
For a plurality of the equipment in which the sign has occurred, a priority calculation unit for calculating the priority of the equipment to perform maintenance and inspection based on the cumulative failure probability,
Equipped with
The sign pattern extraction unit extracts the sign associated with the failure from a plurality of the signs by comparing a rate of change in failure probability between the signs and a threshold value for determining the relationship with the failure. Equipment inspection order setting device characterized by the above. The equipment inspection order setting device according to claim 1 ,
The sign pattern extraction unit, when the sign of the same type continuously exists in the sign pattern, creates the sign pattern by using the sign of the same kind that is continuous as one sign. Inspection order setting device. The equipment inspection order setting device according to claim 1 or 2 ,
The cumulative failure probability calculation unit calculates the cumulative failure probability by Monte Carlo simulation, and sets an upper limit number of times of the Monte Carlo simulation in accordance with the number of signs in the sign pattern. . The equipment inspection order setting device according to any one of claims 1 to 3 ,
Based on the priority calculated by the priority calculation unit, a dispatch limit time calculation unit that calculates a dispatch limit time that is a limit time when the equipment fails unless the maintenance personnel are dispatched is provided. Equipment inspection order setting device. Based on the failure history collected from each of a plurality of facilities, a facility inspection order setting method for setting the priority of the equipment to perform maintenance inspection,
A step of accumulating information on the sign that has been collected from each of the plurality of facilities and information on a failure that has occurred after the sign,
Extracting the sign associated with the failure from the plurality of signs that occurred before the failure based on the sign and the failure, and creating a sign pattern based on the extracted sign.
A step of storing and calculating the transition probabilities between sign and between sign fault in the sign pattern,
Calculating the cumulative failure probability from the sign to the failure when the first sign in the sign pattern occurs, based on the transition probability;
A step of calculating the priority of the equipment for which maintenance is inspected based on the cumulative failure probability for the plurality of equipment in which the sign has occurred ;
Including,
The step of creating the predictive pattern compares the rate of change of the failure probability between the predictive signs and a threshold value for determining the relationship with the failure, thereby determining the predictive sign related to the failure from a plurality of the predictive signs. A facility inspection order setting method characterized by extracting .

Images