Fault determination system and fault determination method

Abstract

In the present invention, an actual measurement data acquisition unit (101) acquires, as post-stoppage actual measurement data, time-series data of measurement values measured in a machine and / or a machine installation environment after the machine stops operating. An estimation data acquisition unit (102): applies a machine operating condition to each of a plurality of failure simulation models (106) corresponding respectively to a plurality of failure factors anticipated for the machine; and from each of the plurality of failure simulation models (106), acquires, as post-stoppage estimated data, time-series data of measurements values estimated to be measured in the machine and / or the machine installation environment after the machine stops operating in a case where a failure occurs in the machine due to each of the plurality of failure factors. A determination unit (103) compares each piece of the post-stoppage estimated data and the post-stoppage actual measurement data, and determines whether any of the plurality of failure factors has caused the failure in the machine.

Description

[Technical field]

[0001] The present disclosure relates to equipment fault diagnosis. [Background technology]

[0002] Patent Document 1 discloses a technique relating to fault diagnosis. In the technology of Patent Document 1, time series data representing the operating state of the equipment at the time of failure for each failure cause is accumulated in a database. In addition, in the technology of Patent Document 1, time series data representing the operating state of the equipment is collected while the equipment is in operation. Furthermore, in the technology of Patent Document 1, time series data approximating the collected time series data is extracted from the database to identify the failure cause. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2004-234302 A Summary of the Invention [Problem to be solved by the invention]

[0004] The technology of Patent Document 1 identifies the cause of a failure using time-series data acquired during operation of the equipment. However, depending on the cause of a failure, there may be cases where a characteristic pattern that allows identification of the cause of the failure does not appear in the time-series data acquired during operation of the equipment. The technique of Patent Document 1 has a problem in that it is not possible to effectively identify the cause of a failure that does not show a characteristic pattern in the time-series data acquired during the operation of such equipment.

[0005] The main objective of the present disclosure is to solve the above-mentioned problems. Specifically, the main objective of the present disclosure is to enable identification of a failure cause even if the failure cause does not show a characteristic pattern in time-series data acquired during operation of the equipment. [Means for solving the problem]

[0006] The fault determination system according to the present disclosure comprises: an actual measurement data acquisition unit that acquires time-series data of measurement values ​​measured in at least one of the device and an installation environment of the device after the device is stopped as actual measurement data after the device is stopped; an estimated data acquisition unit that applies operating conditions of the device to each of a plurality of fault simulation models corresponding to a plurality of fault factors assumed for the device, and acquires, from each of the plurality of fault simulation models, time-series data of measurement values ​​that are estimated to be measured in at least one of the device and an installation environment of the device after operation of the device is stopped, as post-stop estimated data, when a fault due to each of the plurality of fault factors occurs in the device; The device further includes a determination unit that compares each post-stop estimated data with the post-stop actual measured data and determines whether or not a failure due to any of the plurality of failure causes has occurred in the device. Effect of the Invention

[0007] According to the present disclosure, it is possible to identify the cause of a failure even if the cause of the failure does not show a characteristic pattern in the time-series data acquired during operation of the equipment. [Brief description of the drawings]

[0008] [Figure 1] FIG. 1 is a diagram showing an example of a system configuration according to a first embodiment. [Diagram 2] FIG. 2 is a diagram showing an example of a hardware configuration of the fault determination system according to the first embodiment. [Diagram 3] FIG. 2 is a diagram showing an example of a functional configuration of the failure determination system according to the first embodiment. [Figure 4] 4 is a flowchart showing an example of the operation of the failure determination system according to the first embodiment. [Diagram 5] 4 is a flowchart showing an example of the operation of the failure determination system according to the first embodiment. [Figure 6] FIG. 4 is a diagram showing an example of actual measurement data during a normal stop according to the first embodiment. [Figure 7] FIG. 4 is a diagram showing a comparison example between actual measurement data and estimated data during a normal stop according to the first embodiment. [Figure 8] FIG. 4 is a diagram showing a comparison example between actual measurement data and estimated data during a normal stop according to the first embodiment. [Figure 9] FIG. 4 is a diagram showing a comparison example between actual measurement data and estimated data during a normal stop according to the first embodiment. [Figure 10] FIG. 4 is a diagram showing an example of actual measurement data at the time of an abnormal stop according to the first embodiment. [Figure 11] FIG. 4 is a diagram showing a comparison example between actual measurement data and estimated data at the time of an abnormal stop according to the first embodiment. [Figure 12] FIG. 4 is a diagram showing a comparison example between actual measurement data and estimated data at the time of an abnormal stop according to the first embodiment. [Figure 13] FIG. 4 is a diagram showing a comparison example between actual measurement data and estimated data at the time of an abnormal stop according to the first embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] Hereinafter, the embodiments will be described with reference to the drawings. In the following description of the embodiments and in the drawings, the same reference numerals denote the same or corresponding parts.

[0010] Embodiment 1 ***Configuration Description*** FIG. 1 shows an example of a system configuration according to the present embodiment. In this embodiment, a system configured with a failure determination system 100, an air conditioner 200, and a sensor 300 will be described. In this embodiment, the description will be given using an air conditioner 200 as an example of the device. The device does not have to be the air conditioner 200.

[0011] The fault determination system 100 determines whether or not a fault has occurred in the air conditioner 200. Furthermore, if a fault has occurred in the air conditioner 200, the fault determination system 100 determines the cause of the fault. The operation procedure of the failure determination system 100 corresponds to a failure determination method.

[0012] The air conditioner 200 includes an indoor unit, an outdoor unit, and piping that connects the indoor unit and the outdoor unit.

[0013] The sensor 300 is installed in the air conditioner 200 . Although only one sensor 300 is shown in FIG. 1, it is assumed that multiple sensors 300 are present. Specifically, a plurality of sensors 300 are installed in each of the indoor unit, the outdoor unit, and the piping. For example, the sensors 300 are installed before and after the compressor. Furthermore, the sensors 300 are installed before and after the heat exchanger. Furthermore, the sensors 300 are installed before and after the LEV (linear expansion valve). The manager of the air conditioner 200 can install the sensors 300 at any position of the air conditioner 200. Furthermore, the sensors 300 may be installed in the installation environment of the air conditioner 200. For example, the sensors 300 may be installed in a room in which the indoor unit is installed. Furthermore, the sensors 300 may be installed in a building in which the outdoor unit is installed. Each sensor 300 measures a value that represents the state of the air conditioner 200 and / or the state of the environment in which the air conditioner is installed. For example, the sensor 300 measures the refrigerant temperature, refrigerant pressure, room temperature, etc. The value measured by each sensor 300 is called a measurement value. At least one of the multiple sensors 300 has a communication function and transmits time-series data of the measurement values ​​to the fault determination system 100. In the following, for the sake of simplicity, it is assumed that one sensor 300 transmits time-series data of measurement values ​​of a plurality of sensors 300 to the fault determination system 100. However, each sensor 300 may transmit time-series data of measurement values ​​to the fault determination system 100 individually.

[0014] FIG. 2 shows an example of a hardware configuration of the failure determination system 100 according to this embodiment. FIG. 3 shows an example of the functional configuration of the failure determination system 100. First, an example of the hardware configuration of the fault determination system 100 will be described with reference to FIG.

[0015] The fault determination system 100 according to this embodiment is a computer. The fault determination system 100 includes, as hardware, a processor 901, a main memory device 902, an auxiliary memory device 903, and a communication device 904. 3, the fault determination system 100 includes, as functional components, an actual measurement data acquisition unit 101, an estimated data acquisition unit 102, a determination unit 103, and a database 104. Of these, the functions of the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103 are realized by, for example, a program. The auxiliary storage device 903 stores programs for implementing the functions of the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103. These programs are loaded from the auxiliary storage device 903 to the main storage device 902. Then, the processor 901 executes these programs to perform the operations of the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103, which will be described later. FIG. 2 shows a schematic diagram of a state in which a processor 901 is executing a program that realizes the functions of the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103. In FIG. The database 104 is realized by, for example, the auxiliary storage device 903 . The input / output device 905 is, for example, a mouse, a keyboard, a display, a microphone, a speaker, and the like.

[0016] Next, an example of the functional configuration of the failure determination system 100 will be described with reference to FIG.

[0017] The actual measurement data acquisition unit 101 receives actual measurement data from the sensor 300 . The actual measurement data is time-series data of the measurement values ​​measured by the sensor 300. In this embodiment, the actual measurement data acquisition unit 101 receives pre-stop actual measurement data and post-stop actual measurement data as the actual measurement data. The pre-shutdown actual measurement data is time-series data before the operation of the air conditioner 200 is stopped. Specifically, the actual measurement data acquisition unit 101 receives, as pre-shutdown actual measurement data, time series data of measurement values ​​measured by the sensor 300 before normal operation of the air conditioner 200 is stopped. In addition, the actual measurement data acquisition unit 101 receives, as pre-shutdown actual measurement data, time series data of measurement values ​​measured by the sensor 300 before operation of the air conditioner 200 is stopped due to abnormality detection. The post-shutdown actual measurement data is time-series data after the operation of the air conditioner 200 has been stopped. Specifically, the actual measurement data acquisition unit 101 receives, as post-shutdown actual measurement data, time series data of measurement values ​​measured by the sensor 300 after normal operation of the air conditioner 200 has been stopped. In addition, the actual measurement data acquisition unit 101 receives, as post-shutdown actual measurement data, time series data of measurement values ​​measured by the sensor 300 after operation of the air conditioner 200 has been stopped due to abnormality detection. When there is no need to distinguish between the pre-stop actual measurement data and the post-stop actual measurement data, the pre-stop actual measurement data and the post-stop actual measurement data are collectively referred to as the actual measurement data.

[0018] The measured data includes, for example, temperature values, pressure values, and actuator signal values. Temperature values ​​may include, for example, refrigerant temperature values ​​before and after a compressor, refrigerant temperature values ​​before and after a heat exchanger, refrigerant temperature values ​​before and after a LEV, and the like. The pressure values ​​include, for example, refrigerant pressure values ​​before and after the compressor, refrigerant pressure values ​​before and after the heat exchanger, refrigerant pressure values ​​before and after the LEV, and the like. The signal values ​​of the actuators include signal values ​​of actuators installed in the indoor units, signal values ​​of actuators installed in the indoor units, and the like. Furthermore, the actual measurement data may include values ​​for the operation mode (cooling, heating, etc.) of the air conditioner 200, or may include other values. These values ​​of the actual measurement data are merely examples, and values ​​other than these may be included in the actual measurement data. The actual measurement data acquisition unit 101 outputs the actual measurement data to the determination unit 103 .

[0019] The estimated data acquisition unit 102 acquires the estimated data. The estimated data is time-series data of measurement values ​​estimated by each of the multiple fault simulation models 106. In this embodiment, there are fault simulation model A 106A, fault simulation model B 106B, fault simulation model C 106C, etc. When there is no need to distinguish between the fault simulation model A 106A, the fault simulation model B 106B, the fault simulation model C 106C, etc., they are collectively referred to as the fault simulation models 106. Each fault simulation model 106 corresponds to a different fault cause. The fault simulation model 106 will be described in detail later.

[0020] In this embodiment, the estimated data acquisition unit 102 acquires pre-stop estimated data and post-stop estimated data as the estimated data. The pre-shutdown estimated data is time-series data of measurement values ​​that are estimated to be measured by the air conditioner 200 before operation is stopped when a malfunction occurs in the air conditioner 200. More specifically, the estimated data acquisition unit 102 acquires, as pre-shutdown estimated data, time series data of measurement values ​​estimated to be measured by the sensor 300 before normal operation of the air conditioner 200 is stopped when a fault occurs in the air conditioner 200. Furthermore, the estimated data acquisition unit 102 acquires, as pre-shutdown estimated data, time series data of measurement values ​​estimated to be measured by the sensor 300 before operation of the air conditioner 200 is stopped due to abnormality detection when a fault occurs in the air conditioner 200. Moreover, the post-shutdown estimated data is time-series data of measurement values ​​estimated to be measured by the air conditioner 200 after operation is stopped in the event of a malfunction in the equipment. More specifically, the estimated data acquisition unit 102 acquires, as post-stop estimated data, time series data of measurement values ​​estimated to be measured by the sensor 300 after normal operation of the air conditioner 200 is stopped when a fault occurs in the air conditioner 200. Furthermore, the estimated data acquisition unit 102 acquires, as post-stop estimated data, time series data of measurement values ​​estimated to be measured by the sensor 300 after operation of the air conditioner 200 is stopped due to abnormality detection when a fault occurs in the air conditioner 200. The estimated data acquisition unit 102 acquires operating condition data 105 from the database 104, and applies the operating condition data 105 to each failure simulation model 106. Then, the estimated data acquisition unit 102 acquires pre-stop estimated data and post-stop estimated data from each failure simulation model 106. In the following, when there is no need to distinguish between pre-stop estimation data and post-stop estimation data, the pre-stop estimation data and the post-stop estimation data will be collectively referred to as estimation data.

[0021] The types of values ​​contained in the estimated data are the same as the types of values ​​contained in the actual measurement data. For example, if the measured data includes, as temperature values, refrigerant temperature values ​​before and after the compressor, refrigerant temperature values ​​before and after the heat exchanger, and refrigerant temperature values ​​before and after the LEV, the estimated data also includes these values. Similarly, if the measured data includes, as pressure values, refrigerant pressure values ​​before and after the compressor, refrigerant pressure values ​​before and after the heat exchanger, and refrigerant pressure values ​​before and after the LEV, the estimated data also includes these values. The estimated data acquisition unit 102 outputs the estimated data to the determination unit 103 .

[0022] The determination unit 103 compares each estimated data with the actual measurement data, and determines whether or not a fault has occurred in the air conditioner 200 due to any of a plurality of fault causes. Specifically, the determination unit 103 calculates the similarity between each piece of estimated data and the actual measurement data. Then, when there is estimated data whose similarity with the actual measurement data is equal to or greater than a threshold, the determination unit 103 determines that a fault has occurred in the air conditioner 200 due to a fault cause corresponding to the fault simulation model 106 from which the estimated data whose similarity is equal to or greater than the threshold has been obtained. The determination unit 103 outputs the determination result to, for example, a display of the input / output device 905 .

[0023] The database 104 stores operating condition data 105 and a number of fault simulation models 106 .

[0024] The operating condition data 105 indicates the operating conditions of the air conditioner 200. The operating conditions indicated in the operating condition data 105 include, for example, the model name of the indoor unit, the model name of the outdoor unit, the length of the piping connecting the indoor unit and the outdoor unit, the diameter of the piping, the elevation difference between the indoor unit and the outdoor unit, etc. The operating conditions may also include the size of the room in which the indoor unit is installed, heat capacity, wall area, room temperature, indoor humidity, outdoor temperature, outdoor humidity, amount of solar radiation, and the like. These operating conditions are merely examples, and other conditions may be included in the operating conditions.

[0025] The multiple fault simulation models 106 correspond to multiple different fault causes assumed in the air conditioner 200. For example, the fault simulation model A106A corresponds to fault cause A. The fault simulation model A106A generates, as estimated data, time-series data of measurement values ​​estimated to be measured by the sensor 300 when a fault due to fault cause A occurs in the air conditioner 200. Each fault simulation model 106 uses the operating condition data 105 to simulate a state in which a fault occurs in the air conditioner 200 and generates the estimated data. Specific causes of failure are expected to be indoor LEV valve seat wear, refrigerant shortage, outdoor fan stop, indoor LEV stuck open, etc. Indoor LEV valve seat wear is a state in which the indoor unit's LEV valve seat is worn. Refrigerant shortage is a state in which there is a shortage of refrigerant circulating through the indoor and outdoor units. Outdoor fan stop is a state in which the outdoor unit fan is stopped. Indoor LEV stuck open is a state in which the indoor unit's LEV is stuck open. The above failure causes are merely examples, and the failure simulation model 106 may be provided to accommodate failure causes other than these.

[0026] ***Explanation of Operation*** FIG. 4 shows an example of the operation of the failure determination system 100 when the air conditioner 200 has stopped normally after regular inspection, operation check, and the like. "The air conditioner 200 is stopped normally" means that no abnormality is detected in the air conditioner 200 and the air conditioner 200 is stopped as scheduled.

[0027] When the sensor 300 notifies the air conditioner 200 that it has stopped normally (YES in step S11), the actual measurement data acquisition unit 101 acquires actual measurement data from the sensor 300 in step S12. Here, the actual measurement data acquisition unit 101 acquires the actual measurement data after the stoppage. Then, the actual measurement data acquisition unit 101 outputs the post-stop actual measurement data to the determination unit 103 .

[0028] Next, in step S13, the estimated data acquisition unit 102 acquires the estimated data. Here, the estimation data acquisition unit 102 acquires post-stop estimation data from each fault simulation model 106 . More specifically, the estimation data acquisition unit 102 acquires the operating condition data 105 from the database 104. Then, the estimation data acquisition unit 102 applies the operating condition data 105 to each failure simulation model 106, and acquires post-stop estimation data from each failure simulation model 106. Then, the estimation data acquisition unit 102 outputs the post-stop estimation data acquired from each fault simulation model 106 to the judgment unit 103 .

[0029] Next, in step S14, the determination unit 103 calculates the similarity between each piece of estimated data and the actual measurement data. That is, the determination unit 103 calculates the similarity between each piece of post-stop estimated data and the post-stop actual measured data. For example, the determination unit 103 calculates the similarity using an existing technique such as DTW (Dynamic Time Warping).

[0030] If there is post-stop estimation data whose similarity is equal to or greater than the threshold (YES in step S15), in step S16, the determination unit 103 selects the post-stop estimation data whose similarity is highest. In other words, if there are two or more post-stop estimation data whose similarity is equal to or greater than the threshold, in step S16, the determination unit 103 selects the post-stop estimation data whose similarity is highest. Then, in step S17, the determining unit 103 determines that a failure has occurred due to the failure cause of the selected post-stop estimation data. For example, when the similarity of the post-stop estimation data generated by the fault simulation model A 106A is the highest, the determining unit 103 determines that the air conditioner 200 has experienced a fault due to the fault cause A corresponding to the fault simulation model A 106A.

[0031] On the other hand, if there is no post-shutdown estimation data with a similarity equal to or greater than the threshold (NO in step S15), in step S18, the determination unit 103 determines that no malfunction has occurred in the air conditioner 200.

[0032] Thereafter, the determination unit 103 outputs the determination result of step S17 or step S18 to the display of the input / output device 905, for example.

[0033] The operation procedure shown in FIG. 4 will be explained using specific examples shown in FIGS. FIG. 6 shows an example of actual measurement data when the air conditioner 200 stops operating normally. 7 to 9 show examples of estimated data that is estimated when the air conditioner 200 stops operating normally while a fault is occurring.

[0034] 6, "operating" indicates a period during which the air conditioner 200 is operating. "stopped" indicates a period during which the air conditioner 200 is stopped operating. The time series data of the measurement values ​​measured during the "operating" period in Fig. 6 corresponds to the actual measurement data before stopping. Also, the time series data of the measurement values ​​measured by the sensor 300 during the "stopped" period in Fig. 6 corresponds to the actual measurement data after stopping. As described above, when the air conditioner 200 is normally stopped, in step S12, the measured data acquisition unit 101 acquires the measured data after the stop from the sensor 300. Note that, for ease of understanding, Fig. 6 also illustrates the measured data before the stop. FIG. 6( a ) shows the temperature value measured by the sensor 300 . FIG. 6( b ) shows the LEV opening signal measured by the sensor 300 .

[0035] In FIG. 6(a), "room temperature" is the temperature of the room in which the indoor unit is installed. The "indoor gas temperature" is the temperature of the gaseous refrigerant in the indoor unit. The "indoor liquid temperature" is the temperature of the liquid refrigerant in the indoor unit.

[0036] Fig. 7 shows an example of estimated data generated by the failure simulation model 106 corresponding to the failure cause "indoor LEV valve seat wear". In other words, Fig. 7 shows an example of estimated data when the air conditioner 200 stops operating normally while "indoor LEV valve seat wear" is occurring. Fig. 8 shows an example of estimated data generated by the failure simulation model 106 corresponding to the failure cause "refrigerant shortage". In other words, Fig. 8 shows an example of estimated data when the air conditioner 200 stops operating normally while "refrigerant shortage" is occurring. Fig. 9 shows an example of estimated data generated by the fault simulation model 106 corresponding to the fault cause "outdoor fan stopped." In other words, Fig. 9 shows an example of estimated data when the air conditioner 200 stops operating normally in a state where "outdoor fan stopped" has occurred.

[0037] The measured data shown in FIG. 7(a), FIG. 8(a) and FIG. 9(a) are similar to that shown in FIG. 6(a). The measured data shown in FIG. 7(b), FIG. 8(b) and FIG. 9(b) are similar to that shown in FIG. 6(b). Figure 7(c) shows temperature values ​​estimated to be measured when "indoor LEV valve seat wear" occurs in the air conditioner 200. The estimated data in Figure 7(c) corresponds to the actual measurement data in Figure 6(a). Fig. 7(d) shows the LEV opening signal estimated to be measured when "indoor LEV valve seat wear" occurs in the air conditioner 200. The estimated data in Fig. 7(d) corresponds to the actual measurement data in Fig. 6(b). Figure 8(c) shows temperature values ​​estimated to be measured when a "refrigerant shortage" occurs in the air conditioner 200. The estimated data in Figure 8(c) corresponds to the actual measurement data in Figure 6(a). Fig. 8(d) shows the LEV opening signal estimated to be measured when a "refrigerant shortage" occurs in the air conditioner 200. The estimated data in Fig. 8(d) corresponds to the actually measured data in Fig. 6(b). Figure 9(c) shows temperature values ​​estimated to be measured when "outdoor fan stopped" occurs in the air conditioner 200. The estimated data in Figure 9(c) corresponds to the actual measurement data in Figure 6(a). Figure 9(d) shows the LEV opening signal estimated to be measured when "outdoor fan stop" occurs in the air conditioner 200. The estimated data in Figure 9(d) corresponds to the actual measurement data in Figure 6(b). In step S13, the estimation data acquisition unit 102 acquires the post-stop estimation data illustrated in FIGS. As in FIG. 6, in order to facilitate understanding, pre-stop estimation data is also shown in FIG. 7 to FIG. The fault causes shown in Figs. 7 to 9 are fault causes for which no characteristic pattern that allows the fault cause to be identified appears in the time-series data acquired during operation of the air conditioner 200 (actual measurement data before shutdown).

[0038] In step S14, the determination unit 103 calculates the similarity between each of the post-stop estimated data and the post-stop actual measured data. That is, the determination unit 103 calculates the similarity between each of the post-stop estimated data in FIGS. 7 to 9 and the post-stop actual measured data in FIG. The waveform after stop in FIG. 7(c) completely matches the waveform after stop in FIG. 6(a) (FIG. 7(a)). Also, the waveform after stop in FIG. 7(d) completely matches the waveform after stop in FIG. 6(b) (FIG. 7(b)). Therefore, for the post-stop estimated data in FIG. 7, the determination unit 103 obtains a "similarity: 100%." The waveform after stop in FIG. 8(c) does not match the waveform after stop in FIG. 6(a) (FIG. 8(a)). On the other hand, the waveform after stop in FIG. 8(d) completely matches the waveform after stop in FIG. 6(b) (FIG. 8(b)). For this reason, for the post-stop estimated data in FIG. 8, the determination unit 103 obtains a "similarity: 50%". The waveform after stop in FIG. 9(c) does not match the waveform after stop in FIG. 6(a) (FIG. 9(a)). On the other hand, the waveform after stop in FIG. 9(d) completely matches the waveform after stop in FIG. 6(b) (FIG. 9(b)). For this reason, for the post-stop estimated data in FIG. 9, the judgment unit 103 obtains a "similarity: 50%".

[0039] From the above, in step S17, the determination unit 103 determines that "indoor LEV valve seat wear" has occurred in the air conditioner 200. Then, the determination unit 103 outputs a determination result that "indoor LEV valve seat wear" has occurred in the air conditioner 200.

[0040] In this way, the failure judgment system 100 of this embodiment can identify the failure cause even if the failure cause does not show a characteristic pattern in the time series data acquired during operation of the air conditioner 200, by comparing the actual post-shutdown data with multiple post-shutdown estimated data corresponding to multiple failure causes.

[0041] 4 and 6 to 9, an example has been described in which the failure determination system 100 performs analysis using only the post-stop actual measurement data and the post-stop estimated data to identify the cause of a failure of the air conditioner 200. The failure determination system 100 may also identify the cause of a failure of the air conditioner 200 by performing analysis using the pre-stop actual measurement data and the pre-stop estimated data in addition to the post-stop actual measurement data and the post-stop estimated data.

[0042] Next, an example of the operation of the failure determination system 100 when the air conditioner 200 is stopped due to the detection of an abnormality will be shown with reference to FIG. "The air conditioner 200 is stopped due to detection of an abnormality" means that an abnormality is detected in the air conditioner 200, and the air conditioner 200 is stopped urgently.

[0043] For example, when a notification is received from the sensor 300 that the air conditioner 200 has stopped due to detection of an abnormality (YES in step S21), the actual measurement data acquisition unit 101 acquires actual measurement data from the sensor 300 in step S12. Here, the actual measurement data acquisition unit 101 acquires pre-stop actual measurement data and post-stop actual measurement data. Then, the actual measurement data acquisition unit 101 outputs the pre-stop actual measurement data and the post-stop actual measurement data to the determination unit 103.

[0044] Next, in step S13, the estimated data acquisition unit 102 acquires the estimated data. Here, the estimation data acquisition unit 102 acquires pre-shutdown estimation data and post-shutdown estimation data from each failure simulation model 106 . More specifically, the estimation data acquisition unit 102 acquires the operating condition data 105 from the database 104. Then, the estimation data acquisition unit 102 applies the operating condition data 105 to each failure simulation model 106, and acquires pre-stop estimation data and post-stop estimation data from each failure simulation model 106. Then, the estimation data acquisition unit 102 outputs the pre-stop estimation data and the post-stop estimation data acquired from each failure simulation model 106 to the determination unit 103.

[0045] Next, in step S14, the determination unit 103 calculates the similarity between each piece of estimated data and the actual measurement data. That is, the determination unit 103 calculates the similarity between each pair of pre-stop estimated data and post-stop estimated data and a pair of pre-stop actual measured data and post-stop actual measured data. For example, the determination unit 103 calculates the similarity using an existing technique such as DTW.

[0046] If there is estimated data whose similarity is equal to or greater than the threshold value (YES in step S15), in step S16, the decision unit 103 selects the estimated data whose similarity is the highest. Then, in step S17, the determining unit 103 determines that a failure has occurred due to the failure cause of the selected estimated data.

[0047] On the other hand, if there is no estimated data with a similarity equal to or greater than the threshold (NO in step S15), in step S22, the determining unit 103 determines that an abnormality has occurred due to another fault cause, that is, an unanticipated fault cause.

[0048] Thereafter, the determination unit 103 outputs the determination result of step S17 or step S22 to a display of the input / output device 905, for example.

[0049] The operation procedure shown in FIG. 5 will be explained using a specific example shown in FIGS. FIG. 10 shows an example of actual measurement data when the air conditioner 200 stops operating due to detection of an abnormality. 11 to 13 show examples of estimated data that is estimated when the air conditioner 200 stops operation due to abnormality detection in a state in which a malfunction has occurred.

[0050] 10, elements common to those in FIG. 6 will not be described. As described above, when the air conditioner 200 is stopped due to detection of an abnormality, the actual measurement data acquisition unit 101 acquires actual measurement data (actual measurement data before shutdown and actual measurement data after shutdown) from the sensor 300 in step S12. FIG. 10( a ) shows pressure and temperature values ​​measured by the sensor 300 . FIG. 10( b ) shows the LEV opening signal measured by the sensor 300 .

[0051] In FIG. 10(a), "discharge temperature" is the discharge temperature of the refrigerant at the compressor. "high pressure" is the discharge pressure of the refrigerant at the compressor. "low pressure" is the suction pressure of the refrigerant at the compressor.

[0052] Fig. 11 shows an example of estimated data generated by the fault simulation model 106 corresponding to the fault cause "indoor LEV stuck open". In other words, Fig. 11 shows an example of estimated data when the air conditioner 200 stops operation due to abnormality detection in a state where "indoor LEV stuck open" has occurred. Fig. 12 shows an example of estimated data generated by the failure simulation model 106 corresponding to the failure cause "refrigerant shortage." In other words, Fig. 12 shows an example of estimated data when the air conditioner 200 stops operation due to abnormality detection in a state where "refrigerant shortage" is occurring. Fig. 13 shows an example of estimated data generated by the fault simulation model 106 corresponding to the fault cause "outdoor fan stopped." In other words, Fig. 13 shows an example of estimated data when the air conditioner 200 stops operation due to abnormality detection in a state where "outdoor fan stopped" has occurred. The measured data shown in FIG. 11(a), FIG. 12(a) and FIG. 13(a) are similar to that shown in FIG. 10(a). The measured data shown in FIG. 11(b), FIG. 12(b) and FIG. 13(b) are similar to that shown in FIG. 10(b). Figure 11(c) shows pressure values ​​and temperature values ​​estimated to be measured when the "indoor LEV stuck open" state occurs in the air conditioner 200. The estimated data in Figure 11(c) corresponds to the actual measurement data in Figure 10(a). Figure 11(d) shows the LEV opening signal estimated to be measured when the "indoor LEV stuck open" occurs in the air conditioner 200. The estimated data in Figure 11(d) corresponds to the actual measurement data in Figure 10(b). Figure 12(c) shows pressure and temperature values ​​estimated to be measured when a "refrigerant shortage" occurs in the air conditioner 200. The estimated data in Figure 12(c) corresponds to the actual measurement data in Figure 10(a). Figure 12(d) shows the LEV opening signal estimated to be measured when a "refrigerant shortage" occurs in the air conditioner 200. The estimated data in Figure 12(d) corresponds to the actually measured data in Figure 10(b). Figure 13(c) shows pressure values ​​and temperature values ​​estimated to be measured when "outdoor fan stop" occurs in the air conditioner 200. The estimated data in Figure 13(c) corresponds to the actual measurement data in Figure 10(a). Figure 13(d) shows the LEV opening signal estimated to be measured when "outdoor fan stop" occurs in the air conditioner 200. The estimated data in Figure 13(d) corresponds to the actual measurement data in Figure 10(b). In step S13, the estimated data acquisition unit 102 acquires, from each failure simulation model 106, the estimated data (pre-stop estimated data and post-stop estimated data) exemplified in FIGS.

[0053] In step S14, the determination unit 103 calculates the similarity between each of the estimated data and the actual measurement data. That is, the determination unit 103 calculates the similarity between each of the estimated data in FIGS. 11 to 13 and the actually measured data in FIG. The waveform in FIG. 11(c) completely matches the waveform in FIG. 10(a) (FIG. 11(a)). Also, the waveform in FIG. 11(d) completely matches the waveform in FIG. 10(b) (FIG. 11(b)). Therefore, for the estimated data in FIG. 11, the determination unit 103 obtains a "similarity: 100%." The "high temperature" waveform in FIG. 12(c) does not match the "high temperature" waveform in FIG. 10(a) (FIG. 12(a)). On the other hand, the waveform in FIG. 12(d) completely matches the waveform in FIG. 10(b) (FIG. 12(b)). For this reason, the judgment unit 103 obtains a "similarity: 75%" for the estimated data in FIG. 12. The "low temperature" waveform in FIG. 13(c) does not match the "low temperature" waveform in FIG. 10(a) (FIG. 13(a)). Also, the waveform in FIG. 13(d) does not match the waveform in FIG. 10(b) (FIG. 13(b)). For this reason, for the estimated data in FIG. 13, the determination unit 103 obtains a "similarity: 50%."

[0054] From the above, in step S17, the determination unit 103 determines that the "indoor LEV fixed open" has occurred in the air conditioner 200. Then, the determination unit 103 outputs a determination result that the "indoor LEV fixed open" has occurred in the air conditioner 200.

[0055] In the explanation using Figures 5 and 10 to 13, an example has been described in which the failure judgment system 100 identifies the cause of a failure of the air conditioner 200 by performing analysis using actual measured data before shutdown and actual measured data after shutdown and estimated data before shutdown and estimated data after shutdown. Even if the air conditioner 200 stops operating due to the detection of an abnormality, depending on the cause of the failure, the failure determination system 100 may identify the cause of the failure of the air conditioner 200 by performing an analysis using only the actual measured data after the operation was stopped and the estimated data after the operation was stopped.

[0056] ***Description of the Effects of the Embodiment*** In this manner, in this embodiment, the post-stop actual measurement data is compared with a plurality of post-stop estimated data corresponding to a plurality of failure causes, as described with reference to Figures 4, 6 to 9. Therefore, according to this embodiment, even if a failure cause does not show a characteristic pattern that allows the failure cause to be identified in the time-series data acquired during the operation of the equipment, it is possible to identify the failure cause.

[0057] Also, only a part of the procedure described in this embodiment may be performed. Furthermore, at least a part of the procedure described in this embodiment may be combined with a procedure not described in this embodiment. Furthermore, the configuration and procedures described in this embodiment may be modified as necessary.

[0058] ***Additional hardware configuration information*** Finally, a supplementary explanation of the hardware configuration of the failure determination system 100 will be given. The processor 901 shown in FIG. 2 is an integrated circuit (IC) that performs processing. The processor 901 is a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or the like. The main storage device 902 shown in FIG. 2 is a RAM (Random Access Memory). The auxiliary storage device 903 shown in FIG. 2 is a ROM (Read Only Memory), a flash memory, a HDD (Hard Disk Drive), or the like. The communication device 904 shown in FIG. 2 is an electronic circuit that performs data communication processing. The communication device 904 is, for example, a communication chip or a NIC (Network Interface Card).

[0059] The auxiliary storage device 903 also stores an OS (Operating System). At least a part of the OS is executed by the processor 901 . The processor 901 executes programs that realize the functions of the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103 while executing at least a part of the OS. The processor 901 executes the OS, which performs task management, memory management, file management, communication control, and the like. In addition, at least one of information, data, signal values, and variable values ​​indicating the results of processing by the actual data acquisition unit 101, the estimated data acquisition unit 102, and the judgment unit 103 is stored in at least one of the main memory device 902, the auxiliary memory device 903, a register within the processor 901, and a cache memory. Furthermore, the programs for realizing the functions of the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103 may be stored in a portable recording medium such as a magnetic disk, a flexible disk, an optical disk, a compact disk, a Blu-ray (registered trademark) disk, a DVD, etc. Then, the portable recording medium storing the programs for realizing the functions of the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103 may be distributed.

[0060] Furthermore, the "parts" in the actual measurement data acquisition part 101, the estimated data acquisition part 102, and the determination part 103 may be read as "circuits" or "steps" or "procedures" or "processing" or "circuitry". Furthermore, the fault determination system 100 may be realized by a processing circuit. The processing circuit is, for example, a logic integrated circuit (IC), a gate array (GA), an application specific integrated circuit (ASIC), or a field-programmable gate array (FPGA). In this case, the actual measurement data acquisition unit 101, the estimated data acquisition unit 102, and the determination unit 103 are each realized as part of a processing circuit. In this specification, the higher-level concept of a processor and a processing circuit is called "processing circuitry." That is, a processor and a processing circuit are each specific examples of "processing circuitry." [Explanation of symbols]

[0061] 100 fault judgment system, 101 actual measurement data acquisition unit, 102 estimated data acquisition unit, 103 judgment unit, 104 database, 105 operating condition data, 106 fault simulation model, 106A fault simulation model A, 106B fault simulation model B, 106C fault simulation model C, 200 air conditioner, 300 sensor, 901 processor, 902 main memory device, 903 auxiliary memory device, 904 communication device, 905 input / output device.

Claims

1. an actual measurement data acquisition unit that acquires time-series data of measurement values ​​measured in at least one of the device and an installation environment of the device after the device is stopped as actual measurement data after the device is stopped; an estimated data acquisition unit that applies operating conditions of the device to each of a plurality of fault simulation models corresponding to a plurality of fault factors assumed for the device, and acquires, from each of the plurality of fault simulation models, time-series data of measurement values ​​that are estimated to be measured in at least one of the device and an installation environment of the device after operation of the device is stopped, as post-stop estimated data, when a fault due to each of the plurality of fault factors occurs in the device; A failure determination system having a determination unit that compares each post-stop estimated data with the post-stop actual measured data and determines whether or not a failure due to any of the multiple failure causes has occurred in the equipment.

2. The determination unit is Calculating a similarity between each of the post-stop estimated data and the post-stop actual measured data; 2. The fault determination system according to claim 1, wherein, when post-stop estimated data exists whose similarity to the post-stop actual measured data is equal to or greater than a threshold, it is determined that a fault has occurred in the equipment due to a fault cause corresponding to the fault simulation model from which the post-stop estimated data whose similarity is equal to or greater than the threshold was obtained.

3. The determination unit is 3. The fault determination system according to claim 2, wherein, when there are two or more pieces of post-stop estimation data whose similarity to the post-stop actual measurement data is equal to or greater than the threshold value, it is determined that a fault has occurred in the equipment due to a fault cause corresponding to a fault simulation model from which the post-stop estimation data with the highest similarity among the two or more pieces of post-stop estimation data has been obtained.

4. The actual measurement data acquisition unit acquiring time series data of measurement values ​​measured in at least one of the device and an installation environment of the device after normal operation of the device is stopped as actual measurement data after the device is stopped; The estimated data acquisition unit 2. The fault determination system according to claim 1, further comprising: acquiring, from each of the plurality of fault simulation models, time series data of measurement values ​​estimated to be measured in at least one of the equipment and an installation environment of the equipment after normal operation of the equipment is stopped when a fault due to each of the plurality of fault causes occurs in the equipment, as the post-stop estimated data.

5. The actual measurement data acquisition unit acquiring time-series data of measured values ​​measured in at least one of the equipment and an installation environment of the equipment after the operation of the equipment is stopped due to detection of an abnormality in the equipment as actual measurement data after the operation is stopped; The estimated data acquisition unit 2. The fault determination system according to claim 1, further comprising: acquiring, from each of the plurality of fault simulation models, time series data of measurement values ​​estimated to be measured in at least one of the equipment and an installation environment of the equipment after operation is stopped due to detection of an abnormality in the equipment when a fault due to each of the plurality of fault causes occurs in the equipment, as the post-stop estimated data.

6. The actual measurement data acquisition unit Obtaining time series data of measurement values ​​measured in at least one of the equipment and an installation environment of the equipment before operation is stopped due to abnormality detection of the equipment as actual measurement data before operation is stopped; The estimated data acquisition unit From each of the plurality of fault simulation models, time series data of measurement values ​​estimated to be measured in at least one of the equipment and an installation environment of the equipment before operation is stopped due to abnormality detection of the equipment when a fault due to each of the plurality of fault factors occurs in the equipment is acquired as pre-stop estimation data; The determination unit is 6. The fault determination system according to claim 5, wherein for each fault simulation model, the pre-shutdown estimated data and the post-shutdown estimated data are compared with the pre-shutdown actual measurement data and the post-shutdown actual measurement data, and whether or not a fault has occurred in the equipment due to any of the plurality of fault causes is determined.

7. The computer acquires time-series data of measurement values ​​measured in at least one of the device and the installation environment of the device after the device is stopped as actual measurement data after the device is stopped; The computer applies operating conditions of the equipment to each of a plurality of failure simulation models corresponding to a plurality of failure factors assumed for the equipment, and acquires, from each of the plurality of failure simulation models, time series data of measurement values ​​estimated to be measured in at least one of the equipment and an installation environment of the equipment after operation of the equipment is stopped when a failure due to each of the plurality of failure factors occurs in the equipment, as post-stop estimated data; A failure determination method in which the computer compares each post-stop estimated data with the post-stop actual measured data to determine whether or not a failure has occurred in the equipment due to any of the multiple failure causes.

Images