JP2003172567A - Fault diagnosis method, fault diagnosis device, in-shop equipment management system, and recording medium - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily and adequately set the threshold for discriminating a failure for each appliance and each installation place. <P>SOLUTION: The data of the discharge pressure and the pull-down time is corrected to the data with each influential factor eliminated therefrom based on the data which is actually collected (Step S2) after installing the appliance (Step S3). Next, the threshold is determined using a normal distribution curve (or t-distribution curve) for each of the data on in-chamber temperature, the data on the discharge pressure after correction, and the data on the pull-down time (Step S4). <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for managing / controlling various devices such as a showcase, an air conditioner, and lighting arranged in a store from inside or outside the store, and more particularly to a failure diagnosis device and a failure diagnosis method. .

[0002]

2. Description of the Related Art Conventionally, there has been known a technique for diagnosing a failure of a device such as a showcase installed in a store such as a supermarket or a convenience store. For example, in Japanese Unexamined Patent Application Publication No. 10-238920, data regarding the operating state of a device is classified for each operating condition of the device for which the data is obtained, a database regarding the past operating state is constructed, and the current operating state of the device is set. Disclosed is a device operational status management device configured to compare current status data to corresponding historical data in a database to assess the current operational status of the device.

Further, in Japanese Unexamined Patent Publication No. 10-267509, data relating to the operating state of a device is classified and stored for each operating condition of the device from which the data was obtained, and a database relating to the past operating state is constructed. A device operating state management device configured to predict when a device will fail based on the current operating state data and the corresponding past data in the database.

As described above, the above-mentioned two prior arts construct a database in which data on equipment are accumulated, and use the data registered in this database to empirically evaluate the operating state of equipment and predict failure. Met.

[0005]

However, in the case of the above-mentioned prior art, unless careful data about the equipment is acquired, the influence of the fluctuation of the ambient temperature, the case where a plurality of failures occur at the same time, etc. Since it is not possible to accurately evaluate the operating state of equipment and predict failure, it requires a great deal of labor to build a database, and an inference mechanism and the like must also be installed as necessary, which is a human burden /
It is not efficient because of the large cost burden.

Especially when not only showcases but also various devices such as air conditioners and lights installed in a store are to be collectively controlled, enormous data must be stored for each type of device. Instead, the database becomes huge.

An object of the present invention is to provide a failure diagnosis device and a failure diagnosis method capable of efficiently diagnosing or predicting a failure of equipment such as a showcase arranged in a store.

[0008]

A failure diagnosis method according to the present invention collects state data of each freezing / refrigerating showcase including a showcase main body and a refrigerator in a predetermined period within a predetermined learning period, The normal distribution or t distribution is obtained based on the collected state data, each threshold value necessary for determining various failures is determined based on the normal distribution or t distribution, and after the learning period, the threshold value is used. , Diagnose failure occurrence.

As described above, by actually collecting data after installation of a device such as a showcase and further determining a threshold value by a statistical method, a more appropriate threshold value can be set, and a false alarm or vice versa is detected. It is possible to significantly reduce the possibility of occurrence of failure to detect the occurrence.

Further, for example, the discharge side pressure of the refrigerator and the pull-down time of the temperature control temperature of the showcase as the state data are corrected after removing the influence of the ambient environment at the time of data collection. The normal distribution or t distribution is obtained based on the corrected data.

The pull-down time between the discharge side pressure of the refrigerator and the temperature control temperature of the showcase depends on the ambient environment (mainly the ambient temperature).
Fluctuates under the influence of. Therefore, for example, the ambient temperature at the time of collecting each data is stored in association with each other, the relation between the ambient temperature and the data is obtained, and the data is corrected using this, and an appropriate normal distribution or t distribution is obtained. Therefore, an appropriate threshold can be set.

The failure diagnosing device according to the present invention has a normal distribution based on various time-series state data of each freezing / refrigerating showcase including a showcase main body and a refrigerator, which are collected within a predetermined learning period. Alternatively, using a threshold value determining unit that determines a t distribution and determines a threshold value necessary for determining various failures based on the normal distribution or the t distribution, and a threshold value determined by the threshold value determining unit after the learning period, And a failure determination means for determining the presence or absence of a failure.

Further, for example, the failure determination means compares various state data measured at any time with a corresponding threshold value, and when the threshold value is exceeded, the number of times the threshold is exceeded or the number of times the threshold is exceeded is a preset number of times or more. In this case, it is determined that there is a failure.

In the failure diagnosis apparatus of the present invention, since an appropriate threshold value can be set according to the actual installation environment of the equipment, it is possible to immediately determine that there is a failure when the threshold value is exceeded. There is a possibility that the threshold value may be temporarily exceeded even if the failure is not caused by the cause. Therefore, if the threshold value is exceeded a predetermined number of times or more, it may be determined that there is a failure.

The failure diagnosis apparatus, for example, when the failure judgment unit judges that there is a failure, sends an e-mail notifying the occurrence of the failure to an external center apparatus or a portable terminal via a network. It may be configured to further include an occurrence notification unit.

Further, for example, in the failure determination means, when the temperature data in the showcase cabinet, which is one of the various status data, exceeds a threshold value determined in correspondence with the temperature data in the cabinet, an ice bank. When it is diagnosed that the ice bank is dealt with, it may be configured to further include an ice bank coping means for coping with the ice bank by changing the defrosting condition.

Since an ice bank abnormality can usually be dealt with by removing the adhered frost, conventionally, maintenance personnel and the like manually removed it, but on the other hand, the showcase periodically and automatically. There is a defrosting function for defrosting, and if the frost cannot be completely removed by this defrosting function for some reason, an ice bank error occurs. Therefore, by changing the defrosting condition of this defrosting function (shortening the time interval between defrosting operations, lengthening the time taken for one defrosting operation, etc.) Can respond,
The work load on maintenance personnel can be reduced.

Further, for example, the threshold value determining means and the failure determining means may determine the threshold value / determine the failure based on the data under a specific condition among the collected various state data. .

As described above, by setting restrictions on the time-series data used for threshold value determination / fault determination, it is possible to identify disturbance factors and improve the accuracy of threshold setting and fault determination.

Further, for example, the failure determination means may predict the failure occurrence time based on a time series transition of various state data measured at any time and a corresponding threshold value. By predicting the failure time based on the transition speed of the time-series data measured after the threshold is set and the distance to the threshold, the maintenance plan can be accurately constructed.

Further, for example, the failure diagnosis device is an in-store information processing device or a network adapter connected to at least the controller of the freezing / refrigerating showcase via the in-store network, or any network outside the store. It is realized in any information processing device outside the store connected via the.

When implemented in the in-store information processing device, the version upgrade of the function becomes easy. When implemented in a network adapter or any information processing device outside the store, there is no need to provide an in-store information processing device.

Further, for example, when the occurrence of the failure is predicted, the failure prediction notifying means for transmitting the failure prediction information to an external server or a mobile terminal via an electronic mail via a network is further provided. It may be configured as follows.

Further, the present invention is configured as an in-store device management system capable of improving the efficiency of maintenance work based on the failure prediction information notified by e-mail from the failure diagnosis apparatus according to claim 10. You can also

That is, for example, there is provided an in-store device management system in which the failure diagnosis apparatus according to claim 10 and an external server are connected to a network, wherein the external server receives the notification of the electronic mail. Then, based on the failure prediction information, the system configuration is such that a part required for repair is searched for in response to the predicted failure, and the part is ordered to the server of the parts manufacturer.

As described above, when a failure is predicted to occur, parts required for repair can be automatically ordered in advance, parts can be ordered smoothly, and efficient repair can be performed.

Further, for example, an in-store device management system in which the failure diagnosis device according to claim 10 and an external server are connected to a network, wherein a server of a maintenance company among the external servers is When receiving the notification of the regular inspection table storage means for storing the regular inspection table data and the e-mail, the failure prediction information is included in the priority inspection items of the regular inspection day of the store of the e-mail sender in the regular inspection table data. The system configuration has a periodic inspection table updating means for filling in.

By doing this, the day of the regular inspection day,
The maintenance staff will focus on the equipment / location indicated by the failure prediction information entered as the priority inspection item, so if there is any equipment / location where a failure is likely to occur in the near future, overlook it. Therefore, it is possible to reliably repair or remove the cause of failure such as frost / dust, and prevent the failure in advance.

Further, the present invention is not limited to the configuration as the above-mentioned device, but may be configured as a recording medium itself recording a program for realizing a failure diagnosis device or the program itself.

[0030]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram schematically showing the configuration of the entire in-store device management system according to this embodiment.

FIG. 2 is a block diagram showing the internal structure of the store 100A shown in FIG. Here, for convenience of explanation, only the store 100A is shown in detail, but the stores 100B and 100C have substantially the same configuration as the store 100A.

The in-store device management system shown in FIGS. 1 and 2 is provided separately for each of various devices such as a showcase 1 (freezing / refrigerating open showcase), an air conditioner 2 and other lights in each store 100A to 100C. Controller (showcase controller 9, air conditioner controller 1
1, a heat storage controller 12, a lighting controller 21) are provided, and an in-store PC 8 and an information processing device outside the store (store headquarters server 6, service) are provided via an NA (network adapter) 3, a router 4, and a network 5. It is configured to connect to the headquarters server 7 etc.). Then, in this in-store device management system, each controller collects data from the devices under its jurisdiction over a certain period of time, and transmits the collected data to an in-store PC (hereinafter referred to as in-store PC) 8. . As a specific example, the showcase controller 9 collects the data of the showcase 1 and the refrigerator 10 and transmits the data to the in-store PC 8. The air conditioner controller 11 collects the data of the air conditioner 2 and sends it to the in-store PC 8. The heat storage controller 12 collects the data of the heat storage tank 13 and stores the data in the in-store PC 8
Send to. Alternatively, the collected data may be transmitted to the information processing device outside the store via the network 5. It should be noted that these controllers are, of course, not dedicated to data collection.

Then, the in-store PC 8 or any information processing device outside the store which has received the data from these controllers performs the failure diagnosis / prediction of each device based on the received data. A specific failure diagnosis / prediction method will be described in detail later.

In FIG. 1, store 100A and store 100
B and the store 100C are stores in which the showcase 1 and the like are arranged. Further, the air conditioner 2, the vending machine 14, the security system 15 and the like may be installed in these stores. In addition, here, these stores 100
It is assumed that A to 100C are locally distributed stores such as supermarkets that are chain stores.

Further, the store headquarters server 6 is connected to each store 100.
It is a terminal device of the headquarters mechanism that supervises A to 100C, instructing each store the type and price of the product to be provided, and also grasps the sales situation of each store 100A to 100C. The service headquarters server 7 has each store 100A-100
C is a terminal device of the headquarters mechanism that centrally manages the content of services to be provided to customers. Security headquarters server 16
Is a terminal device of the headquarters mechanism that controls the security mechanism related to entry / exit management, fire, and crime prevention in each store 100A to 100C.

As described above, in the in-store device management system of this example, the store headquarters server 6, the service headquarters server 7, the security headquarters server are not simply managed for each store, but the products, services, security, etc. of the store. 1
6 has a common headquarters mechanism for each of the stores 100A to 100C, and centrally manages each of the stores 100A to 100C for each function.

The store owner server 17 is a terminal device owned by the owner of each store 100A to 100C. The device owner / server 18 operates from each store 100A to
When a device such as a showcase installed in 100C is a rental device rented from a manufacturer or the like, it is a terminal device owned by the owner of this rental device. The equipment maintenance company / server 19 is for each store 100A-10
It is a terminal device owned by a maintenance company that maintains and manages equipment such as the showcase 1 installed at 0C. still,
Here, for convenience of description, it is assumed that the store owners, device owners, and device maintenance companies of the stores 100A to 100C are the same person.

Further, each of the stores 100A to 100C, the store headquarters server 6, the service headquarters server 7, the security headquarters server 16, the store owner server 17, the equipment owner server 18, the equipment maintenance company server 19,
Each is connected to the network 5 via the router 4. Specifically, this network 5 is a public line network (telephone line, ISDN, etc.), the Internet, or an ATM.
Is a network, etc., and each terminal device is a TC
Communication is performed using the P / IP protocol. Therefore, it is also possible to perform file transfer (FTP) or electronic mail transfer on the Internet between each store 100A to 100C and each terminal device.

Therefore, the failure diagnosis and prediction of each device in each store 100A-100C is performed by each store 100A-100C.
Not only the in-store PC 8 in the store, but also each terminal device such as the store headquarters server 6, the service headquarters server 7, the security headquarters server 16, the store owner server 17, the equipment owner server 18, the equipment maintenance company server 19 and the like. You can also Alternatively, it may be performed by NA3 described later.

The system in each store 100A to 100C shown in FIG. 2 includes a store controller 20, a showcase controller 9, an air conditioner controller 11, a heat storage controller 12, a lighting controller 21, a vending machine 14, a security system 15 and the like. Is connected to the NA3 by a multi-drop method (a method of connecting via various controllers, vending machines, etc. as shown in the figure), and provides notifications and instructions of information related to the status of each device. You can do it.

Specifically, for example, the heat storage controller 1
2 receives data indicating load conditions of the refrigerator 22 and the refrigerator 23 from the showcase controller 9 and the air conditioner controller 11, and controls the heat storage tank 13 based on the received data (load condition of the refrigerator). To do.

The NA3 provides a homepage such as "display of device status / input of device setting", for example, on the "browser" screen of the terminal device to which access is permitted,
As described above, a function that enables you to display the status and settings of the connected device (hereinafter referred to as the homepage function), and when a problem occurs with the connected device, the administrator or in advance automatically A network adapter having a function (hereinafter, referred to as an automatic mail sending function) of notifying a designated / registered person of a trouble occurrence by e-mail or the like. Applicant has filed).

By using this NA3, each store 1
Monitoring and management of various devices in 00A to 100C can be centrally performed externally (for example, the store headquarters server 6 etc.), or a failure occurrence etc. is promptly notified to the device maintenance company / server 19 etc. by e-mail. can do. The NA3 can also be formed as an option board having an interface for accommodating a serial communication line and mounted on the PC 8 or the like in the store. Also, N
Instead of A3, a personal computer or the like having a home page function and a mail sending function may be provided, or the NA3 may be directly attached to various devices in the store.

The store controller 20 is a controller for performing energy saving control of the entire store, and is specifically constituted by a personal computer or the like. Showcase controller 9
Is a controller that controls the refrigerator 22 in order to adjust the temperature of the commodities in the showcase 1 and the refrigerator 10. The air conditioner controller 11 is a controller that controls the air conditioner 2 for adjusting the inside of the store to a predetermined temperature by using the refrigerator 23. The heat storage controller 12 is
It is a controller that controls the heat storage tank 13 that stores ice based on the load states of the refrigerator 22 and the refrigerator 23.

The lighting controller 21 is a controller for controlling various lighting devices arranged on the ceiling or the like in the store, and also controls the fluorescent lamp of the showcase 1.
Specifically, for example, when performing the dimming of the fluorescent lamp of the showcase 1, dimming control corresponding to the illumination control system disclosed in JP-A-6-325874 is performed.

The vending machine 14 is an owner-managed vending machine which is installed in a store and sells beverages and foods. The vending machine 14 is not provided with a controller for the vending machine 14 because it has little relation to a showcase or the like.

The security system 15 is a system for security management of the entire store based on data from a sensor provided on the door 24, a fire alarm 25, a crime prevention device 26 and the like. In addition, this security system 15
Does not have a separate controller because it originally has the function of a controller.

The in-store PC 8 is a personal computer installed in the store, and manages the types and prices of products sold in the store, counts sales, and based on time series data received from each controller. Failure diagnosis of each device.
The PHS adapter 27 attached to the in-store PC 8
Via the store owner server 17 or equipment owner
It is also possible to contact the server 18 or the like by using a voice message. Specifically, for example, failure prediction information described later is transmitted.

In the above description of the embodiment, various devices such as the showcase controller 9 and the air conditioner controller 11 are connected by the multi-drop method, but the present invention is not limited to this. It is also possible to use various network topologies such as ring type and ring type. In addition, the stores 100A to 1 through the network 5
00C and the store headquarters server 6 and the like can communicate with each other, but the present invention is not limited to this. For example, the communication is possible via a transmission line such as a wireless communication system or satellite communication. Good.

Next, the structure of the showcase 1 will be described. Here, an open showcase will be described as an example with reference to FIG. FIG. 3 shows a side sectional view of an example of the open showcase.

The open showcase 32 shown in FIG.
The main body 27, the display shelf 28, the opening 29, the refrigerator 30, the refrigerant pipe 31, the evaporator 33, the blower 34, the duct 35, the outlet 36, and the like, the thermistor 45 for the outside air temperature, the thermistor 46 for the inside temperature, and the temperature Temperature control thermistor 47,
The defrosting temperature thermistor 48 and the like are installed. Note that these are basically installed indoors (in a store), and in order to distinguish them from the refrigerator 30, which is an outdoor unit, they will be hereinafter referred to as the main body side of the open showcase 32.

The main body 27 is the open showcase 3 concerned.
It is a basic housing forming 2 and has an opening (opening 29) in one direction so that goods can be put in and taken out from here. The display shelf 28 is a shelf for placing products.

The refrigerator 30 generates a high temperature and high pressure liquid refrigerant and supplies it to the main body side of the open showcase 32.
That is, the high-temperature, high-pressure liquid refrigerant generated by the refrigerator 30 passes through the refrigerant pipe 31 and passes through the open showcase 3
2 It is supplied to the main body side. This becomes a low-temperature low-pressure liquid refrigerant by an expansion valve (not shown) and is sent to the evaporator 33. This cools the air around the evaporator 33. The cooled air is blown into the duct 35 by the blower 34, and is further sent from the outlet 36 into the inside 37 of the open showcase 32 to cool the whole inside 37. Outlet 3
A part of the cool air sent from the inside 6 to the inside 37 is sucked into the suction port 38 while forming an air curtain as shown by an arrow in the figure. This is again sent by the blower 34, creating a circulation of cold air.

The outside air temperature thermistor 45 is
This is a temperature sensor for detecting the temperature around the open showcase 32. The temperature inside the thermistor 46 is a temperature sensor for detecting the temperature of the inside 37 of the open showcase 32. The controlled temperature thermistor 47 is a temperature sensor for detecting the temperature of the air sent to the duct 35 by the blower 34. Thermistor 4 for defrosting temperature
Reference numeral 8 is a temperature sensor for detecting the temperature near the evaporator 33.

The refrigerator 30 will be described in detail.
The refrigerator 30 has a compressor 39, a condenser 40, and the like, as shown in FIG.

The low-temperature low-pressure liquid refrigerant sent to the evaporator 33 becomes a low-temperature low-pressure gas refrigerant, and becomes the refrigerant pipe 31.
And is returned to the refrigerator 30 side. First, this is sent to the compressor 39 and compressed into a high-temperature, high-pressure gas refrigerant. This gas refrigerant is sent to the condenser 40, and the condenser 40
By radiating heat, the liquid refrigerant becomes a high-temperature, high-pressure liquid refrigerant, which again passes through the refrigerant pipe 31 and is sent to the open showcase 32 side. The refrigerant pipe 31 is arranged in a loop to circulate the refrigerant.

Further, the refrigerator 30 is provided with a temperature sensor and a pressure sensor. That is, as shown in FIG.
A suction (low pressure) side temperature sensor 41, a suction (low pressure) side pressure sensor 42, a discharge (high pressure) side temperature sensor 43, and a discharge (high pressure) side pressure sensor 44 are provided. Suction (low pressure)
The side temperature sensor 41 and the suction (low pressure) side pressure sensor 42 are
The sensor is provided to measure the temperature and pressure of the low-temperature low-pressure gas refrigerant before entering the compressor 39. The discharge (high pressure) side temperature sensor 43 and the discharge (high pressure) side pressure sensor 44 are sensors provided to measure the temperature and pressure of the high temperature and high pressure gas refrigerant compressed by the compressor 39.

In the case of an inverter type refrigerator,
If it is expected that the allowable range of the refrigeration cycle equipment and electric / electronic parts will be exceeded due to an increase in load or transient disturbance, a protection control operation different from the normal operation control is started. For example, protection control such as reducing the operating frequency when the discharge (high pressure) side pressure is too high is performed. By performing such protection control, it is possible to prevent the failure of the refrigerator 30 itself, but there may be a problem that the temperature inside the showcase is increased.

Next, in the following, when various failures occur,
The signal change characteristics of the various sensors installed in the open showcase 32 and the refrigerator 30 will be described. First, a case where a frosting failure (ice bank) occurs in the evaporator 33 will be described.

FIG. 4 (a) is an external view of the evaporator 33 with frost completely removed, and FIG. 4 (b) shows the amount of frost 49 adhered to the limit amount where frost can be completely removed by the defrosting operation. It is an external view of the evaporator 33 in a state.

FIG. 5 mainly shows the state of FIG.
The graph shows the changes over time in the temperatures measured by the thermistor 45 for the outside air temperature, the thermistor 46 for the inside temperature, and the thermistor 47 for the temperature control until the state of (b) is reached (and thereafter). The vertical axis of the graph shown in the figure is the measured temperature, and the horizontal axis is the time.

In FIG. 5, time t1 is in the state of FIG. 4 (a), and time t2 is in the state of FIG. 4 (b).
The outdoor temperature thermistor 45 is an open showcase 3
Since the temperature around 2 is measured, it is not affected by the state of frost adhering to the evaporator 33. In the example shown in the figure, it is assumed that the ambient temperature is almost constant. On the other hand, as the amount of frost attached to the evaporator 33 increases, it becomes more difficult for the cool air to be blown,
Internal temperature thermistor 46 and controlled temperature thermistor 4
Temperature measured at 7 (temperature in the chamber 37, duct 35)
Gradually increases as shown in the figure.

Next, FIGS. 6 (a), 6 (b), 7 (a),
A case where the condenser 40 of the refrigerator 30 is clogged and fails will be described with reference to FIG. FIG. 6A is an external view of the condenser 40 with the dust 50 completely removed, and FIG. 6B is an external view of the condenser 40 with the dust 50 clogged.

7 (a) and 7 (b) are mainly shown in FIG. 6 (a).
Change from the state of FIG. 6 to the state of FIG. 6B (and thereafter) with time of the temperature / pressure of the refrigerant measured by the discharge (high pressure) side temperature sensor 43 and the discharge (high pressure) side pressure sensor 44, respectively. Is shown in the graph. In the graphs shown in (a) and (b), the horizontal axis represents time, the vertical axis (a) represents temperature, and (b) represents pressure.

In FIGS. 7 (a) and 7 (b), time t3 is in the state of FIG. 6 (a), and time t4 is in the state of FIG. 6 (b). When the condenser 40 is clogged and the heat transfer area is reduced, heat radiation from the gas refrigerant is not normally performed, and therefore the temperature and pressure of the gas refrigerant on the discharge (high pressure) side rise according to the degree of clogging. . As the temperature and pressure of the gas refrigerant on the discharge (high pressure) side rise, the temperature and pressure of the refrigerant guided to the evaporator 33 also rise, so the temperature of the inside 37 of the showcase rises.

Furthermore, the temperature and pressure of the refrigerant returning from the evaporator 37 to the refrigerator 30 side, that is, the gas refrigerant on the suction (low pressure) side also rises. When the pressure of this gas refrigerant rises, in the case of the above-mentioned inverter type refrigerator, the protection control operation for lowering the operating frequency is started as described above. By entering the protection control operation, the refrigerator 30 may escape the failure,
If the operating frequency is lowered, the temperature and pressure of the refrigerant introduced to the evaporator 33 will further rise, and the inside 37 of the showcase will not be cooled at all.

Furthermore, if the clogging of the condenser 40 is further aggravated by continuing the operation as it is, finally the refrigerator 3
0 is abnormal due to high pressure and stops. Next, the phenomenon when the gas refrigerant leaks or is insufficient will be described below.

FIG. 8 shows the pull-down time of the inside temperature 51 (measured by the inside temperature thermistor 46) and the temperature control temperature 52 (measured by the temperature control thermistor 47) after the defrosting operation. The graph shown is shown.

The left side of the figure shows changes over time in the internal temperature 51 and the controlled temperature 52 when there is no leakage / insufficiency of the gas refrigerant and on the right side there is leakage / insufficiency of the gas refrigerant. First, as shown on the left side of the figure, in the state where there is no leakage / insufficiency of the gas refrigerant, the defrosting operation causes the internal temperature 51 and the controlled temperature 52 to rise rapidly, but after the defrosting operation ends. Drops sharply. On the other hand, when there is a leak / insufficiency of the gas refrigerant, the cooling capacity of the evaporator 33 decreases, so that the inside temperature 51 and the controlled temperature 52 decrease after the defrosting operation ends, as shown on the right side of the figure. It will take more time to pull down.

The signal change characteristics of the various sensors for detecting the state data of the open showcase 32 and the refrigerator 30 have been described above in the case of various failures. Based on the signal (measurement data) of, the threshold setting and the subsequent failure diagnosis / prediction processing are performed.

The failure diagnosis / prediction method according to this embodiment will be described in detail below with reference to FIG. 9 and the like. The prediction is to judge that a failure is likely to occur in the near future.

FIG. 9 shows threshold value determination / fault diagnosis /
It is a flowchart figure for demonstrating a prediction process. As described above, in the in-store device management system shown in FIGS. 1 and 2, data (various sensors) of devices (showcase 1, refrigerator 10, etc.) managed by each controller (showcase controller 9, etc.), respectively. (Measurement data, etc.) are collected over time at regular time intervals, and the collected time-series data is transmitted to the in-store PC 8.

In this example, the in-store PC 8 uses the data transmitted as described above to execute the process described below. However, the process is not limited to this, and the process is NA3. Alternatively, the above data may be transferred to an arbitrary information processing device outside the store (for example, each server in FIG. 1 (service headquarters server 7, etc.)) via the network 5 to perform processing. You may If the function as the failure diagnosis apparatus according to the present example is realized by the in-store PC 8, the function can be easily upgraded, and if realized by the NA3 or any information processing apparatus outside the store, the in-store PC 8 can be realized. It becomes unnecessary to provide.

The in-store PC 8 is installed within a specific period (for example, when a showcase and a refrigerator are delivered, installed in the store, and after the power is turned on until the operation of the device becomes stable) within a specific period in which normal data collection is considered possible. During a period from the start after a certain period of time has elapsed to the elapse of an arbitrary period), learning processing (database formation) is performed based on the time-series data (step S1).

That is, in the non-defrosting section (that is, when defrosting is not performed; the temperature inside the showcase, etc., temporarily rises during defrosting, so it cannot be used for learning). The average value and the instantaneous value of the temperature inside the showcase, the average value of the discharge pressure of the refrigerator, and the average value of the temperature outside the store are obtained and stored. Further, the data of the internal temperature pull-down time is naturally collected in the defrosting section (after the defrosting operation, etc.), the average value of the temperature in the store at that time is obtained, and these data are stored (step S2). ).

After collecting the above data within a predetermined period,
Based on this data, a threshold value determination process using a normal distribution curve described later is performed (or a t distribution curve may be used, although not particularly shown / explained), but before that, the discharge pressure of the refrigerator is , And the internal temperature pull-down time are corrected.

That is, of the above data, the discharge pressure of the refrigerator and the pull-down time of the temperature inside the refrigerator are affected by seasonal fluctuations (mainly temperature changes) and fluctuate even if no failure occurs. To go.

Therefore, these data are obtained by obtaining the relationship between each of the collected average values of the discharge pressure of the refrigerator and the pull-down time of the temperature inside the refrigerator, and each influencing factor (mainly the temperature inside / outside the store). Needs to be corrected so that it is not affected by temperature. Since the discharge pressure of the refrigerator is mainly affected by the temperature outside the store, and the internal temperature pull-down time is mainly affected by the temperature inside the store, these relationships are obtained and the collected data is corrected (step S3).

Here, an example of the relationship between the discharge pressure of the refrigerator and the temperature outside the store will be described with reference to FIG. 10, but the relationship between the pull-down time in the refrigerator and the temperature inside the store is the same. And ask.

FIG. 10 is a relational diagram of discharge pressure-outside store temperature created based on the data collected in step S2. The illustrated example is created based on about 15 pieces of data, whereby the regression line (y = 0.
2x + 9.7) (correlation coefficient; 0.76). As shown in the figure, since the discharge pressure data fluctuates with a certain tendency influenced by the temperature outside the store, each collected data is obtained when the temperature outside the store is the same. It must be corrected to a power value. Therefore, a certain temperature (hereinafter, referred to as standard temperature) is determined in advance, and each data of the discharge pressure is corrected to the value at the standard temperature by using the above relational expression (regression straight line).

The internal temperature pull-down time is similarly corrected by obtaining the relational expression with the in-store temperature. Then, next, a process for determining a threshold value is performed (step S
4). The threshold value is obtained using a normal distribution curve. The normal distribution curve is created using the collected data as it is for the showcase interior temperature, and is created using the data corrected in step S3 for the refrigerator discharge pressure and the interior temperature pulldown time.

First, the temperature inside the showcase will be described as an example. FIG. 11 is a diagram showing a histogram and a normal distribution curve created by collecting temperature data (about 40 pieces) inside the showcase actually installed in the supermarket.

As shown in the figure, the inside temperature data is obtained so that a normal distribution curve 53 centered around -1 to -0.5 ° C can be drawn without any particular correction.
A method of obtaining the threshold value using the normal distribution curve 53 will be described below.

First, in general, statistically, when testing whether or not the distribution of certain measurement data is the same as the learned distribution, normalization of the normal distribution of the learning data (enclosed by the normal distribution curve The area where the measured data is located is 1) and the probability that the measured data is located is determined. Generally, values such as 5% and 1% are used. That is, in the example shown in FIG. 11, the threshold value is set so that the area 54 on the right side of the threshold value (indicated by diagonal lines) is about 1% or 5% of the total area surrounded by the normal distribution curve 53. Determine the value. At that time, the final threshold value may be determined in consideration of some margin.

Regarding the discharge pressure of the refrigerator and the pull-down time of the internal temperature, the normal distribution curve is obtained by using the data corrected in step S3, and the threshold value is determined in the same manner as above.

Here, the fact that the correct threshold value can be obtained by performing the above correction will be described by taking the discharge pressure of the refrigerator explained in FIG. 10 as an example. 12
12A is a histogram created using the data before correction and its normal distribution curve, FIG. 12B is an example of the histogram created using the data after correction and its normal distribution curve, and FIG. It is a figure which shows the relationship of FIG.12 (a) and FIG.12 (b).

As is clear from comparison between FIGS. 12 (a) and 12 (b), before the correction, the data outside the store is greatly affected by the outside temperature of the store, but after the correction, the outside temperature of the store is changed. The width of the distribution of the data is quite small because the influence is removed. FIG. 12C is a diagram showing this in a more visually understandable manner.

A more appropriate threshold value can be determined by using the normal distribution curve 56 obtained from the corrected data. The same applies to the internal temperature pulldown time.

As described above, since the data is actually collected on the spot for a predetermined period after the device is installed, and the threshold value is determined based on this data, it depends on the environment in which each device is installed (if the device is not actually installed, (Because there is a part where the influence of the environment etc. is unknown), it is possible to determine the threshold value, and in this example, by further correcting the collected data as necessary, by determining the threshold value using the normal distribution curve, A more appropriate threshold value will be set.

After the threshold is determined as described above, the abnormality (failure) of the device is determined using this threshold. Basically, data is collected at any time or at predetermined time intervals, and this data is compared with a threshold to determine whether it is abnormal or normal. The data collection is the same as in step S2 described above, and in the non-defrosting section, the average value and the instantaneous value of the temperature inside the showcase, the average value of the discharge pressure of the refrigerator, and the average value of the temperature outside the store are obtained. In the defrosting section, the data of the internal temperature pull-down time and the average value of the temperature in the store are obtained. Then, from these data, it is determined whether the evaporator is frosted (ice bank), the heat exchange performance of the condenser is reduced as typified by clogging of the condenser, and the cooling performance is reduced as typified by refrigerant leakage. In addition, here, the threshold relating to the showcase internal temperature is the threshold A, the threshold relating to the discharge pressure of the refrigerator is the threshold B, and the threshold relating to the internal temperature pull-down time is the threshold C.
The following description will be given.

First, the abnormality determination of frost formation (ice bank) on the evaporator will be described. First, it is determined whether or not the average value of the showcase internal temperatures obtained by collecting the data is equal to or more than the threshold value A (step S6).

If it is equal to or larger than the threshold value A (may be equal to or larger than the threshold value A for n consecutive times) (step S6)
YES), although the cause is uncertain, it is determined that some abnormality has occurred and an "abnormality" signal is issued (step S).
7). In this example, considering the possibility of false alarm, even if the "abnormal" signal is issued about once, it is not immediately judged as a failure, and the "abnormal" signal is issued m times in succession. On the other hand (step S8, YES), it is assumed that the temperature inside the showcase is surely rising, and the "evaporator frost (ice bank)" signal is issued (step S9).

On the other hand, the average value of the temperature inside the showcase is
Even if it is less than the threshold value A (step S6, NO),
Instantaneous value of the inside temperature of the showcase (a plurality of time series data, and the average value is obtained based on this)
From this, the rate of change of the internal temperature is obtained (step S10), and when the rate of change exceeds a preset threshold value (when it can be considered that an abnormally rapid change has occurred) (step S10). (S11, YES), an "evaporator frost (ice bank)" signal is issued (step S9).

Also, "frost formation on the evaporator (ice bank)"
Even if the signal is not issued (the threshold is not exceeded), if the temperature inside the warehouse tends to rise, the temperature inside the warehouse rises (in units such as per day, per week, etc.). It is also possible to predict the failure occurrence time and display / notify the prediction result by calculating the rising speed) and calculating the time required to reach the threshold value based on this. By predicting the failure time in this way, the maintenance plan can be accurately constructed.

Next, the abnormality determination of the heat exchange performance deterioration of the condenser will be described. In this case, first, it is necessary to correct the collected data (the average value of the discharge pressure of the refrigerator) using the relational expression obtained in step S3 so as to be the value under the standard temperature environment (step S12).

Then, it is judged whether the corrected discharge pressure data is equal to or more than the threshold value B (step S13). If it is equal to or greater than the threshold B (may be equal to or greater than the threshold B consecutively n times) (step S13, YES), the cause is uncertain, but it is determined that some abnormality has occurred, and “ An "abnormal" signal is issued (step S14). In this example, the "abnormal" signal is set to 1 in consideration of the possibility of false alarm.
Even if it is reported about once, it is not judged to be a failure immediately but "abnormal"
When the signal is issued m times consecutively (step S1
5, YES), it is determined that the discharge pressure of the refrigerator has surely increased, and a "condenser heat exchange performance degradation (clogging)" signal is issued (step S16).

On the other hand, the average value of the temperature inside the showcase is
If it is less than the threshold B (step S13, N
O), the process returns to step S5. Finally, the abnormality determination of the cooling performance deterioration will be described. This is similar to the deterioration of heat exchange performance (clogging) of the condenser, first, the collected data (internal temperature pull-down time) is corrected using the relational expression obtained in step S3 (step S17). Then, it is determined whether the corrected data is equal to or more than the threshold value C (step S18), and if it is equal to or more than the threshold value C (step S18,
YES), an "abnormal" signal is issued (step S19),
When this continues m times (step S20, YES)
Determined as "cooling performance deterioration (refrigerant leakage)" (step S
21).

Even if the alarm such as the "frosting of the evaporator (ice bank)" signal is issued once, as a result of confirmation by the maintenance personnel, there is no abnormality (if it is a false alarm). ) And it is determined that insufficient learning is the cause of the false alarm, the data collected at that time (data used for comparison / judgment with the threshold) is also stored in the database (learning period). It may be added to the database of the data collected in 1.) and the threshold value may be set again.

FIGS. 13 (a) to 13 (d) are visual representations of the above-described learning to determination in the same graph format. Here, the discharge pressure of the refrigerator will be described as an example.

FIG. 13A shows the learning stage (step S3 above). As already described, here, first, a regression line is created in order to obtain the correlation between the temperature outside the store and the discharge pressure from the collected data to create the correction formula. Next, along this regression line, the outside-store temperature of each data is converged to a preset standard temperature. This has already been described with reference to FIG. 10, but FIG. 13 (a) visually shows this convergence on a graph.

FIG. 13B shows the threshold value setting step (step S4 above). Here, the threshold value is determined from the data width of the discharge pressure at the standard temperature, which is converged in the learning stage. In FIG. 13B, the manner of determining the threshold value is visually represented on a graph.

FIG. 13C shows the measurement stage (from step S5 onward; during actual operation), and shows an example of measuring the temperature outside the store when the discharge pressure data is acquired.
FIG. 13D shows the determination stage (in this example, the above steps S12 and S13; other steps S17 and 18 also).

That is, the discharge pressure data measured in the above measuring step is corrected to a value at a preset standard temperature (corrected along a regression line), and the corrected value is compared with a set threshold value. , Is visually represented on the graph.

Finally, the automatic coping process for frost formation (ice bank) on the evaporator will be described. Generally, frost is attached to the evaporator 33 of the showcase even if there is no failure. Therefore, for example, in the case of meat and fresh fish, defrosting operation (heating with a heater and intake of outside air) is performed every 6 hours,
Melting frost that has adhered. However, if the inside of the store is in a high temperature and high humidity state for a long time, or if a failure such as a refrigerant leak occurs, a frost amount more than the limit amount that can be melted by the defrosting operation is attached, and the evaporator 33 is gradually attached. Frost accumulates, and finally it becomes a clogged state called ice bank. Therefore, normally, when the "evaporator frost (ice bank)" signal is issued, maintenance personnel etc. go to the site,
Frost adhered to the evaporator 33 was melted by manually removing frost or by spraying water in some cases.

In this example, frost formation on the evaporator (ice bank)
By carrying out the automatic countermeasure processing described above, it becomes unnecessary for maintenance personnel to go to the site one by one and work. In particular,
For example, after the "evaporator frost (ice bank)" signal is issued, the defrosting operation is immediately performed even if the defrosting operation is not performed (every 6 hours described above), and then for a while. Performs a process such as prolonging the time of one defrosting operation or shortening the interval for performing the defrosting operation (shorter than every 6 hours).

The time-series data collected in the above step S1 may be targeted only at data under a specific condition (for example, opening date). In this way, by setting restrictions on the time-series data to be collected, it is possible to separate disturbance factors and improve the accuracy of threshold setting and determination of presence / absence of a failure.

Here, as described below, a failure due to electronic mail is caused by a network system (for example, the in-store device management system shown in FIG. 1) in which the device for predicting the failure as described above is connected to a network such as the Internet. Prompt maintenance work by transmitting diagnostic / prediction information, performing maintenance accordingly, and automatically creating parts and priority inspection items in the periodic inspection table
Can be done accurately / efficiently. This will be described below.

As described above, an example of the network system according to this example is already shown in FIG. 1 (further, an example of the system configuration in each store is shown in FIG. 2), but here, FIG. 14 and FIG. An example of the network system shown in FIG.

First, the structure common to FIGS. 14 and 15 will be described. First, in both figures, in each store (A store, B
The system configuration of the store is simple, so each showcase 71
And its refrigerator 72 are shown. Showcase 71
2 is a configuration corresponding to the “showcase 1 + showcase controller 9” in FIG. 2, and the refrigerator 72 is a configuration corresponding to the refrigerator 22 in FIG. 2 (the configuration is shown as the refrigerator 30 in FIG. 3). Is. Further, the network connection device 73 in the system configuration in the store is
C8, NA (network adapter) 3, and router 4
It is assumed that the configuration is equivalent to. Further, the network connection device 73 can be connected to a network and can communicate with an external information processing device. In the illustrated example, the network connection device 73 of each store and the headquarters / center 75 (corresponding to the store headquarters server 6). ), Branch server 76, maintenance company server 77 (equipment maintenance company / server 1)
9) is connected by the LAN 78, but is not limited to this.

Then, in FIG. 14, in addition to the above configuration,
For maintenance personnel, mobile phones, notebook PCs, P
A portable information terminal 74 such as a DA is provided so that each store or server can notify failure information to a specific portable information terminal 74 via the LAN 78 and the Internet 79.

Here, if a specific example is explained,
The head office of a certain supermarket (headquarters / center 75) is located in Tokyo, and it constantly collects the operating status information of the showcases 71 of chain stores within its jurisdiction sent from the server 76 of each branch office. The branch office server 76 sequentially collects information on chain stores in the jurisdiction area.

One day, the network connection device 73 of store A
Has performed a failure diagnosis. Network connection device 73 (as described above, failure diagnosis device (for example, in-store PC
8) includes the function of determining the failure content of the showcase or the refrigerator and predicting the failure, as described above, and here, for example, "the condenser 40 of the refrigerator 30 is used. It has been diagnosed as "clogging failure of" and notified of this by e-mail. Network connection device 7
3 sends an electronic mail to the headquarters / center 75, the branch server 76, the maintenance company server 77, etc. Also,
In the case of failure prediction, there is plenty of time before the actual failure occurs, but in the case of failure diagnosis as in the above example, it is often necessary to take immediate action. The electronic mail is also transmitted to the portable information terminal 74.

The maintenance person who received this e-mail said, for example, "If the condenser 40 of the refrigerator 30 is clogged, all the showcases connected to the refrigerator will not be cooled and will be displayed. I decided that I should deal with it immediately because all the products were ruined, and immediately went to store A. When I arrived at store A, I knew that the refrigerator was clogged, so immediately After cleaning the clogged part cleanly, it was confirmed that the temperature transitioned normally, so I contacted the center, branch office and maintenance company and finished the work. ”Etc. Enables swift response before problems occur.

The automatic creation / transmission of electronic mail may be realized as described below, for example, but is not limited to the method described below. First, a message text associated with each failure content and each failure prediction content is stored in advance in a storage device (not shown) in the network connection device 73, and a message text corresponding to the determined failure content is stored from here. Search and retrieve it and insert it in the body of the email. Further, the mail address of the destination to be transmitted (in this example, the headquarters / center 75, branch server 7
(6, maintenance company server 77 email address)
Is registered, and this is read and used as an e-mail destination. In addition, the mobile information terminal 7 of each maintenance person
When the e-mail address of No. 4 and the schedule of each maintenance person are stored and it is determined that it is necessary to notify the maintenance person as well (the judgment criterion is determined to be unnecessary in the case of failure prediction, In the case of a failure determination (that is, it is determined to be necessary if a failure actually occurs), the maintenance person in question is determined from the above schedule, and the email address of the portable information terminal 74 of this maintenance person is also determined. Use as an email destination. The sender of the e-mail is the e-mail address of the store that is stored in advance.

Incidentally, the portable information terminal 7 of the person in charge of maintenance.
4 may be notified by the server 77 such as a maintenance company upon receiving a notification from the network connection device 73.

Next, as shown in FIG. 15, each store is connected to a headquarters / center 75, a branch office server 76, a maintenance company server 77 by a LAN 78, and a server 80 of a parts maker via the Internet 79. This section describes a network configuration that is connected to and enables automatic ordering of parts.

Here, if a specific example is explained,
One super headquarters is in Tokyo, and each branch's server 76
The operating status information of the showcases 71 etc. of the chain stores in the jurisdiction area sent from is always collected. The branch office server 76 sequentially collects information on chain stores in the jurisdiction area.

One day, the network connection device 73 of store B
Has made a failure prediction of "cooling performance deterioration (refrigerant leakage)". The network connection device 73 includes a headquarters / center 75, a branch office server 76, a maintenance company server 7, etc.
The failure prediction content is notified to the e-mail at 7. In this case, since it is a failure prediction, it is determined that it is not necessary to notify the portable information terminal 74 of the maintenance person.

Here, for example, in the branch server 76,
It is assumed that the component names required for repairing the failure are stored in a table format, for example, in association with each failure prediction content. From this, the branch server 76 that received the mail
Searches the above table with "cooling performance deterioration (refrigerant leakage)" as a key to find the corresponding part name. Here, it is assumed that the "pipe through which the refrigerant passes". As a result, the branch office server 76 automatically orders the parts manufacturer / server 80 via the Internet for the parts required for repair (in this example, "the pipe through which the refrigerant flows") in response to the failure prediction. be able to. In response to this, the parts manufacturer promptly sends the "refrigerant-passing pipe" to the maintenance company.

On the other hand, the maintenance company server 77 stores in advance, in association with each failure prediction content, tools and parts necessary for repairing the failure and the repair content in, for example, a table format. You may do it.

FIG. 16 shows an example of this table. As illustrated, in the table 90, the tools / parts 9 required for repair are associated with each failure prediction content (failure location 91).
2. The repair content 93 is stored.

When the maintenance company server 77 receives the failure prediction notification by the electronic mail, it searches the table 90 using this notification as a key and displays the corresponding information on the display. For example, when a notification of "cooling performance deterioration (refrigerant leakage)" is received by the above e-mail, "refrigerant leakage" is indicated as a failure point, and "refrigerant, piping, lubricating oil, ..." As tools / parts required for repair. "," Refrigerant replenishment, honeycomb cleaning, ... "is displayed as the repair content.

As a result, an employee (maintenance worker, etc.) of the maintenance company can send parts required for repair (for example, the above-mentioned "pipe through which the refrigerant flows") from the parts manufacturer side without particular order. However, since the tools / parts required for repair, repair details, etc. can be known by looking at the above display screen, it becomes easier to carry out preparations and actual repair work in advance, and even an inexperienced person can perform repairs efficiently. Will be able to.

Further, when it is determined that there is a failure, the repair may be directly requested from each shop side to a service company or the like that undertakes the repair of each failed portion. For example, the table shown in FIG. 17, for example, is stored in the in-store PC 8 in advance, the table is searched based on the failure determination content, and the corresponding service company is displayed.

[0125] The store clerk or the like of the store can look at this display and contact the service company to request repair.
Even an inexperienced store clerk can quickly respond.

In the table 100 shown in FIG. 17, the contact information 102 (telephone number, company name, etc.) of the service company that undertakes the repair is stored in association with each failure point 101. It should be noted that, on the part manufacturer side, not all ordered parts are shipped immediately. For example, those that do not require urgent replacement of fluorescent lamps are shipped on a regular shipping day, for example. To do. In such a case, for example, if there is failure prediction information from the store B that "the life of the fluorescent lamp is approaching", and the fluorescent lamp has been ordered but has not been shipped, "Refrigerant The fluorescent lamp may be shipped to the maintenance company at the same time as the "passing tube".

In this example, since the failure is predicted, even if it takes some time before the parts arrive, it does not have a great influence on the store or the customer. After receiving the parts, the person in charge of the maintenance company heads to shop B, goes to the shop B, repairs the "tube through which the refrigerant flows", finds the bad part of the fluorescent lamp, and then performs the replacement work.

Thus, parts can be ordered smoothly,
Efficient parts replacement is possible. The configuration of FIG. 14 will be described again. For example, one day, the network-connected device 73 of store A predicts that a "frosting failure" will occur in the near future in the showcase of Case No. 5 of system 1, for example. It is assumed that the server 77 such as the maintenance company is notified by e-mail. In this case, since it is a failure prediction, it is determined that it is not necessary to notify the portable information terminal of the maintenance person.

The application of the server 77 of the maintenance company or the like which received this, for example, focuses on the items related to frost formation in the showcase of case No. Fill out the instructions to check. That is, the maintenance company server 77 stores regular inspection schedule table data, which stores data indicating when, for which shop, and who performs the regular inspection, and the column of priority items. Shall exist. The application first searches for the regular inspection schedule table data by using "A store" as a key, and detects the closest regular inspection date of the A store. Here, it is assumed that this is two days later. Next, in the regular inspection schedule table, the contents notified by the above-mentioned e-mail are stored in the column of the priority item of the data corresponding to "two days later" and "A store". For example, “show 1 case No. 5 showcase frost” is stored.

Two days later, the maintenance person prints out the inspection sheet from the personal computer of the company, and with reference to this, the frost on the showcase of Case No. 5 of system 1 of "A store" is emphasized. You can judge that it should be inspected. Then, head to store A. Upon arrival at store A, the person in charge inspects the evaporator 33 of the showcase of Case No. 5 of system 1 with special emphasis, so even if there is a lot of frost, he or she may inadvertently miss it during normal regular inspection work. There is a possibility that it will happen, but that will not happen and it will surely remove the frost.

By doing so, when a portion that is likely to cause a failure is predicted, this can be surely reflected in the maintenance work at the actual site, the failure can be prevented, and the system life can be shortened. Can be long.

FIG. 18 is a hardware block diagram of an information processing apparatus for executing the learning processing, threshold setting / fault diagnosis / prediction / warning notification processing, etc., and an example of a storage medium storing a program for executing these processing. FIG.

In FIG. 18, the information processing apparatus 60 is C
PU 61, storage unit 62 (including portable storage medium 62a),
The memory 63, the display unit 64, the operation unit 65, the input / output interface unit 66, the communication unit 67, and the like. The information processing device 60 is the in-store PC 8, NA (network adapter) 3, device maintenance company / server 19, and the like.

The CPU 61 is a central processing unit that controls the entire information processing apparatus 60. The storage unit 62 is a storage device such as an HDD that stores at least programs for implementing the learning process and the failure diagnosis / prediction process described above. Alternatively, the storage unit 62 is a combination of a portable storage medium 62a and its drive reading device (for example, a flexible disk (FD) and a flexible disk drive (FD)).
D)). The portable storage medium 62a has the above F
In addition to D, there are CD-ROM, memory card, DVD, MO and the like.

The memory 63 temporarily stores the program stored in the storage unit 62 and stores the program in the CPU.
A RAM or the like to be executed by 61. The display unit 64 is
A display or the like, for example, displays as shown in FIGS. 15 and 16.

The operation section 65 is a keyboard, a mouse or the like, and the user or the like uses the operation section 65 to perform a desired operation. The input / output interface unit 66 is an interface for inputting state data of various devices (showcase, refrigerator, etc.) via some kind of communication line such as the multi-drop type or ring type.

The communication unit 67 has a generally well-known structure that enables communication with other information processing devices via the router (dial-up router) or the network 8, for example.

Here, the present invention is not limited to the device itself such as an information processing device, but a computer readable recording in which programs for implementing various functions according to the present invention are stored when used by a computer. It can also be configured as a medium (storage medium) itself.

In this case, the "recording medium" is, for example, as shown in FIG.
CD- 8 shown as an example of the portable storage medium 62a in FIG.
A portable storage medium such as the ROM 68 and the flexible disk 69, or the like (of course, not limited to these examples, any other medium such as MO, DVD, removable hard disk, etc. may be used as long as it falls within the category of the "portable storage medium"). However, it also includes a “recording medium” of any external information processing device (not shown) capable of communicating via the network 8. Also, of course,
A storage device (RAM / ROM, hard disk, etc.) in the information processing device 60 is also included.

[0140]

As described above in detail, according to the failure diagnosis method and the failure diagnosis apparatus of the present invention, even if there is a difference due to each device or the installation environment, it is actually measured after installation. By correcting the actually measured data based on the obtained data and using statistical processing such as normal distribution, a more appropriate threshold value can be set. As a result, an appropriate threshold can be set for each device / installation location, and the probability of false alarms is extremely low. In addition, if the learning is performed first, then only the obtained thresholds are stored / held and a simple comparison process with these thresholds may be performed. Therefore, it is not necessary to construct a database that accumulates a huge amount of data. And the processing load is very small.

Further, by performing the failure prediction, the maintenance staff or the like can take action before the failure occurs, and the occurrence of the failure can be prevented in advance, so that the loss of the sales opportunity due to the failure can be almost eliminated. In particular, by notifying the outside by e-mail or the like, it becomes possible to promptly grasp the details of the failure and take countermeasures (repair, etc.).

Further, as described above, the device for performing the failure prediction transmits the failure prediction information by e-mail and orders the parts in accordance with the network system connected to the network such as LAN and the Internet. You can quickly respond to maintenance work and maintain the system without breaking the refrigerator or showcase.
In addition, the products displayed in the store are not deteriorated, and the quality can be maintained and the consumer can be provided with a safe product. In addition, by automatically creating priority inspection items in the periodic inspection table based on the failure prediction information, maintenance work can be enhanced and serious breakdowns of refrigerators and showcases can be expected to decrease. . From this, it can be expected that consumers can always be provided with safe and good-quality products.

[Brief description of drawings]

FIG. 1 is a diagram schematically showing an overall configuration of an in-store device management system.

FIG. 2 is a block diagram showing a configuration inside a store 100A shown in FIG.

FIG. 3 is a side sectional view of an example of an open showcase.

FIG. 4A is an external view of the evaporator with frost completely removed, and FIG. 4B is an external view of the evaporator with frost attached.

FIG. 5 is a diagram showing an example of changes over time in temperatures measured by the thermistors for the outside air temperature, the inside temperature, and the controlled temperature.

6A is an external view of the condenser in which dust is completely removed, and FIG. 6B is an external view of the condenser in which dust is clogged.

FIG. 7A is a diagram showing an example of a change over time in the refrigerant temperature measured by a discharge (high pressure) side temperature sensor, and FIG. 7B is a time change of the refrigerant pressure measured by a discharge (high pressure) side pressure sensor. is there.

FIG. 8 is a diagram showing pull-down times for the internal temperature and the controlled temperature after the defrosting operation is started.

FIG. 9 is a flowchart for explaining threshold value determination / fault diagnosis / prediction processing.

FIG. 10 is a relationship diagram of discharge pressure-outside temperature of a store.

FIG. 11 is a diagram showing a histogram and normal distribution curve of showcase interior temperature data.

FIG. 12 shows discharge pressure data of a refrigerator, (a) is an example before correction, (b) is an example of a histogram after correction and its normal distribution curve, and (c) shows a relationship between (a) and (b). It is a figure.

13A to 13D are diagrams visually showing the states from learning to determination described above in the same graph format.

FIG. 14 is a diagram (part 1) for explaining how to notify the failure diagnosis / prediction result to the outside and improve the efficiency of maintenance work according to the present example.

FIG. 15 is a diagram (part 2) for explaining how to notify the failure diagnosis / prediction result to the outside and improve the efficiency of the maintenance work according to the present example.

FIG. 16 is a diagram showing a list of tables stored in advance in a server such as a maintenance company.

FIG. 17 is a diagram showing a list of tables stored in advance in an in-store PC.

FIG. 18 is a diagram showing a hardware configuration of an information processing apparatus that executes various processes according to the present invention, and a diagram showing an example of a storage medium that stores a program that executes these processes.

[Explanation of symbols]

1 showcase 2 air conditioners 3 NA (network adapter) 4 router 5 network 6 store headquarters server 7 Service headquarters server 8 in-store PC 9 Showcase controller 10 refrigerator 11 Air conditioner controller 12 Heat storage controller 13 Heat storage tank 14 vending machines 15 Security system 16 Security Headquarters Server 17 store owner server 18 equipment owner server 19 Equipment maintenance companies / servers 20 Store control controller 21 Lighting controller 22 refrigerator 23 refrigerator 24 doors 25 fire alarm 26 Security Equipment 27 body 28 display shelves 29 opening 30 refrigerator 31 Refrigerant piping 32 open showcase 33 Evaporator 34 blower 35 duct 36 Blowout port (honeycomb) 37 Inside 38 Suction mouth 39 compressor 40 condenser 41 Suction (low pressure) side temperature sensor 42 Suction (low pressure) side pressure sensor 43 Discharge (high pressure) side temperature sensor 44 Discharge (high pressure) side pressure sensor 45 Outdoor temperature thermistor 46 Internal temperature thermistor 47 Thermistor for temperature control 48 Thermistor for defrosting temperature 49 Frost 50 garbage 51 Internal temperature 52 Temperature control 53 Normal distribution curve 54 areas 55 Normal distribution curve 56 Normal distribution curve 60 Information processing equipment 61 CPU 62 storage 62a portable storage medium 63 memory 64 display 65 Operation part 66 I / O interface section 67 Communications Department 68 CD-ROM 69 Flexible disk 71 showcase 72 Refrigerator 73 Network connection equipment 74 Personal Digital Assistant 75 Headquarters / Center 76 branch server 77 Maintenance company server 78 LAN 79 Internet 80 Parts Maker / Server 90 tables 91 Failure point 92 Tools / parts required for repair 93 Repair content 100 tables 101 Failure point 102 Service company contact information

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Hiroyuki Yoshimura             1-1 Tanabe Nitta, Kawasaki-ku, Kawasaki-shi, Kanagawa             Within Fuji Electric Co., Ltd. (72) Inventor Kazuhiro Sakai             1-1 Tanabe Nitta, Kawasaki-ku, Kawasaki-shi, Kanagawa             Within Fuji Electric Co., Ltd. F term (reference) 3L045 AA02 BA01 CA02 DA01 LA14                       LA15 LA16 LA17 LA18 MA02                       MA09 MA11 PA01 PA02 PA04                       PA05                 5H223 AA11 BB02 DD07 DD09 EE06                       EE29 EE30 FF06

Claims (15)

[Claims] 1. Within a predetermined learning period, state data of each freezing / refrigerating showcase including a showcase body and a refrigerator is collected at a predetermined cycle, and a normal distribution or t is collected based on the collected state data.
A failure characterized by obtaining a distribution, deciding each threshold necessary for determining various failures based on the normal distribution or the t distribution, and diagnosing failure occurrence using the threshold after the learning period. Diagnostic method. 2. The pressure on the discharge side of the refrigerator as the state data and the pull-down time of the temperature control temperature of the showcase are corrected after removing the influence of the ambient environment at the time of data collection, and after the correction. The fault diagnosis method according to claim 1, wherein the normal distribution or the t distribution is obtained based on the data of (1). 3. Collected within a predetermined learning period,
The normal distribution or t distribution is obtained based on various time-series state data of each freezing / refrigerating showcase including the showcase body and the refrigerator, and various failures are determined based on the normal distribution or t distribution. A fault deciding device for deciding a necessary threshold, and a fault deciding device for deciding whether or not there is a fault using the threshold decided by the threshold deciding device after the learning period, . 4. The failure determination means compares various state data measured at any time with a corresponding threshold value, and when the threshold value is exceeded, or when the number of times the threshold is exceeded is a preset number of times or more. The failure diagnosis device according to claim 3, wherein it is determined whether or not there is a failure. 5. When the failure determination means determines that there is a failure, an email notifying the occurrence of the failure is sent,
The failure diagnosis device according to claim 3 or 4, further comprising failure occurrence notification means for transmitting to an external server or a mobile terminal via a network. 6. The show case internal temperature data, which is one of the various status data, in the failure determination means,
When an ice bank is diagnosed by exceeding a threshold value determined corresponding to the inside temperature data, it further has an ice bank coping means for coping with the ice bank by changing a defrosting condition. The failure diagnosis device according to any one of claims 3 to 5. 7. The threshold value determination means and the failure determination means perform the threshold value determination / fault determination based on data under a specific condition among the collected various state data. 3. The failure diagnosis device described in 3. 8. The failure diagnosis device according to claim 3, wherein the failure determination means predicts a failure occurrence based on a time series transition of various state data measured at any time and a corresponding threshold value. 9. The failure diagnosis device is connected to at least the controller of the freezing / refrigerating showcase via an in-store network via an in-store information processing device or a network adapter, or via an arbitrary network outside the store. 4. It is realized by an arbitrary information processing device connected outside the store.
9. The failure diagnosis device according to any one of 8 to 8. 10. The method according to claim 8, further comprising failure prediction notifying means for sending the failure prediction information to an external server or a mobile terminal via an electronic mail via a network when the failure occurrence is predicted. Fault diagnosis device. 11. An in-store device management system in which the failure diagnosis device according to claim 10 and an external server are connected to a network, wherein the external server receives the notification of the electronic mail, An in-store device management system characterized in that a part required for repair corresponding to the predicted failure is searched for based on the failure prediction information, and the part is ordered from a server of a parts manufacturer. 12. An in-store device management system in which the failure diagnosis apparatus according to claim 10 and an external server are connected to a network, wherein a server of a maintenance company among the external servers is a regular inspection. Periodic inspection table storage means for storing table data, and, upon receiving the notification of the email, enter the failure prediction information in the priority inspection items of the regular inspection day of the store of the sender of the email in the periodic inspection table data. An in-store device management system, comprising: 13. An in-store device management system in which the failure diagnosis apparatus according to claim 10 and an external server are connected to a network, wherein each of the external servers is a server of a maintenance company. Tools / parts required for repairs in association with the forecast content,
Repair method storage means for storing repair content, and a tool / part necessary for repairing the predicted failure by searching the repair method storage means based on the failure prediction information when receiving the notification of the electronic mail. An in-store device management system, comprising: a repair method display means for displaying and displaying repair content. 14. The in-store information processing device, a contact address storage unit that stores a contact address for repairing in association with each failure, and a contact for repairing the failure determined by the failure determination unit. 10. The failure diagnosis device according to claim 9, further comprising contact display means for searching for and displaying a contact destination storage means. 15. When used in a computer, based on various time-series state data of each freezing / refrigerating showcase, which is composed of a showcase main body and a refrigerator, and is collected in a predetermined learning period, the regularity thereof is obtained. Distribution or t
A function of determining a distribution and determining a threshold value necessary for determining various failures based on the normal distribution or the t distribution; and a function of determining the presence or absence of a failure using the determined threshold value after the learning period. The computer-readable recording medium storing a program for realizing the following.