METHODS FOR CONFIGURING AND OPERATING A THERMAL ENERGY STORAGE SYSTEM AND THERMAL ENERGY STORAGE SYSTEM

Abstract

Provided is a method for configuring a thermal energy storage system including the following steps:

providing a thermal energy storage device for storing heat,

providing a plurality of temperature sensors at different locations of the thermal energy storage device for measuring temperatures at the different locations,

providing a control device of the thermal energy storage system for reading measurement data of the plurality of temperature sensors,

generating a numerical model for at least one first temperature sensor of the plurality of temperature sensors based on the measured temperatures of the plurality of temperature sensors means of machine learning, and

storing the numerical model by a control device, for configuring the thermal energy storage system,

Furthermore, a thermal energy storage system and a method for operating a thermal energy storage system is also provided.

Claims

1. A method for configuring a thermal energy storage system, comprising the following steps: providing a thermal energy storage device for storing heat, providing a plurality of temperature sensors at different locations of the thermal energy storage device for measuring temperatures at the different locations, providing a control device of the thermal energy storage system for reading measurement data of the plurality of temperature sensors, generating a numerical model for at least one first temperature sensor of the plurality of temperature sensors based on measured and/or simulated temperature values of the plurality of temperature sensors by means of machine learning, and storing the numerical model by control device of the thermal energy storage, for configuring the thermal energy storage system.

2. The method according to claim 1, wherein the numerical model is generated on the basis of a temperature distribution of a heat transfer fluid of the thermal energy storage device.

3. The method according to claim 1, wherein the numerical model is generated on the basis of a pressure and/or mass flow distribution of a heat transfer fluid of the thermal energy storage device.

4. The method according to claim 1, wherein the numerical model is generated for multiple temperature sensors of the plurality of temperature sensors.

5. The method according to claim 1, wherein for generating the numerical model, a computational fluid dynamic model and/or a finite element method model and/or a discrete element method model of the thermal energy storage device for representing temperatures of a plurality of volume elements of the thermal energy storage device is used, wherein the numerical model is based on properties of the thermal energy storage device.

6. A thermal energy storage system, comprising a thermal energy storage device for storing heat, a plurality of temperature sensors, distributed at different locations of the thermal energy storage device for measuring physical parameters at the different locations, and a control device for reading measurement data of the plurality of temperature sensors, wherein a numerical model for at least one first temperature sensor the plurality of temperature sensors based on measured physical parameters of the plurality of temperature sensors by means of machine learning is stored in the control device, wherein the control device is configured for predicting a physical parameter at the at least one first temperature sensor by means of physical parameters of at least a group of the plurality of temperature sensors and the numerical model.

7. The thermal energy storage system disaccording to claim 6, wherein the thermal energy storage system is configured by the method of: providing a thermal energy storage device for storing heat, providing a plurality of temperature sensors at different locations of the thermal energy storage device for measuring temperatures at the different locations, providing a control device of the thermal energy storage system for reading measurement data of the plurality of temperature sensors, generating a numerical model for at least one first temperature sensor of the plurality of temperature sensors based on measured and/or simulated temperature values of the plurality of temperature sensors by means of machine learning, and storing the numerical model by a control device of the thermal energy storage, for configuring the thermal energy storage system.

8. A method for operating the thermal energy storage system according to claim 6, comprising the following steps: measuring temperatures at the different locations by a plurality of temperature sensors, processing the measured temperatures with a numerical model by the control device of the thermal energy storage system, and determining a temperature of a location of a first virtual temperature sensor of the thermal energy storage system, based on the numerical model and the measured temperatures.

Description

BRIEF DESCRIPTION

[0041] Some of the embodiments will be described in detail, with reference to the following figures, wherein like designations denote like members, wherein:

[0042] FIG. 1 shows a schematic side view of an embodiment of a thermal energy storage system according to the second aspect of the invention;

[0043] FIG. 2 shows a schematic view of a sensor configuration according to an embodiment of the invention;

[0044] FIG. 3 shows a diagram illustrating the accuracy of virtual sensors in comparison with regular sensors;

[0045] FIG. 4 shows a flow chart of the inventive method according to the first aspect of the invention; and

[0046] FIG. 5 shows a flow chart of the inventive method according to the third aspect of the invention.

DETAILED DESCRIPTION

[0047] Elements with the same function and effectiveness are denoted each in FIGS. 1 to 5 with the same reference numbers.

[0048] In FIG. 1, an embodiment of a thermal energy storage system 1 according to the second aspect of embodiments of the invention is shown in a schematic side view. The thermal energy storage system 1 is configured as an electro thermal energy storage system 1. The thermal energy storage system 1 comprises a thermal energy storage device 2 for storing heat. Within the thermal energy storage device 2, a plurality of temperature sensors 3 is shown. Moreover, one or more not illustrated other sensors, such as pressure sensors, can be provided. A few temperature sensors 3 are configured as virtual temperature sensors 3a by means of the numerical model. The numerical model is stored in a control device 4 of the thermal energy storage system 1 for reading measurement data of the plurality of temperature sensors 3.

[0049] In FIG. 2, a temperature sensor 3 configuration according to embodiments of the invention is shown in a schematic view. The temperature sensors 3 shown as filled circles are used as input data for generating the numerical model by machine learning. The temperature sensors 3 shown as open circles are virtual temperature sensors 3a, generated by the numerical model.

[0050] In FIG. 3, a diagram illustrating the accuracy of virtual temperature sensors 3a in comparison with regular temperature sensors 3 is shown. As can be inferred from this diagram, the real temperature sensor 3 data and the virtual temperature sensor 3a data show a clear correlation. This means that, by means of the numerical model, virtual temperature sensors 3a with high accuracy can be provided.

[0051] In FIG. 4, a flow chart of the inventive method according to the first aspect of embodiments of the invention is shown. In a first step 10, a thermal energy storage device 2 for storing heat is provided. In a second step 20, a plurality of temperature sensors 3 is provided at different locations of the thermal energy storage device 2 for measuring temperatures at the different locations. In a third step 30, a control device 4 of the thermal energy storage system 1 for reading measurement data of the plurality of temperature sensors 3 is provided. In a fourth step 40, a numerical model for at least one first temperature sensor 3 of the plurality of temperature sensors 3 is generated based on the measured physical parameters of the plurality of temperature sensors 3 by means of machine learning. In a fifth step 50, the numerical model is stored in the control device 4 for configuring the thermal energy storage system 1. Now, the thermal energy storage system 1 is configured to compensate a broken temperature sensor 3 with a virtual temperature sensor 3a by means of the numerical model.

[0052] In FIG. 5, a flow chart of the inventive method according to the third aspect of embodiments of the invention is shown. Since the inventive method according to the third aspect of embodiments of the invention is based on the inventive method according to the first aspect of embodiments of the invention, the numbering of the method steps is continued. In a sixth step 60, physical parameters are measured at the different locations by a plurality of temperature sensors 3 of the thermal energy storage system 1. In a seventh step 70, the measured physical parameters are processed with the numerical model by the control device 4 of the thermal energy storage system 1. In an eighth step 80, a physical parameter of a location of a first virtual temperature sensor 3a of the thermal energy storage system 1 is determined by the control device 4, based on the numerical model and the measured physical parameters. In other words, by these means, physical values at different locations within the thermal energy storage device 2 can be determined without the need of generating real measurement data at these locations.

[0053] Although the present invention has been disclosed in the form of preferred embodiments and variations thereon, it will be understood that numerous additional modifications and variations could be made thereto without departing from the scope of the invention.

[0054] For the sake of clarity, it is to be understood that the use of “a” or “an” throughout this application does not exclude a plurality, and “comprising” does not exclude other steps or elements.