Prediction system, prediction method, and program

Abstract

In order to be able to predict an abnormality in a drain pump provided in an air conditioner with higher accuracy, a prediction system is provided with an air conditioner provided with a drain pump, and a control unit that acquires data on the number of rotations of the drain pump or the current value of the drain pump, and determines the abnormality of the drain pump on the basis of a change in the data for a specified period of time. And outputting a prediction result for predicting the abnormality of the drainage pump.

Description

Technical Field

[0001] The present invention relates to a prediction system, a prediction method and a program. Background Art

[0002] In an air conditioner (air conditioner) equipped with a drain pump, a technology for determining the degree of contamination of the drain pump is known. For example, the following technology is known, which detects the current value or the rotation speed of the drain pump, and when the current value of the drain pump is above a specified value or the rotation speed of the drain pump is below a specified value, determines the degree of contamination of the drain pump that requires repair (for example, refer to Patent Document 1).

[0003] [Prior art literature]

[0004] [Patent Literature]

[0005] [Patent Document 1] Japanese Patent Application Publication No. 2005-283057 Summary of the invention

[0006] [Problems to be solved by the invention]

[0007] However, there is a problem that since the current value (or rotation speed) of the drainage pump varies greatly among drainage pumps, false detection is prone to occur if only the instantaneous value of the current value (or rotation speed) is used to determine abnormality as in the prior art.

[0008] The present invention can predict an abnormality of a drain pump provided in an air conditioner with higher accuracy.

[0009]

Methods to solve the problem

[0010] The prediction system involved in the first embodiment of the present invention is a prediction system having an air conditioner with a drain pump and a control unit, wherein the control unit obtains data on the number of rotations of the drain pump or the current value of the drain pump, and outputs a prediction result of predicting an abnormality of the drain pump based on changes in the data in a specified time period.

[0011] According to the first aspect of the present invention, abnormality of the drain pump provided in the air conditioner can be predicted with higher accuracy.

[0012] A second aspect of the present invention is the prediction system according to the first aspect, wherein the control unit outputs the prediction result based on a representative value of the data in the designated time period.

[0013] The third aspect of the present invention is a prediction system according to the first aspect or the second aspect, wherein the control unit predicts abnormality of the drainage pump based on an average value of the data of a first specified time period and the data of a second specified time period shorter than the first specified time period.

[0014] The fourth mode of the present invention is a prediction system according to the third mode, wherein the control unit outputs the prediction result predicting the abnormality of the drainage pump when the deviation between the average value of the data in the first specified time period and the average value of the data in the second specified time period exceeds a threshold.

[0015] A fifth aspect of the present invention is the prediction system according to any one of the first to fourth aspects, wherein the control unit further predicts abnormality of the drain pump based on environmental data including temperature data or humidity data. Thus, the prediction system can predict abnormality of the drain pump with higher accuracy.

[0016] A sixth aspect of the present invention is a prediction system according to any one of the first to fifth aspects, wherein when the air conditioner or the drain pump is not operating, the control unit does not use the data of the time period when the air conditioner or the drain pump is not operating, but uses the data before the time period to predict the abnormality of the drain pump. Thus, the prediction system can predict the abnormality of the drain pump with higher accuracy.

[0017] A seventh aspect of the present invention is the prediction system according to the sixth aspect, wherein the control unit determines a time period during which the drain pump is not operating based on environmental data including temperature data or humidity data.

[0018] The eighth mode of the present invention is a prediction system according to any one of the first to fifth modes, wherein the control unit determines whether drainage occurs based on environmental data including temperature data or humidity data, removes the data of the number of rotations of the drainage pump or the current value of the drainage pump during a time period when drainage does not occur, and predicts abnormality of the drainage pump.

[0019] A ninth aspect of the present invention is the prediction system according to the eighth aspect, wherein the control unit further determines whether the drainage has occurred using information indicating whether the air conditioner is operating in a designated mode.

[0020] The tenth mode of the present invention is a prediction system according to any one of the first to ninth modes, wherein the prediction system has an edge device that collects the data from the air conditioner, and the control unit obtains the data averaged in the air conditioner or the edge device.

[0021] The 11th method of the present invention is a prediction system according to the 1st method, wherein the control unit uses the following model to predict the abnormality of the drainage pump, and uses the data when the drainage pump is normal and the data when the drainage pump is abnormal as training data to perform machine learning to obtain a prediction model.

[0022] The 12th method of the present invention is to use the following model to predict the abnormality of the drainage pump based on the prediction system recorded in the 1st method, and use the image representing the data when the drainage pump is normal and the image representing the data when the drainage pump is abnormal as training data to perform machine learning to obtain a prediction model that has been learned.

[0023] The prediction method involved in the 13th embodiment of the present invention is a prediction system having an air conditioner with a drain pump and a control unit, wherein the control unit obtains data on the number of rotations of the drain pump or the current value of the drain pump, and outputs a prediction result of predicting an abnormality of the drain pump based on changes in the data in a specified time period.

[0024] The program involved in the 14th embodiment of the present invention causes a computer to execute, in a prediction system having an air conditioner with a drain pump and a control unit, the following steps: obtaining data on the number of rotations of the drain pump or the current value of the drain pump, and outputting a prediction result of predicting an abnormality of the drain pump based on changes in the data during a specified time period.

[0025] Another aspect of the present invention is achieved by a recording medium recording the program according to the fourteenth aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0026]

Figure 1

[0027]

Figure 2

[0028]

Figure 3

[0029]

Figure 4

[0030]

Figure 5

[0031]

Figure 6

[0032]

Figure 7

[0033]

Figure 8

[0034]

Fig. 9

[0035]

Fig.10

[0036] In the following, each embodiment will be described with reference to the drawings. In addition, in this specification and the drawings, the same reference numerals are given to components having substantially the same functional structure, and repeated descriptions are omitted.

[0037] <System Structure>

[0038] Figure 1 1 is a diagram showing an example of a system configuration of a prediction system according to an embodiment. The prediction system 1 includes an air conditioner (air conditioner) 10 having a drain pump 11 and a control unit 101. The prediction system 1 is a system in which the control unit 101 obtains data on the number of revolutions (or current value) of the drain pump 11 and predicts an abnormality of the drain pump based on changes in the data in a specified time period.

[0039] Figure 1 In the example of FIG. 1 , the prediction system 1 includes a local controller 20, which is communicatively connected to the air conditioner 10 via a designated communication interface to control the air conditioner 10. In addition, the prediction system 1 includes a prediction server 100, which includes a control unit 101 and can communicate with the local controller 20 via a communication network 2 such as the Internet or a LAN (local area network).

[0040] In addition, if Figure 1 The system configuration of the prediction system 1 shown is an example. For example, the control unit 101 may include the local controller 20. In addition, the prediction server 100 may be configured by a plurality of information processing devices. Here, as an example, the prediction server 100 includes the control unit 101, and the following description is given.

[0041] The air conditioner 10 is configured such that, for example, during cooling operation, dew (drain water) is generated in a heat exchanger provided in the air conditioner 10 , and the generated dew water is stored in a container called a drain pan.

[0042] The drain pump 11 is a pump that sucks dew stored in the drain pan and discharges it to the outside through the drain pipe. When the drain pump fails, dew accumulates inside the air conditioner 10. When the amount of dew exceeds the allowable range, the air conditioner 10 detects the failure (drain pump failure) and stops operating.

[0043] When the drain pump fails, the user cannot operate the cold room, so the prediction system 1 outputs a prediction result for predicting the abnormality of the drain pump in advance so that, for example, a service person in charge can handle it on site before the drain pump 11 is completely blocked.

[0044] The local controller 20 has, for example, a computer structure, and controls the air conditioner 10 by executing a specified program recorded (stored) in a recording medium. In addition, the local controller 20 involved in the present embodiment has a function of acquiring measurement data including the number of rotations of the drain pump 11 or the current value of the drain pump 11 measured by the air conditioner 10 periodically (for example, every 1 minute) at each specified time (for example, every 1 hour). Furthermore, the local controller 20 has a function of sending the following data to the prediction server 100 through the communication network 2, and averaging the measurement data (first measurement data) obtained from the air conditioner 10 at each specified time (second measurement data). In addition, the local controller 20 is an example of an edge device.

[0045] As another example, the air conditioner 10 averages the measurement data at each designated time, and the local controller 20 may obtain the averaged measurement data from the air conditioner 10. In this case, the first measurement data obtained by the local controller 20 from the air conditioner 10 and the second measurement data sent by the local controller 20 to the prediction server 100 may be the same data.

[0046] The prediction server 100 has a computer configuration, and executes a predetermined program recorded (stored) in a recording medium, thereby executing a prediction process for predicting abnormality of the drainage pump 11 .

[0047] (Overview of prediction processing)

[0048] Figure 2 , 3 FIG. 1 is a schematic diagram for explaining a prediction process according to an embodiment of the present invention. Figure 2 The horizontal axis of the graph 200 shown represents the number of days that have passed since the start of measurement, and the vertical axis represents the number of revolutions of the drainage pump 11. In addition, the graph 200 plots measurement data (raw data) 201 of the number of revolutions of the drainage pump 11, a cumulative average 202 of the measurement data 201, and a three-day moving average 203 of the measurement data 201.

[0049] The rotation speed of the drain pump can be determined by measuring the number of pulses, for example, and the rotation speed can be obtained from the measured number of pulses. In this case, if one pulse is 24 rpm, for example, the rotation speed is measured at intervals of 24 rpm. In addition, in the rotation speed measurement data 201, for example Figure 2 As shown, a measurement deviation 204 of ±48 rpm (±2 pulses) is included.

[0050] Thus, the measured data 201 of the rotation speed of the drain pump 11 has large deviations. In addition, there are deviations (individual differences) among the drain pumps 11. Therefore, as in the prior art, there is a problem that false detection is easily caused by judging abnormality only by the instantaneous value of the rotation speed.

[0051] Therefore, the control unit 101 involved in this embodiment, for example, the local controller 20 stores (accumulates) the measurement data 201 sent every hour in the storage unit. Here, the storage unit storing the measurement data 201 can be, for example, a memory device possessed by the prediction server 100, or a memory server external to the prediction server 100.

[0052] In addition, the control unit 101 obtains the cumulative average 202 of the measurement data 201 stored in the storage unit and the 3-day moving average 203 of the measurement data 201. Here, the 3-day moving average 203 is the average value (moving average) of the measurement data 201 of the most recent 3 days. In addition, 3 days is an example of a designated time period (second designated time period) for abnormal prediction of the drainage pump 11. The second designated time period may be other days (for example, about 1 to 5 days) other than 3 days.

[0053] In addition, the cumulative average 202 is, for example, the average value of the measurement data 201 of the first specified time period which is sufficiently longer than the second specified time period. The cumulative average 202 may be, for example, the average value (moving average value) of the measurement data 201 of the most recent 30 days, or the cumulative moving average value from the time when the measurement is started, etc. Here, the cumulative average 202 is made to be the average value of the measurement data 201 of 30 days, and the following description is given.

[0054] When the drain pump 11 tends to be blocked, Figure 2 As shown, the rotation speed of the drain pump gradually decreases, and the value of the 3-day moving average 203 decreases steadily. On the other hand, the cumulative average 202 is an average value of a period of time sufficiently longer than 3 days, and therefore has less change in value than the 3-day moving average 203 .

[0055] Therefore, the control unit 101 according to the present embodiment outputs a prediction result of predicting abnormality of the drainage pump 11 based on the cumulative average 202 (the data average value of the first designated time period) and the 3-day moving average 203 (the data average value of the second designated time period shorter than the first designated time period). For example, the control unit 101 outputs a prediction result of predicting abnormality of the drainage pump 11 when the deviation between the cumulative average 202 (the data average value of the first designated time period) and the 3-day moving average 203 (the data average value of the second designated time period) exceeds a threshold value.

[0056] As a specific example, the control unit 101 Figure 3As shown, the difference 301 between the three-day moving average 203 and the cumulative average 202 is calculated, and when the calculated difference 301 exceeds a threshold 302, a prediction result predicting an abnormality of the drainage pump 11 is output. Here, the threshold 302 is predetermined, for example, by outputting a prediction result several days before an abnormality occurs in the drainage pump 11.

[0057] Through the above-described processing, according to the prediction system 1 according to the present embodiment, it is possible to predict abnormality of the drain pump 11 included in the air conditioner 10 with higher accuracy.

[0058] In the above description, the rotation speed of the drain pump 11 (the rotation speed of the motor included in the drain pump) is used as the measurement data 201 , but the measurement data 201 may be the current value of the drain pump 11 (the current value flowing in the drain pump 11 ).

[0059] In addition, when the drain pump 11 tends to be clogged, the rotation speed of the drain pump 11 decreases and the current value of the drain pump 11 increases. In either case, the control unit 101 can output a prediction result predicting an abnormality of the drain pump 11 when the deviation between the 3-day moving average 203 of the measurement data 201 and the cumulative average 202 of the measurement data 201 exceeds a threshold.

[0060] In addition, the prediction result of the abnormality of the drain pump 11 can be notified to the air conditioner 10 or the local controller 20, for example, and a specified display can be made, or it can be notified to the manager who manages the air conditioner 10. In addition, the prediction result can be, for example, a message such as "The drain pump is dirty, so please check it." It can also be information indicating the time before the drain pump 11 becomes abnormal.

[0061] <Hardware Structure>

[0062] The prediction server 100 and the local controller 20 have, for example, Figure 4 The hardware structure of the computer 400 is shown. In addition, the prediction server 100 may be composed of a plurality of computers 400.

[0063] Figure 4 The computer 400 includes, for example, a control unit 101, a memory 401, a storage device 402, a communication device 403, a display device 404, an input device 405, a drive device 406, and a bus 408.

[0064] The control unit 101 is, for example, a processor such as a CPU (central processing unit) that implements various functions by executing a specified program stored in a storage medium (recording medium) such as the memory device 402 or the memory 401. In addition, the control unit 101 may include a processor such as a GPU (graphics processing unit) or a DSP (digital signal processor) in addition to the CPU. In addition, the control unit 101 may be a device such as an ASIC (application integrated circuit) or an FPGA (field programmable gate array).

[0065] The memory 401 includes, for example, a volatile memory RAM (Random Access Memory) used as a work area of ​​the control unit 101 and a non-volatile memory ROM (Read Only Memory) storing a startup program of the control unit 101. The memory device 402 is a large-capacity storage device that stores programs such as an OS (operating system) and application programs, various data, information, etc., and is implemented by, for example, an SSD (Solid State Drive) or an HDD (Hard Disk Drive).

[0066] The communication device 403 includes one or more communication interfaces or communication devices for communicating with external devices. For example, the communication device 403 includes a NIC (network card) for connecting the computer 400 to the communication network 2 and communicating with other devices. In addition, the communication device 403 may also include, for example, a communication interface for connecting the air conditioner 10 to the computer 400.

[0067] The display device 404 is a display device or device that displays a display screen. The input device 405 is an input device that receives input from the outside, such as a keyboard, a pointing device, or a touch screen. In addition, the display device 404 and the input device 405 can be an integrated display input device such as a touch screen display.

[0068] The drive device 406 is a device for connecting a recording medium 407 recording (storing) a designated program to the computer 400. The recording medium 407 here includes a medium that records information optically, electrically, or magnetically, such as a CD-ROM, a floppy disk, a magneto-optical disk, etc. In addition, the recording medium 407 may include a semiconductor memory that records information electrically, such as a ROM, a flash memory, etc. The bus 408 is commonly connected to the above-mentioned various structural elements and transmits, for example, address signals, data signals, and various control signals.

[0069] <Processing Flow>

[0070] Next, a plurality of embodiments will be described regarding the processing flow of the prediction method according to the present embodiment.

[0071] [First embodiment]

[0072] Figure 5is a sequence diagram showing an example of processing of the prediction system according to the first embodiment. Figure 1 An example of prediction processing repeatedly performed by the prediction system 1 described in .

[0073] In step S501, the air conditioner 10 collects measurement data including the rotation speed of the drain pump 11 or the current value of the drain pump provided in the air conditioner 10. For example, the air conditioner 10 measures the rotation speed of the drain pump 11 regularly (for example, every 1 minute) and stores it in a storage unit provided in the air conditioner 10.

[0074] In step S502, the local controller 20 requests data acquisition from the air conditioner 10. In response, in step S503, the air conditioner 10 sends the collected measurement data (first measurement data) to the local controller 20. For example, the local controller 20 sends a data acquisition request to the air conditioner 10 at a specified time interval (for example, every 1 hour). In addition, when the air conditioner 10 receives the data acquisition request, it sends the measurement data of the most recent specified time period (for example, 1 hour) as the first measurement data to the local controller. Alternatively, when the air conditioner 10 receives the data acquisition request, it calculates the average value of the measurement data of the most recent specified time period, and sends the calculated average value as the first measurement data to the local controller 20.

[0075] In step S504, the local controller 20 sends the average value of the measured data of the most recent designated time period as the second measured data to the prediction server 100 based on the data (first measured data) received from the air conditioner 10. For example, when the first measured data received from the air conditioner 10 is the measured data of the most recent designated time period, the local controller 20 calculates the average value of the first measured data and sends the calculated average value as the second measured data to the prediction server 100. In addition, when the first measured data received from the air conditioner 10 is the average value of the measured data of the most recent designated time period, the local controller 20 sends the first measured data to the prediction server 100 as the second measured data.

[0076] In step S505 , the control unit 101 of the prediction server 100 stores (accumulates) the data (second measurement data) received from the local controller 20 or data processed from the received data in a storage unit such as the memory device 402 .

[0077] For example, when the drain pump 11 is not running (stopped), the measured data 201 of the rotation number and the current value of the drain pump 11 is 0. Therefore, when the cumulative average 202 or the three-day moving average is calculated using the data of the time period when the drain pump 11 is not running, a correct prediction result may not be obtained. Therefore, as an example, the control unit 101 can replace the data of the time period when the drain pump 11 is not running with the data before the time period when the drain pump 11 is not running, and store it in the storage unit.

[0078] For example, when the operation of the drainage pump 11 stops for one day due to vacation or going out, the control unit 101 can replace the data of the one-day stop of the drainage pump 11 with the above data and store them in the storage unit. In addition, when the drainage pump 11 stops for one hour, the control unit 101 can replace the data of the one-hour stop of the drainage pump 11 with the data of the previous hour and store them in the storage unit.

[0079] This is an appropriate example, and the control unit 101 may directly store the data received from the local controller 20 in the storage unit. In this case, the control unit 101 may calculate the cumulative average 202 or the three-day moving average 203 by excluding the data of the time period when the drainage pump 11 is stopped.

[0080] In step S506 , the control unit 101 calculates the cumulative average 202 and the 3-day moving average 203 based on the rotation speed (or current value) of the drain pump 11 included in the data stored in the storage unit, and calculates the difference 301 between the cumulative average 202 and the 3-day moving average 203 .

[0081] Furthermore, the control unit 101 executes the process of step S507 when the calculated difference 301 exceeds the preset threshold value 302. In step S507, the control unit 101 outputs the prediction result of predicting the abnormality of the drainage pump 11.

[0082] pass Figure 5 The processing of the prediction system 1 is, for example, Figure 3 As shown, before the abnormality of the drain pump 11 occurs, the prediction result of the abnormality of the drain pump 11 can be output with higher accuracy.

[0083] [Second embodiment]

[0084] The second embodiment describes a processing example in which the control unit 101 predicts abnormality of the drain pump 11 based on environmental data including temperature data or humidity data.

[0085] Figure 6 is a sequence diagram showing an example of processing of the prediction system according to the first embodiment. Figure 1Another example of the prediction process repeatedly executed by the prediction system 1 described in . In addition, since the basic process flow is the same as that of the first embodiment, the detailed description of the same process content as that of the first embodiment is omitted here.

[0086] In step S601, the air conditioner 10 collects measurement data including the number of rotations of the drain pump 11 or the current value of the drain pump, environmental data, and operating data of the air conditioner 10. Here, the environmental data includes, for example, temperature data indicating the room temperature and humidity data indicating the relative humidity. In addition, when the air conditioner 10 cannot obtain the humidity data, the environmental data may be only the temperature data. In addition, the operating data includes, for example, data indicating the operating mode (cold room, dehumidification, heating room, etc.) of the air conditioner 10, and data such as the evaporation temperature measured by the temperature sensor of the heat exchanger of the air conditioner 10. In addition, the data of the evaporation temperature may be included in the environmental data.

[0087] In step S602, the local controller 20 requests data acquisition from the air conditioner 10. In response, in step S603, the air conditioner 10 transmits the collected data including the measurement data (first measurement data), environmental data, and operation data to the local controller 20.

[0088] In step S604 , the local controller 20 transmits data including the second measurement data, which is the average value of the first measurement data received from the air conditioner 10 , the environmental data, the operating data, and the like to the prediction server 100 .

[0089] In step S605 , the control unit 101 of the prediction server 100 stores (accumulates) the data received from the local controller 20 or data processed from the received data in a storage unit such as the memory device 402 .

[0090] In step S606, the control unit 101 determines the time period during which the water is not drained based on the data stored in the storage unit. For example, the control unit 101 performs the following steps: Figure 7 The drainage judgment process shown.

[0091] Figure 7 : is a flowchart showing an example of the drainage determination process according to the second embodiment. Figure 6 In step S606, the control unit 101 performs an example of processing. The control unit 101 performs, for example, data stored in the storage unit, data in which the air conditioner 10 is in cooling operation. Figure 7 processing.

[0092] In step S701, the control unit 101 calculates the dew point temperature (temperature at which condensation occurs) using the temperature data and the humidity data, or the temperature data. For example, the control unit 101 obtains the water vapor pressure from the temperature data indicating the room temperature and the humidity data indicating the relative humidity, and obtains the temperature at which the water vapor pressure becomes the saturated water vapor pressure. In addition, when there is no humidity data, the control unit 101 calculates the dew point temperature by assuming the value of the relative humidity.

[0093] In step S702 , the control unit 102 acquires the evaporation temperature from the operation data. Here, the evaporation temperature is the temperature of the heat exchanger measured by the sensor of the heat exchanger of the air conditioner 10 .

[0094] In step S703, the control unit 101 determines whether the evaporation temperature is below the calculated dew point temperature. When the evaporation temperature is below the dew point temperature, condensation occurs, and in step S704, the control unit 101 determines that drainage is performed. On the other hand, when the evaporation temperature is not below the dew point temperature, the control unit 101 determines that drainage is not performed in step S705.

[0095] Here, return Figure 6 , the description of the sequence diagram continues. In step S607, the control unit 101 removes the data of the time period when the water is not drained from the measurement data stored in the storage unit (for example, the number of revolutions of the drainage pump 11), and calculates the difference 301 between the cumulative average 202 and the 3-day moving average 203. This is because when the water is not drained, the drainage pump 11 has no load, and the number of revolutions of the drainage pump 11 increases. Therefore, by removing the data of the time period when the water is not drained, the prediction accuracy of the abnormality of the drainage pump 11 can be further improved.

[0096] Next, the control unit 101 executes the process of step S608 when the calculated difference 301 exceeds the preset threshold value 302. In step S608, the control unit 101 outputs the prediction result of predicting the abnormality of the drainage pump 11.

[0097] pass Figure 6 By processing, the prediction system 1 can further improve the accuracy of the prediction result of the abnormality of the drainage pump 11.

[0098] [Third embodiment]

[0099] In the first and second embodiments, the control unit 101 outputs a prediction result predicting abnormality of the drain pump 11 when the deviation between the cumulative average 202 and the three-day moving average 203 of the measurement data of the rotation speed (or current value) of the drain pump 11 exceeds a threshold value.

[0100] The third embodiment describes an example in which the control unit 101 uses the following model to predict when the drain pump 11 is abnormal: a prediction model that has been learned by machine learning using data of a specified time period when the drain pump 11 is normal and data of a specified time period when the drain pump 11 is abnormal as training data.

[0101] Figure 8 FIG. 2 is a diagram showing an example of a system configuration of a prediction system according to a third embodiment. Figure 8 As shown, the prediction system 1 involved in the third embodiment is Figure 1 The prediction system 1 described above includes a prediction model 801 in addition to the system configuration.

[0102] The prediction model 801 is a prediction model that has been learned by machine learning in a manner that predicts the abnormality of the drainage pump 11 by using the data of the specified time period when the drainage pump 11 is normal and the data of the specified time period when the drainage pump 11 is abnormal as training data. The control unit 101 inputs the measurement data of the specified time period stored in the storage unit (for example, the cumulative average 202 and the 3-day moving average 203) into the prediction model 801, thereby obtaining a prediction result indicating whether the drainage pump 11 has an abnormality. In addition, the measurement data of the specified time period may be only the 3-day moving average 203.

[0103] In addition, the prediction model 801 may be learned by the prediction server 100 using the data stored in the storage unit by the control unit 101, and the prediction model 801 learned by other information processing devices may be set in the prediction server 100.

[0104] <Processing Flow>

[0105] Fig. 9 is a sequence diagram showing an example of processing of the prediction system according to the third embodiment. The processing is given by Figure 8 An example of a prediction process repeatedly performed by the prediction system 1 described in Figure 5 The processing of the prediction system according to the first embodiment described in is the same as that of the first embodiment, and thus detailed description of the processing contents that are the same as those of the first embodiment will be omitted here.

[0106] In step S901 , the air conditioner 10 collects measurement data including the rotation speed of the drain pump 11 included in the air conditioner 10 or the current value of the drain pump.

[0107] In step S902, the local controller 20 requests the air conditioner 10 to obtain data. In response, the air conditioner 10 transmits the collected measurement data (first measurement data) to the local controller 20 in step S903.

[0108] In step S904 , the local controller 20 transmits the average value of the measurement data in the latest designated time period as the second measurement data to the prediction server 100 based on the data (first measurement data) received from the air conditioner 10 .

[0109] In step S905 , the control unit 101 of the prediction server 100 stores (accumulates) the data (second measurement data) received from the local controller 20 or data processed from the received data in a storage unit such as the memory device 402 .

[0110] In step S906, the control unit 101 inputs the measurement data of the designated time period stored in the storage unit into the learned prediction model 801. For example, the control unit 101 calculates the cumulative average 202 and the three-day moving average 203 of the measurement data of the number of revolutions (or current value) of the drain pump 11 stored in the storage unit, and inputs the calculated data into the learned prediction model 801. Thus, the learned prediction model 801 outputs a prediction result indicating whether an abnormality has occurred in the drain pump 11.

[0111] Next, the control unit 101 executes the process of step S907 when predicting that the drainage pump 11 will have an abnormality. In step S907, the control unit 101 outputs the prediction result of predicting the abnormality of the drainage pump 11.

[0112] In this way, the prediction system 1 can predict the abnormality of the drain pump 11 using the following model, using the data when the drain pump 11 is normal and the data when the drain pump 11 is abnormal as training data, and performing machine learning to obtain the learned prediction model 801.

[0113] also, Fig. 9 The processing described in is an example. For example, the processing of the prediction system involved in the third embodiment can be as follows Figure 6 In the second embodiment described in FIG. 1 , the abnormality of the drainage pump 11 is predicted by excluding data of a time period in which drainage is not performed from the data stored in the storage unit.

[0114] [Fourth embodiment]

[0115] In the third embodiment, the prediction model 801 is subjected to machine learning using data when the drain pump 11 is normal and data when the drain pump 11 is abnormal as training data.

[0116] In the fourth embodiment, the prediction model 801 is a prediction model that has been learned by machine learning using images representing data when the drain pump 11 is normal and images representing data when the drain pump 11 is abnormal as training data.

[0117] Here, the image representing the data when the drain pump 11 is normal and the image representing the data when the drain pump 11 is abnormal can be, for example, Figure 2 As shown, an image of a graph 200 is shown in which data of a cumulative average 202 and a 3-day moving average 203 are plotted.

[0118] <Processing Flow>

[0119] Fig.10 is a sequence diagram showing an example of processing of the prediction system according to the fourth embodiment. The processing is given by Figure 8 Another example of a prediction process repeatedly performed by the prediction system 1 described in Fig.10 In the process shown, the processes of steps S901 to S905 and S907 are the same as Fig. 9 The processing of the prediction system involved in the third embodiment described in is the same as that of the third embodiment, so the description is omitted here.

[0120] In step S1001, the control unit 101 generates a graph of the change trend of the cumulative average 202 and the three-day moving average 203 of the measurement data (for example, the number of rotations of the drainage pump 11) stored in the storage unit. Figure 2 An image of chart 200 is shown.

[0121] In step S1002, the control unit 101 inputs the created image into the learned prediction model 801. Thus, the learned prediction model 801 outputs a prediction result indicating whether or not an abnormality has occurred in the drainage pump 11.

[0122] Next, when predicting that the drain pump 11 has an abnormality, the control unit 101 executes the process of step S907 .

[0123] In this way, the prediction system 1 can predict the abnormality of the drain pump 11 using the following model, using the image of the data when the drain pump 11 is normal and the image of the data when the drain pump 11 is abnormal as training data, and performing machine learning on the learned prediction model 801.

[0124] In addition, the processing of the prediction system involved in the fourth embodiment is also as follows Figure 6 According to the second embodiment described in FIG. 1 , the abnormality of the drainage pump 11 can be predicted by excluding the data of the time period in which drainage is not performed from the data stored in the storage unit.

[0125] As described above, according to the embodiments of the present invention, abnormality of the drain pump 11 included in the air conditioner 10 can be predicted with higher accuracy.

[0126] It should be understood that the above describes the embodiments, but various changes in form and details are possible without departing from the spirit and scope of the patent claims.

[0127] For example, in each of the above-mentioned embodiments, an average value such as the cumulative average 202 or the three-day moving average is used to predict abnormality of the drainage pump 11. However, the present invention is not limited thereto, and the average value may be another representative value such as the median or mode.

[0128] Furthermore, in each of the above-mentioned embodiments, the prediction server 100 includes the control unit 101, but the control unit 101 may be included in the local controller 20. Furthermore, the control unit 101 may be realized by, for example, a virtual computer on a cloud.

[0129] This application claims the priority of basic application No. 2022-136193 filed in the Japan Patent Office on August 29, 2022, the entire contents of which are incorporated herein by reference.

[0130] [Reference Signs]

[0131] 1 Prediction system

[0132] 2 Communication Network

[0133] 10 Air conditioner

[0134] 11. Drain pump

[0135] 20 Local Controller

[0136] 100 Prediction Server

[0137] 101 Control Department

[0138] 200 Charts

[0139] 202 Cumulative Average

[0140] 203 3-day moving average

[0141] 302 Threshold

[0142] 400 Computer

[0143] 407 Recording Media

[0144] 801 Prediction Model.

Claims

1. A prediction system comprising an air conditioner having a drain pump and a control unit, The control unit, acquiring data of the number of rotations of the drain pump or the current value of the drain pump, Based on the change of the data in a designated period of time, a prediction result predicting abnormality of the drainage pump is output. 2 . The prediction system according to claim 1 , wherein the control unit outputs the prediction result based on a representative value of the data in the designated time period. 3 . The prediction system according to claim 1 , wherein the control unit predicts abnormality of the drainage pump based on an average value of the data in a first designated time period and the data in a second designated time period shorter than the first designated time period.

4. The prediction system according to claim 3, wherein the control unit outputs the prediction result predicting abnormality of the drainage pump when a deviation between an average value of the data in the first designated time period and an average value of the data in the second designated time period exceeds a threshold value. 5 . The prediction system according to claim 1 , wherein the control unit predicts abnormality of the drain pump further based on environmental data including temperature data or humidity data.

6. According to the prediction system according to any one of claims 1 to 5, the control unit does not use the data of the time period when the air conditioner or the drain pump is not operating, but uses the data before the time period to predict the abnormality of the drain pump. 7 . The prediction system according to claim 6 , wherein the control part determines a time period during which the drain pump is not operated based on environmental data including temperature data or humidity data.

8. According to the prediction system according to any one of claims 1 to 5, the control unit determines whether drainage occurs based on environmental data including temperature data or humidity data, removes the data of the number of rotations of the drainage pump or the current value of the drainage pump during a time period when drainage does not occur, and predicts abnormality of the drainage pump. 9 . The prediction system according to claim 8 , wherein the control unit further determines whether the drainage occurs by using information indicating whether the air conditioner is operated in a designated mode.

10. The prediction system according to any one of claims 1 to 9, comprising an edge device for collecting the data from the air conditioner, The control unit acquires the data averaged in the air conditioner or the edge device.

11. The prediction system according to claim 1, wherein the control unit predicts the abnormality of the drain pump using a prediction model that has been learned by machine learning using the data when the drain pump is normal and the data when the drain pump is abnormal as training data.

12. According to the prediction system of claim 1, the control unit predicts the abnormality of the drainage pump using the following model: an image representing the data when the drainage pump is normal and an image representing the data when the drainage pump is abnormal are used as training data to perform machine learning on a learned prediction model.

13. A prediction method, in a prediction system having an air conditioner with a drain pump and a control unit, The control unit, acquiring data of the number of rotations of the drain pump or the current value of the drain pump, Based on the change of the data in a designated period of time, a prediction result predicting abnormality of the drainage pump is output.

14. A program, in a prediction system having an air conditioner with a drain pump and a control unit, Causes the computer to execute: a process of acquiring data of the number of rotations of the drain pump or the current value of the drain pump, and A process of outputting a prediction result of predicting abnormality of the drainage pump based on a change in the data during a designated period of time.

Images