The invention relates to a field device in which device-related data can be intelligently stored, thereby increasing the lifetime of the data memory and thus the possible service life of the field device.
In process automation technology, field devices are often used to record or influence certain process variables. For the recording of the respective process variable, the field device comprises, depending upon the type, a specific sensor unit in which a corresponding measuring principle is implemented. Depending upon the design, the respective field device type can thus be used, for example, to measure a fill-level, a flow rate, a pressure, a temperature, a pH value, and/or a conductivity. A wide variety of such field devices are manufactured and distributed by the Endress+Hauser corporate group.
In general, a large amount of device-related data must be saved or modified for field devices. These can, for example, be measured values, diagnostic data, or configuration data of the device display. However, data that must be constantly modified, such as meters for operating hours, must also be stored. However, for this purpose, only limited power is available to the respective field device type, mainly due to explosion protection specifications, such as “4-20 mA” or “Profibus®.” Therefore, an EEPROM-based data memory (“electrically erasable programmable read-only memory”), rather than a flash-based data memory, is typically used to store device-related data. In addition, at least one of these data memories is arranged in a disconnectable manner on or in the field device. In this way, sensitive, device-related data may be transferred or assigned to a replacement field device in the event of a defect in the field device so that, for example, recalibration can be avoided on the replacement device.
However, a disadvantage of EEPROM-based memories is that they can be reliably rewritten only for a limited number of 104 to 105 write cycles. In extreme cases, this means that the field device can no longer be used after just a few months, since no further data can be stored. To be sure, FRAM-based data memories (“non-volatile ferroelectric random access memory”) are also highly energy-efficient and reliable for 106 to 1014 write cycles, i.e., they can be rewritten without errors. However, FRAM-based data memories are disproportionately more expensive per memory capacity than EEPROM-based memories.
One possible strategy to increase the usability of the field device with EEPROM-based field device memories despite the limited number of write cycles is to not store dynamic data at all, or to store it in a highly compressed form. In addition, the corresponding data processing unit of the field device can perform so-called “wear leveling” when writing to the data memory, according to which the write cycles are distributed evenly over the memory, so that individual memory cells are not written to too often, while other memory cells are written to only very rarely.
However, in particular with complex measurement applications of a field device that are associated with a high turnover of device-related data (e.g., with a high measurement rate and measurement values with a large volume of data, as is the case with imaging or radar-based measurement principles), such measures are not sufficient to ensure enduring operational readiness of the field device. Therefore, the invention is based upon the object of realizing a field device with extended operational readiness.
This object is achieved by a field device comprising at least the following components:
Thus, the device-related data may be deleted from the buffer memory unit after copying to the long-term memory unit, in order to be able to record new data. Therefore, the idea according to the invention is to accumulate or buffer-store device-related data in a buffer memory that is, for example, at least 10× smaller than the long-term memory unit, but can often be rewritten, up to a “worthwhile” amount of data, and then to store it as a common data packet in a large, but less expensive, long-term memory. Thus, the long-term memory unit needs to be written to less frequently, potentially increasing the lifetime of the entire field device. Thereby, the idea according to the invention can be applied not only to a FRAM-based memory as an internal buffer memory unit and an EEPROM-based memory as a long-term memory unit, but also to any other memory types that have corresponding maximum memory cycles. The idea according to the invention is implemented most efficiently if the data processing unit copies not only a part of the data, but the entire data volume of device-related data as a common data packet from the buffer unit if the defined data volume or time interval is reached.
In particular, so that a replacement device can be set up on the basis of the data stored in the long-term memory unit in the event of a defect in the field device, it is advantageous for the long-term memory unit to be connected to the central data processing unit via a disconnectable and reconnectable interface.
Corresponding to the field device according to the invention, the object underlying the invention is achieved by a method for storing device-related data in the field device according to one of the preceding embodiment variants. Accordingly, the method comprises the following method steps:
The invention is explained in more detail with reference to the following FIGURE. In the FIGURE:
For an understanding of the invention, the structure of a field device 1 according to the invention is shown schematically in
Regardless of the measuring principle or the application, the sensor unit 13 generates, on the one hand, measured values in the form of digital data xs. On the other hand, the sensor unit 13 is usually assigned calibration data xs or application-specific data xs, such as measuring ranges or limit values. In particular, such data xs of the sensor unit 13 must be retrieved recurrently, such that they must accordingly be stored in the field device 1. Corresponding data xd, xc of other units 14, such as configuration data xd of a display 14 or setting data xc of an interface to a higher-level unit 2 such as a process control system or a decentralized database, are also accordingly to be stored in the field device 1.
According to the invention, a central data processing unit 11 of the field device 1 stores such device-related data xs, xd, xc after their receipt in a FRAM-based buffer memory unit 111 assigned to it or in a comparable memory medium that is rewritable without errors for at least 106 write cycles. Thereby, the data processing unit 11 stores incoming data xs, xd, xc in the buffer memory unit 111 until the buffer memory unit 111 is occupied up to a defined data volume [xmax]. Thereby, the maximum data volume [xmax] can be defined as the memory capacity of the buffer memory unit 111. Once such data volume [xmax] has been reached in the buffer memory unit 111, the central data processing unit 11 copies as far as possible all the buffer-stored, device-related data xs, xd, xc as a common data packet [X] from the buffer memory unit 111 to an external long-term memory unit 12, which is based upon an EEPROM memory or a corresponding memory limited to a maximum of 5*105 write cycles, but which results in lower memory costs compared to the FRAM-based buffer memory unit 111. If, for example, very little data xs, xd, xc is received temporarily due to an unmodified measurement situation, a time period of, for example, one hour, one day, or one month can also be defined in addition to the data volume [xmax], after which the data xs, xd, xc is copied automatically, i.e., without having reached the data volume [xmax], from the buffer memory unit 111 to the long-term memory unit 12.
By collecting the device-related data xs, xd, xc in the buffer memory unit 111 according to the invention and then copying them to the long-term memory unit 12 in the form of a common dataset [X], the write cycles to the EEPROM memory 12 are thus significantly reduced, which in turn can extend the operational readiness of the field device 1. At the same time, there is at least no need to retain a large FRAM-based data storage 111, making it possible to keep memory space costs for the field device 1 low. It is also advantageous that, with this method, very dynamic data xs, xd, xc may also be stored in the long-term storage unit 12.
This strategy according to the invention for storing device-related data xs, xd, xc results in a potentially longer service life of the field device, the greater the memory capacity of the long-term memory unit 12 is compared to the buffer memory unit 111. Therefore, it is advantageous if the long-term memory unit 12 ideally has a memory capacity that is at least twenty times greater than that of the buffer memory unit 111. In practice, it is therefore useful for the long-term memory unit 12 to have a size of at least 16 kilobytes, depending upon the application of the field device 1. To further minimize the write cycles to the long-term memory unit 12, in addition to the storage method according to the invention, the central data processing unit 11 can also copy the data packet [X] to the long-term memory unit 12 by “wear leveling” in manner that is as evenly distributed as possible.
With the embodiment of the field device 1 according to the invention shown in
Number | Date | Country | Kind |
---|---|---|---|
10 2021 109 400.3 | Apr 2021 | DE | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/EP2022/057301 | 3/21/2022 | WO |