EP3736517B1 - Method for condition monitoring, condition evaluation and automatic maintenance request for refrigerated cabinets - Google Patents

Description

The invention relates to a method for condition monitoring, condition assessment and maintenance of locally widely distributed refrigerated cabinets with a measuring box assigned to the respective refrigerated cabinet, which includes a microcontroller and a transmitter/receiver unit.

In food wholesale and retail, mainly industrial central systems were originally used to operate refrigerators and freezers (composite systems). The condition monitoring of these central large systems is carried out using their own control system or a higher-level building management system. The safety requirements are very high due to the enormous downtime costs in the event of a malfunction. The effort for this continuous condition monitoring is therefore also economically justifiable.

Due to increasingly frequent changes in concepts on the sales floor and the legal requirements (F-gas regulations), the use of self-sufficient small cooling units with environmentally friendly refrigerants is making more and more sense. The smaller mobile, so-called plug-in refrigerators, can be designed as refrigerators and/or freezers. The plug-in refrigerated cabinets have the advantage that they can be used flexibly, as they only require a power connection in the area and can therefore be adapted to the requirements of the end customer in wholesale and retail. This means that end customers can present the refrigerated cabinets according to their needs, for the respective goods, at the optimal location for them on their sales floor. The plug-in refrigeration units have very different design criteria, inconsistent controls with non-standardized interfaces, and often changing details of the technical equipment. The variety of manufacturers and therefore the number of device variants is particularly large in the European market. The end customer expects an optimal maintenance service, which should, if possible, be carried out by one provider of this service and not by different and changing manufacturers of refrigeration cabinets.

Under these circumstances, it is impossible to implement a uniform control standard with standardized status parameters and an equally standardized data transmission interface in the respective refrigeration cabinets with a wide range of customers and technology and in this way to realize continuous status monitoring of the different model types. Integration into the local infrastructure (Wi-Fi or ISDN connection) is often not possible and would require a technician to reconfigure every time the refrigerated cabinet is moved.

The end customer expects trouble-free operation without having to constantly find out about the status of their devices. Although the end customer is generally responsible for the proper cleaning of the refrigeration cabinets and their surroundings, proper preventive cleaning of the refrigeration cabinets and their surroundings cannot always be guaranteed. In addition, any necessary cleaning in the housing is only permitted by trained specialist personnel.

The plug-in refrigerated cabinets are also exposed to a wide variety of environmental conditions. They can be used mobile, so that on the one hand their location can change on the sales floor, but on the other hand it is not uncommon for them to be set up at a completely different sales location. Due to these different environmental conditions, even a simple status estimate based on the operating time is extremely inaccurate.

All of this means that either very low maintenance intervals have to be chosen in order to check the condition of the system and, if necessary Maintenance measures must be carried out, or a relatively high risk of failure of the refrigerated cabinet and the associated costs in the form of loss of goods and/or replacement of the refrigerated cabinet or the unit of the refrigerated cabinet must be accepted.

US 2018/340729 A1 discloses a system with a measurement box for monitoring power consumption and temperature characteristics of cold storage devices.

It is therefore the object of the invention, starting from a method of the type mentioned at the beginning, to develop a particularly easy to install, retrofittable and universally applicable system and a method to be carried out with it and, at the same time, to enable the condition assessment of locally widely distributed refrigeration cabinets of a wide variety of model types , as well as to automatically request maintenance service based on the condition assessment and optimize material procurement.

To solve this problem, the invention proposes a method with a measuring box assigned to the respective refrigerated cabinet, which includes a microcontroller and a mobile radio interface, in which the measuring box is installed between the power connection of a refrigerated cabinet and the power plug of the refrigerated cabinet, the measuring box the electrical current - and voltage characteristics of the refrigeration cabinet are recorded, the recorded characteristics are further processed by the microcontroller into compressed data packets, the data packets are forwarded to a higher-level central server via the mobile radio interface, the data packets provided are evaluated using software and a condition assessment of the refrigeration cabinet is carried out based on the evaluation .

The measuring box is therefore simply connected via plug-and-play between the power supply and the connection plug of the refrigerator. To reduce the amount of data, the measured current and voltage characteristics are compressed into data packets. Not all measured data need to be included. Rather, these can be filtered over time or according to certain events, such as a significant change in power consumption or similar. The performance parameters, such as the active power factor and the active and Reactive power can be calculated. The data packets preprocessed in this way are then forwarded to the higher-level server via the mobile communications interface. With the data packets received, the server can carry out a condition assessment of the refrigerated cabinet using stored samples.

The invention further provides that the measuring box recognizes the model type of the refrigeration cabinet when it is switched on for the first time after installation based on stored sample characteristics and the characteristics recorded after switching on and transmits it to the server. This automatic detection means that it is not necessary to manually set which model type the measuring box is connected to on the measuring box. It also makes the use of the measuring box even more flexible because the connected model types can be changed as often as you like.

The possibility of updating the program and data from a higher-level server via the existing radio interface means that new patterns can be reloaded and the program can be optimized remotely.

Based on the information determined in this way, namely the condition assessment, the location of the measuring box and the model type, as well as a maintenance tour database and personnel database (availability and skills), the optimal time for the next maintenance can now be determined using software. This means that employees' maintenance tours can be optimized without running the risk of the refrigerated cabinet failing due to late maintenance. Because of the high number of monitored refrigeration cabinets and their geographically very wide distribution, there is a high potential for optimization in the condition assessment, in conjunction with the already planned maintenance tours.

It also makes sense if a time-scheduled and location-defined material procurement is automatically initiated based on the data determined in conjunction with a warehouse database. Through this measure, warehousing and material procurement can be optimized in such a way that, on the one hand, storage costs are minimized and, on the other hand, they are always the right ones

Spare parts are available at the right time and in sufficient quantities.

Furthermore, the invention provides that the ambient temperature is additionally recorded by the measuring box and made available to the higher-level server. This allows further conclusions to be drawn about the operating and environmental conditions of the refrigerated cabinet. They also serve to improve model creation in the higher-level computer.

Figur 1:

Ein Blockschaltbild eines Aufbaus zur Durchführung eines erfindungsgemäßen Verfahrens;

Figur 2:

Ablaufdiagramm einer Initialisierung bei einem erfindungsgemäßen Verfahren;

Figur 3:

Ablaufdiagramm bei der laufenden Datenerfassung

Figur 4:

Ablaufdiagramm des serverseitigen Datenempfangs

Figur 5:

Ablaufdiagramm der serverseitigen Datenverarbeitung

Finally, the method according to the invention provides that the measuring box can transmit its location to the server. This measure has the advantage that the measuring box can be used flexibly in different locations. In addition, by determining the location, conclusions can be drawn about the ambient conditions of the refrigerated cabinet. Transmitter/receiver unit and diagrams explained in more detail. Show it:

Figure 1:

A block diagram of a structure for carrying out a method according to the invention;

Figure 2:

Flowchart of an initialization in a method according to the invention;

Figure 3:

Flowchart for ongoing data collection

Figure 4:

Flowchart of server-side data reception

Figure 5:

Server-side data processing flowchart

Figures 6 and 7 : Diagrams of exemplary measured values In Figure 1 a measuring box is designated by the reference number 1. The measuring box 1 has a power supply input 11, a radio interface 12 and a power supply output 13. The power supply output 13 is connected to the power supply input 21 of a refrigerator 2. The measuring box 1 records the performance data of the refrigerated cabinet 2 by means of a current and voltage recording unit 14. In addition, the measuring box 1 has additional sensors 15 which, for example, record the ambient temperature or humidity. The captured data is preprocessed and compressed by a microcontroller 10. The freely programmable microcontroller 10 of the measuring box 1 is equipped with an algorithm that enables the acquisition and evaluation of the raw data, calculation of the performance variables and temporary storage of the calculated data. For example, the microcontroller 10 is able to detect and compensate for incorrect measurements. The data is then sent to a higher-level central server 3 by means of a transmitting/receiving unit 12. The data is transmitted bidirectionally to the higher-level server 3, for example, via the mobile network according to the GPRS/GSM standards or similar. The bidirectional data transmission also makes it possible to implement remote maintenance/remote programming (program update) and remote parameterization (data synchronization).

In this exemplary embodiment, the server 3 comprises two computer units 31, 32, which, depending on the complexity of the overall system, can also be implemented in one physical unit.

The server 3 receives the data via the radio interface 311 of the computer unit 31. The computer unit 31 carries out data processing and forwards the processed data to the computer unit 32 via its interface 321 via an interface 312. The computer unit 32 serves as a master computer (operation control).

The data processing in the individual process steps is described in more detail below. In Figure 2 The initiating start-up routine of the microcontroller 10 is shown as a flow chart. The aim of this start-up routine is in particular to recognize which type of refrigerator is connected to the measuring box 11.

In a first step, an automatic offset correction is carried out when measuring current and voltage. The offset is determined and saved once by the microcontroller 10 during initialization. The initial data is then saved.

The first measurement data is then recorded by the refrigeration cabinet 2. It can be useful to carry out a large number of start cycles in order to determine the most accurate information possible about the type of refrigeration cabinet 2. The data is then compressed, stored internally as characteristic values and finally sent to the higher-level server 3.

Figure 3 shows the process of data collection and processing by the locally installed measuring box. First, measurement data is recorded, and this process is carried out either continuously or at intervals following a specific rule. Once a sufficient amount of data is collected, the data is compressed and stored internally as normal values. The normal values are also sent to the higher-level server 3. By compressing the data beforehand, the amount of data to be transferred is already minimized. Once the normal values have been defined, the subsequently recorded measurement data is compared with the normal values. As long as the measured values do not deviate from the normal value or are still within a specified tolerance band, the measured values are not processed further and the measurement data continues to be recorded cyclically according to a specified rule.

However, if the measured values deviate from the normal value, this data is also compressed and stored internally and sent to the higher-level server.

Figure 4 shows a flowchart of data reception and data allocation at the higher-level server 3. After the data has been received, it is checked in a first step and then, if necessary, converted to the desired data format. The data is then saved internally. It is also checked whether the data transfer was complete. If the data received is not yet complete or is incorrect have been transmitted, data reception continues. If necessary, the server 3 can also request the measurement box 10 to send further data or to send the data again.

Figure 5 shows the flow chart of the data analysis on the central, higher-level server 3. Here, on the one hand, the data from a local measuring box 1 is processed and analyzed and, on the other hand, the data from all widely distributed measuring boxes 1 are processed and analyzed cumulatively.

When processing the data from all measuring boxes 1, data from the individual measuring boxes 1 are first categorized based on their characteristic values. The data is therefore primarily divided into groups and/or classes of different types of refrigerated cabinets. Further specifications are then formed from the data obtained in this way and the data assigned to the individual refrigeration units is divided into further specified clusters. For example, the installation location, the ambient temperature, the frequency of defrosting processes and/or the type and quantity of typical goods filling can be taken into account. The classification is carried out by artificial intelligence, which is able to develop patterns from the massively accumulated data sets and categorize data accordingly. Of course, it is also possible to manually update specific data if this appears useful. The data is then graphically processed and stored in a database.

If special features arise, for example a new cluster formed by artificial intelligence, or a data set does not appear plausible, a special report is created and stored in a database. The special reports can then be reviewed and examined manually and/or by artificial intelligence directly or at regular intervals.

When analyzing the data of a single local measuring box 1, the data is compared with that from the specific cluster to which the local measuring box 1 was previously assigned. These are therefore not the static characteristics of the type of refrigerated cabinet (cf. Fig. 2 ), but about the clusters formed by artificial intelligence. The data stored here is not purely static, but rather can also reveal tendencies regarding the degree of wear and contamination of the refrigerated cabinet 2. For this purpose, the data is compared with each other. If the measurement data is within a tolerance band, nothing needs to be done. The routine maintenance cycle that is assigned to the refrigeration cabinet 2 in the respective cluster can be maintained. If the data deviates from the tolerance band, a message is sent to operational planning so that the changed circumstances can be taken into account and, if necessary, earlier or later maintenance can be carried out. Early detection is particularly important here, so that maintenance work is brought forward if possible and the refrigerator does not break down. Rather, with the method according to the invention, maintenance can be carried out on site during a routine operation if a technician is there (on site anyway) to troubleshoot another refrigerator. In addition, the storage and provision of spare parts for maintenance use can also be optimized based on the data obtained and the insights gained from it. Material and resource planning can be done manually, but also automatically, by the higher-level server 3.

In parallel to the status monitoring described above, a permanent device-specific energy consumption determination is carried out so that comparative information can be obtained by analyzing the incoming data to assess energy consumption. The assessment of the absolute energy consumption is subject to process-dependent accuracy and is of secondary importance. It essentially serves to reduce the energy consumption of the recorded devices through adapted (condition-dependent) maintenance cycles and is therefore a building block for implementing predictive maintenance.

Figures 6 and 7 show exemplary measurement curves that were determined during complex, week-long tests in which various disruption, wear and contamination scenarios were simulated. In Figure 5 A contamination simulation was carried out by covering the ventilation slots on the refrigerator. After a routine defrost period, the cover was removed from the vents. The power consumption of the refrigerator therefore drops relatively abruptly. At the real one This would correspond to the condition after cleaning. The real, i.e. not simulated, pollution occurs gradually and is also very location-dependent due to the pollution conditions in the area. For example, the ventilation slots of a refrigerator located in a fish department usually become dirty much more slowly than the ventilation slots of a refrigerator located in the fruit and vegetable department. The speed of the pollution process can be taken into account in the clustering described previously.

Furthermore, in Figure 6 a performance change in the load-carrying curve can be seen after an overload of goods has been simulated by covering the air outlets in the storage space of the refrigerated cabinet.

This significantly increases the load running time. However, the peak performance drops slightly, as shown in particular Figure 7 is visible.

If you compare the two disruption scenarios "contamination" and "overloading of goods", it becomes clear that energy consumption monitoring is not sufficient because both cases lead to increased energy consumption. While maintenance is necessary when it is dirty, when it comes to covering goods, an instruction to the operator to load the refrigerated cabinet correctly is sufficient. The type of fault can be easily determined based on the performance curves.

In general, a detailed performance analysis can already be used to draw fairly precise conclusions about the condition of the refrigeration cabinets from the individual parameters. Based on the data volumes collected from thousands of refrigeration cabinets using the method according to the invention, these conclusions are further specified, in particular through the use of artificial intelligence. This means that maintenance and repair can be planned and carried out much more efficiently.

Claims (5)

1. Method for condition monitoring, condition assessment, and maintenance of locally widely spaced refrigeration equipment (2), comprising a measuring box (1) which is assigned to the respective item of refrigeration equipment (2) and which comprises a microcontroller (10) and a transceiver unit (12), in which

   - the measuring box (1) is installed between the mains connection of the refrigeration equipment (2) and the power connector of the refrigeration equipment (2),

   - the measuring box (1) detects the current and voltage characteristics of the refrigeration equipment (2),

   - the measuring box (1) additionally detects the ambient temperature,

   - the measuring box (1) identifies the model type of the refrigeration equipment (2), upon first start-up after installation, on the basis of stored pattern characteristics and the characteristics detected after start-up, and transmits said model type to a superordinate central server (3),

   - the identified characteristics are further processed by the microcontroller, to form compressed data packets,

   - the data packets are forwarded via the transceiver unit (12) to the server (3),

   - the ambient temperature is provided to the server (3),

   - the measuring box (1) transmits its location to the server (3),

   - the provided data packets are evaluated by means of software,

   - a condition assessment of the cooling equipment (2) is carried out on the basis of the evaluation, and

   - the optimum timepoint for the next maintenance is ascertained by means of the software on the basis of the condition assessment, the location of the measuring box (1), and the model type, as well as a maintenance tour database and personnel database.
2. Method according to claim 1, characterised in that a temporally terminated and locally fixed material procurement is automatically initiated, on the basis of the condition assessment, the model type of the cooling equipment, and the location of the measuring box, as well as a storage database.
3. Method according to either of the preceding claims, characterised in that the characteristics stored on the server are evaluated by means of artificial intelligence, and the artificial intelligence means is designed to be self-learning, such that patterns are developed and said patterns, in conjunction with historic data, allow for conclusions to be drawn regarding the condition and the use conditions of the cooling equipment.
4. Method according to any of the preceding claims, characterised in that the measuring box (1) additionally detects the atmospheric humidity and provides this to the superordinate server (3).
5. Method according to any of the preceding claims, characterised in that the superordinate central server (3) accesses the measuring box (1) via the transceiver unit (12).

Images