A smart electricity meter

Abstract

The present invention relates to a smart electricity meter used to predict electrical energy consumption without the need for a server connection.

Description

[0001] A SMART ELECTRICITY METER

[0002] Technical Field

[0003] The present invention relates to a smart electricity meter used to predict electrical energy consumption without the need for a server connection.

[0004] State of the Art

[0005] In the known state of the art, electrical energy measurement is legally carried out with electricity meters. In recent years, electrical energy measurement has been carried out, especially with electronic electricity meters aka smart meters. These meters have both measurement and recording features. In addition, with the communication connection features on the meters, they can send the data to remote units or terminals. The data collected in the server centers is used for energy billing, estimation of energy demands, and other analyses through different software.

[0006] Efficient use of energy resources is of great importance in energy generation on a macro scale and correct planning of resources. It is a very critical issue for electricity generation, transmission, and distribution operators to predict the load demand that will be happening in the future. On a micro-scale, knowing the energy bill amount in advance is a great benefit to all communities, both technically and economically. Predicting energy usage before the due time is a crucial challenge in our resource-limited world.

[0007] To solve this problem, centralized demand forecasting software running on large servers is available. These software collect the data they need through various communication methods. These data are calculated by various methods on high-power server hardware and forecasting values are generated. There are no products that solve this problem without the need for a centralized server software and communication need.

[0008] In the known state of the art, there are various patent and utility model applications regarding the subject. The invention CN111598330A provides an energy estimation method for household consumers. For three-phase home subscribers in low-voltage networks, it performs load identification by sampling electrical current, and voltage wave models, and calculating power and energy values. It performs statistical operations and forecasts with the weather forecast information and correction data it receives over the internet connection. The invention in question is only a method for three-phase electricity consumers and is not a device that officially uses meter data obtained from metrological measurements and is spread throughout all electricity consumers. To implement this method and use the data obtained, a server connection and internet connection are needed.

[0009] As a result, there is a need to develop a device that can operate on its own without the need for a server connection, which produces its load characteristics for all electrical energy consumers for their use.

[0010] Summary of the Invention

[0011] The purpose of the invention is to provide energy consumers with a measuring device that selfestimates the amount of consumption they will make in the future without the need for central servers. With this invention, electricity meters are used to predict energy usage for consumers and energy provider companies after certain periods with the help of its artificial intelligence and deep learning-based processing capacity.

[0012] Another purpose of the invention is to display these predicted values on the screen of electricity meters. The estimated data is recorded in the device and transferred to the other systems via communication technologies.

[0013] Another purpose of the invention is to reveal a measurement and estimation infrastructure with a very dense resolution spread throughout the country with the broadest perspective in the most efficient way.

[0014] Figures

[0015] Figure 1 : Smart Electronic Electric Meter Principal Block Diagram

[0016] Figure 2: Forecasting Algorithm Flowchart

[0017] 1. Smart electronic electricity meter

[0018] 2. Electronic printed circuit board

[0019] 3. Connection terminal

[0020] 4. Measuring circuits and measuring chip

[0021] 5. Main processor chip

[0022] 6. Embedded basic processing unit

[0023] 7. Embedded memory unit

[0024] 8. External memory unit

[0025] 9. Embedded forecasting unit

[0026] 10. Forecasting processor

[0027] 11. Optical communication port 12. Remote communication port

[0028] 13. Display

[0029] 14. Real-time clock

[0030] 15. Button

[0031] 16. Buzzer

[0032] 17. Circuit breaker

[0033] 18. Current inputs and outputs and voltage connections

[0034] 19. LED

[0035] Detailed Description of the Invention

[0036] The present invention relates to a smart electricity meter used to predict electrical energy consumption without the need for a server connection. The invention is a smart electronic electricity meter (1) that works on a prepaid or credited basis and can predict the consumption values that will occur in the future by itself without connecting to any remote server or any other system, by using artificial intelligence, machine learning by deep learning, and features it contains with the samples it takes from the measured electrical energy values, and performs functions depending on this prediction data, The invention is related to a new smart electricity meter structure that can be used for all types of electrical energy consumers, which selfimproves learning and prediction mechanisms.

[0037] The smart electronic electricity meter (1) contains at least one electronic printed circuit board

[0038] (2) together with its outer casing and mechanical parts. The measuring circuits and measurement chips (4) on this electronic printed circuit board (2) measure the electric current and voltage with the current-voltage connections (18) connected to the connection terminals

[0039] (3). The consumed electrical values are calculated with the help of these circuits. On the electronic printed circuit board (2), there is the main processor chip (5) to perform the functions of the smart electronic electricity meter (1). The embedded basic processing unit (MCU) (6) in the main processor chip (5) processes the electrical values it receives from the measuring circuits and measurement chips (4), stamps it with the date & time information it receives from the real-time clock (14), and displays it on the display (13) for consumers to see. It also records this information in the embedded memory unit (7) on the processor chip (5) and in the external memory unit (8) on the electronic printed circuit board (2). The embedded basic processing unit (6) also sends the requested information and makes the desired changes according to the communication requests received via the optical communication port (11) or the remote communication port (12). The remote communication port (12) is RS485, RS232, CS20, Fiber Optic or Ethernet and the remote communication port (12) is pluggable modular communication units such as PLC, BPL, RF, GSM, GPRS, 3G, 4G or LTE. In cases where the smart electronic electricity meter (1) is de-energized, the basic embedded processor unit (6) with the help of the supercapacitor, battery units, and auxiliary electronic circuits displays the values on the display (13) by pressing the selector button (15) on the electronic printed circuit board (2) and responds to limited requests from the optical communication port (11).

[0040] Under the normal energized working conditions, the embedded basic processing unit (6) generates electrical energy consumption values according to the billing period in the operating mode of credit systems where payment is made against the invoice and stored them in the memory units. In the prepaid operation mode, in addition to the above operations, it calculates the energy values consumed and calculates how much is left of the loaded credit according to the energy values and displays it on the display (13). When the credit is low, it warns consumers by sending an audible warning with the help of a buzzer (16) and a visual warning with the help of an alarm LED (19), and the warning can be seen on the display (13).

[0041] The energy of the consumer whose credit has expired is cut off by sending a command to the circuit breakers (17). When consumers load the credit, circuit breakers (17) close the circuit and provide energy flow to the consumer usage. In cases where more strict safe operation is required, the circuit breakers (17) do not close automatically after the loaded credit, and the authorized persons are expected to press the selector button (15).

[0042] In the operating mode, where payment is made on credit, the embedded basic processing unit (6) does not automatically send a command to the circuit breakers (17). Circuit open and close commands are sent only in response to requests received via the optical communication port (11) or the remote communication port (12).

[0043] The forecasting processor (10) on the electronic printed circuit board (2), which performs machine learning and contains artificial intelligence and the embedded forecasting units (9) on the main processor chip communicate with the basic processor unit and the embedded basic processing unit (6) over i2c, SPI, and other processor buses at certain periods. Metrologically verified energy values (Wh) are transferred to the forecasting units by this communication. The basic processor unit is the main processor, and the embedded basic processor unit (6) is a part of the main processor. The forecasting processor (10) has deep learning algorithms it performs electronically or optically or both as hybrid structure.

[0044] The forecasting process (shown in figure 2) is done on the flow chart line below. a) The embedded forecasting unit (9), and the forecasting processor (10) collect the values of active energy consumption (Wh), active power (W), current (I), voltage (V), and date, hour, minute, and second information from the embedded basic processing unit (6) with a high resolution. With the data set formed by this metrological metering data, an estimation of the moment t0is made. By receiving the data from the metrological meter, the invention surpasses energy analyzers and many similar devices. b) In this layer, artificial intelligence processors (embedded forecasting unit (9) and the forecasting processor (10)) perform analysis, compare the estimated data at the time with the actual values of t0and find the deviation (6). c) In this layer, Al processors (embedded forecasting unit (9), and the forecasting processor (10)) learn and improve themselves. The embedded forecasting unit (9), and the forecasting processor (10) develop themselves and learn the energy consumption habits of the consumer they are connected to through deep learning.

[0045] If data is coming from the external environment via optical communication port (11), and the remote communication port (12), it is included in this process. External correction coefficients, temperature data, and local calendar information on a yearly or monthly basis can be transferred to the Al processor(s) (embedded forecasting unit (9), and the forecasting processor (10)) via remote communication port (12). d) In this layer, artificial intelligence processors (embedded forecasting unit (9), and the forecasting processor (10)) generate Wh forecasted energy values that will occur at the moment t . According to this newly learned situation, the estimation algorithm also updates estimation data other than when it deems necessary. For example, during an hourly forecast, daily, weekly, monthly, and annual forecasts are also changed if necessary. e) The predicted data is written to the external memory unit (8) with the help of the embedded basic processing unit (6). These data, which are stored in the memory unit with date and time information, can be observed and read locally with an optical communication port (11) or remotely with a remote communication port (12).

[0046] Estimation data can be displayed in different formats on the meters. The embedded basic processing unit (6) receives the prediction data, displays it on the display (13), and sends it to different systems via optical communication port (11) and remote communication port (12). The data format differs appropriately from the protocol description used but is expressed in the following form; Value, period (127.57 kWh, Ih)

[0047] In this example, it is estimated that the value after 1 hour will be 127.57 kWh.

[0048] In the prepayment mode, this is represented only by duration. (23d, 18h, 55m).

[0049] For example, it was estimated that the amount of credit in it would expire after 23 days, 18 hours, and 55 minutes. This forecasting may also include a specific day and time. For example, the credit over date&time (17-02-2024 17:40)

[0050] The invention is applicable to the industry. All parts of the invention, including electronic printed circuit boards, plastic sleeves, and connection terminals, can be produced today with different techniques. The forecasting method developed with the invention also realizes the forecasting of gas and water meter data. Therefore, the invention is applicable to industry.

Claims

CLAIMS1. A smart electricity meter (1), for predicting electrical energy consumption without need for a server connection, characterized in comprising;• An embedded basic processing unit (6) structured to process electrical values received from measuring circuit and measurement chip (4), stamps date and time information received from a real-time clock (14), and displays on a display (13),• A forecasting processor (10) structured to perform machine learning and having artificial intelligence and is positioned on electronic printed circuit board (2),• An embedded forecasting unit (9) structured to communicate with the basic processor unit and the embedded basic processing unit (6) over processor buses at certain periods and positioned on the main processor chip (5).

2. A smart electricity meter (1) according to claim 1, wherein said electronic printed circuit board (2) comprises an outer casing and mechanical parts.

3. A smart electricity meter (1) according to claim 1, wherein the embedded basic processing unit (6) structured to send requested information and make desired changes according to communication requests received via optical communication port (11) or remote communication port (12).

4. A smart electricity meter (1) according to claim 3, wherein said remote communication port (12) is RS485, RS232, CS20, Fiber Optic, Ethernet or pluggable modular communication units.

5. A smart electricity meter (1) according to claim 3, wherein said pluggable modular communication unit is PLC, BPL, RF, GSM, GPRS, 3G, 4G or LTE.

6. A smart electronic electricity meter (1) according to claim 1, wherein said forecasting processor (10) having deep learning algorithm structured to perform electronically or optically or both as hybrid structure.

7. A smart electricity meter (1) according to claim 1, wherein said basic processing unit (6) structured to generate electrical energy consumption values according to billing period.

8. A smart electricity meter (1) according to claim 1, wherein said embedded basic processing unit (6) structured to calculate energy values consumed and to calculate how much is left of the loaded credit.

9. A smart electricity meter (1) according to claim 1, characterized in further comprising a buzzer (16) structured to send an audible warning to customer in case of low credit.

10. A smart electricity meter (1) according to claim 1, characterized in further comprising an alarm LED (19) structured to send visual warning to customer in case of low credit.

11. A forecasting method in the smart electricity meter (1) without need for a server connection according to claim 1; characterized in comprising steps of;• Collecting values of taken by metrologically active energy consumption (Wh), active power (W), current (I), voltage (V), and date, hour, and minute via the embedded forecasting unit (9), and the forecasting processor (10) from the embedded basic processing unit (6),• Performing analysis, compare estimated data at time (t- ) with actual values of t0and finding deviation (6) via the embedded forecasting unit (9) and the forecasting processor (10),• Learning and improving values via the embedded forecasting unit (9), and the forecasting processor (10),• Transferring external correction coefficients, temperature data, and local calendar information on a yearly or monthly to the embedded forecasting unit (9), and the forecasting processor (10) via remote communication port (12),• Generating Wh forecasted energy values t at momentvia embedded forecasting unit (9), and the forecasting processor (10),• Writing predicted data to the memory unit (8) with help of an embedded basic processing unit (6),• Observing and reading predicted these data locally with an optical communication port (11) or remotely with a remote communication port (12).

Images