Power monitoring device for power generating equipment

The power monitoring device uses AI to verify renewable energy usage by learning environmental correlations and detecting deviations, ensuring legitimate power generation in P2P markets.

JP7613319B2Active Publication Date: 2025-01-15TOYOTA JIDOSHA KK
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
JP2021138777
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-08-27
Publication Date
2025-01-15
Estimated Expiration
2041-08-27

AI Technical Summary

Technical Problem

Existing systems fail to verify whether electricity traded in P2P markets is generated using renewable energy, as highlighted by Japanese Patent Application Publication No. 2015-230544.

Method used

A power monitoring device that includes a first registration section for specification information, a sensing section to measure output, a determination section to assess normalcy, and a communication section to verify if the power generation device is using renewable energy, utilizing AI to learn the environment-output correspondence and detect deviations indicating illegal replacements.

Benefits of technology

Enables monitoring of power generation using renewable energy, preventing illegal use of non-renewable devices in P2P markets by identifying and flagging unauthorized power generation equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007613319000001
    Figure 0007613319000001
  • Figure 0007613319000002
    Figure 0007613319000002
  • Figure 0007613319000003
    Figure 0007613319000003
Patent Text Reader

Abstract

To monitor whether power generated by a generation device is power generated by using renewable energy.SOLUTION: An AI edge device (10) includes: a specification information registration part (11) for storing registered specification information of a PV (4) that generates power using solar light; a trace information registration part (12) storing registered trace information up to installing the PV; a sensing part (13) for measuring an output of the PV; a PV output determination part (14) for determining whether the output of the PV is abnormal; and a communication part (15) for communicating with a tracking system (3). The PV output determination part (14) performs: learning a correspondence between an environment of the PV during the generation and an output of the PV; collating an output pattern of the PV estimated by using the learning result with the output pattern of the PV actually measured; and outputting abnormality information to the tracking system in a case where the output patterns of the aforementioned both outputs are mutually different.SELECTED DRAWING: Figure 5
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present disclosure relates to a power monitoring device for a power generation device that generates power using renewable energy. [Background technology]

[0002] In recent years, the liberalization of electricity has made it possible to buy and sell electricity on the market, and the introduction of so-called P2P (Peer to Peer) electricity trading, in which electricity is bought and sold directly between consumers and owners (individuals or corporations) of power generation equipment that generates electricity using renewable energy such as solar power, rather than only between power companies, is being considered. Accordingly, various devices have been proposed for monitoring the output (power generation amount) of power generation equipment.

[0003] For example, Japanese Patent Laid-Open Publication No. 2015-230544 (Patent Document 1) discloses a device that calculates the ratio between the amount of power procured from a power generation facility that generates power using solar light and the amount of power procured from a power company. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2015-230544 A Summary of the Invention [Problem to be solved by the invention]

[0005] In recent years, increasing the ratio of renewable energy use and reducing carbon dioxide emissions have been recognized as social issues. In the future, it is desirable to monitor in some way whether electricity traded as electricity generated using renewable energy was actually generated using renewable energy in the P2P electricity trading market. However, JP 2015-230544 A does not mention such issues or countermeasures.

[0006] The present disclosure has been made to solve the above-mentioned problems, and an object of the present disclosure is to make it possible to monitor whether or not electricity generated by a power generation device is electricity generated using renewable energy. [Means for solving the problem]

[0007] The power monitoring device according to the present disclosure is a power monitoring device for a power generation device that generates power using renewable energy, and includes a first registration unit in which specification information of the power generation device is registered, a second registration unit in which distribution information up until the installation of the power generation device is registered, a sensing unit that measures the output of the power generation device, a determination unit that determines whether the output of the power generation device is abnormal, and a communication unit that communicates with an external system. The determination unit includes a learning unit that learns a correspondence relationship between the environment during power generation of the power generation device and the output of the power generation device, a comparison unit that compares a first output, which is the output of the power generation device estimated using the learning result by the learning unit, with a second output, which is the output of the power generation device measured by the sensing, and an output unit that outputs abnormality information to the external system via the communication unit when it is determined by the comparison unit that the first output and the second output are deviated. The abnormality information includes the specification information registered in the first registration unit, the distribution information registered in the second registration unit, and information indicating that the power generation device is not being used properly.

[0008] In the above power monitoring device, the correspondence between the environment during power generation of the power generation device and the output of the power generation device is learned. Then, when a first output, which is the output of the power generation device estimated using the learning result, deviates from a second output, which is the output of the power generation device actually measured, it is assumed that the power generation device has been illegally replaced with another power generation device, and the power monitoring device outputs abnormality information to an external system. This abnormality information includes specification information and distribution information of the power generation device. Therefore, the external system that receives the abnormality information can recognize that the power generation device specified by the abnormality information is being used illegally. As a result, it becomes possible to monitor whether or not the power generated by the power generation device is power generated using renewable energy. Effect of the Invention

[0009] According to the present disclosure, it is possible to monitor whether or not the electric power generated by the power generation device is electric power generated using renewable energy. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram illustrating an example of a configuration of a power transmission and distribution system. [Diagram 2] FIG. 1 is a diagram illustrating an example of a configuration of a P2P system. [Diagram 3] FIG. 1 is a diagram illustrating an example of a bidding method in a general trading market. [Figure 4] FIG. 1 is a diagram illustrating an example of a bidding method in a direct trading market. [Diagram 5] FIG. 1 is a diagram illustrating an example of the configuration of an AI edge device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference characters, and the description thereof will not be repeated.

[0012] <Overall configuration of power transmission and distribution system> 1 is a diagram showing a schematic example of a configuration of a power transmission and distribution system 1 including a power monitoring device according to the present embodiment. The power transmission and distribution system 1 includes a P2P (Peer to Peer) system 2, a tracking system 3, a solar power generation device (hereinafter also referred to as "PV" (Solar Photovoltaics)) 4, an AI (Artificial Intelligence) edge device 10, a power transmission network PL, a charger / discharger equipment EC, and an electric vehicle EV.

[0013] Traditionally, electricity was supplied exclusively from power plants managed by electric power companies to electricity consumer facilities (for example, factories, company buildings, commercial facilities, homes, stores, etc.) through a power transmission line network PL (system power grid).

[0014] In the power transmission and distribution system 1 according to this embodiment, buying and selling of power between individuals, that is, P2P power trading, is carried out.

[0015] In addition, with the spread of mobile objects such as electric vehicles (EVs) equipped with large-capacity storage batteries, such as electric cars and hybrid cars, charger / discharger equipment EC is installed at various locations in the power transmission network PL and at power consumers in the power transmission and distribution system 1. These charger / discharger equipment EC enable electric vehicles (EVs) to be connected and charge and discharge their storage batteries. Electric vehicles (EVs) can be used as a means of storing electricity, and electricity can be bought and sold using mobile objects such as electric vehicles (EVs) through P2P trading.

[0016] The P2P system 2 is a platform for conducting P2P electricity buying and selling transactions in a P2P electricity trading market in which electric vehicles EVs and electricity consumers participate. The P2P system 2 includes a communication device that communicates with electric vehicles EVs and electricity consumers via a communication network, and a computer that accesses the P2P electricity trading market and makes bids for buying and selling electricity.

[0017] When a mobile object such as an electric vehicle EV and an electricity consumer wish to trade electricity, the electric vehicle EV or electricity consumer submits a bid to the P2P system 2 that manages the P2P electricity trading market in which the electricity trade is to be carried out, with bidding conditions including, for example, the time period in which electricity is to be sold or purchased, the amount of electricity to be sold or purchased per unit time period, and the price at which it is sold or purchased. The P2P system 2 uses an arbitrary algorithm to effect an agreement for an electricity trading transaction between a seller and a buyer who meet the bidding conditions, and processes bids for which no other party meeting the conditions is found as non-contracted.

[0018] The P2P system 2 is configured to be able to communicate with charger / discharger equipment EC at each location and with electric vehicles EVs. Information on the usage status of the charger / discharger equipment EC (whether it can be used, whether it is in use, the amount of charging and discharging performed, etc.) is transmitted from the charger / discharger equipment EC at each location to the P2P system 2. The P2P system 2 is configured to notify a moving object such as an electric vehicle EV of which charger / discharger equipment EC is available for use.

[0019] FIG. 2 is a diagram showing a schematic example of the configuration of the P2P system 2. In the P2P system 2, as shown in FIG. 2, there are a plurality of business agent agents, a plurality of residential agents, and a plurality of mobile agents. Each business agent manages the bids and agreements of large power consumers with a large amount of power demand, such as relatively large facilities, for example, factories, commercial facilities, railway stations, and airports. Each residential agent manages the bids and agreements of small power consumers with a normal amount of power demand, such as homes and small and medium-sized stores. Each mobile agent manages the bids and agreements of mobile bodies, such as electric vehicles (EVs). The electric vehicles (EVs), large power consumers, and small power consumers each execute bids via the agents in charge.

[0020] The P2P energy trading market in this embodiment includes a "general trading market" in which electricity consumers trade electricity transmitted through the power transmission network PL, and a "direct trading market" in which a mobile object moves into an electricity consumer's facility and trades electricity transmitted without passing through the power transmission network PL. Note that a mobile object can transmit electricity through a charger / discharger facility EC connected to the power transmission network PL, so it is also possible to trade electricity in the general trading market.

[0021] As shown in Figure 2, the general trading market receives bids from multiple mobile entities and electricity consumers through business agents, residential agents, and mobile agents, respectively, and a deal is made when the bidding conditions of the seller and buyer match.

[0022] Fig. 3 is a diagram showing a schematic example of a bidding method in a general trading market. In a general trading market, as shown in Fig. 3, multiple buyers and sellers submit bids for pairs of price and energy amount (p, q) and (P, Q) for each unit time slot (1, 2, ... n). Note that a unit time slot is a time span (usually 30 minutes) set in the market, and trading of energy amount is performed for each amount of energy transmitted within the unit time slot (power x length of unit time slot).

[0023] On the other hand, in the direct trading market, one market is established for fixed electricity consumers who have installed charger / discharger equipment EC, as shown in Fig. 2. Each of these direct trading markets may be established on a server managed by each electricity consumer, or multiple direct trading markets may be collectively established on a server.

[0024] Fig. 4 is a diagram showing a schematic example of a bidding method in a direct trading market. In a direct trading market, as shown in Fig. 4, one consumer submits a buy or sell bid for each unit time slot (1, 2, ... n), and buy or sell bids for each unit time slot (1, 2, ... n) are accepted from multiple mobile entities, and a bid that meets the conditions is executed.

[0025] Returning to FIG. 1, the PV 4, the AI ​​edge device 10, and the tracking system 3 will be described. The PV 4 is installed at the facility of each power consumer. The PV 4 is a device that generates electricity using sunlight, which is a type of renewable energy. The PV 4 includes a solar panel that generates electricity through the photoelectric effect, an inverter that boosts the voltage of the electricity generated by the solar panel, and the like. Note that instead of or in addition to the PV 4, a power generation device that generates electricity using renewable energy other than sunlight (for example, wind power, geothermal power, biomass, etc.) may be installed at the facility of each power consumer.

[0026] The AI ​​edge device 10 is a power monitoring device for monitoring the output of the PV 4. The AI ​​edge device 10 is installed, for example, at the output stage of a solar panel in the PV 4 (the part before the voltage is boosted by the inverter) or at the output stage of an inverter (the part after the voltage is boosted by the inverter). Note that, in the case where a smart meter is provided between the facility of each power consumer and the power transmission network PL, the AI ​​edge device 10 may be installed between the smart meter and the PV 4.

[0027] The AI ​​edge device 10 has preregistered specification information and trace information (distribution information) of the PV4 to be monitored. The specification information of the PV4 includes information capable of identifying the specifications and individual PV4, such as the model number and serial number of the PV4. The trace information of the PV4 includes information capable of identifying the distribution route of the PV4, such as the factory shipment of the PV4, the name of the installation contractor, and the handover of the installation work. The AI ​​edge device 10 transmits the specification information and trace information of the PV4 to the tracking system 3.

[0028] The AI ​​edge device 10 uses AI to learn the correspondence between the environment during power generation by the PV4 (e.g., the date and time of power generation, weather, etc.) and the output of the PV4, and uses the learning results to monitor whether the PV4 is being used properly (whether the PV4 really generated power using renewable energy). The AI ​​edge device 10 transmits the monitoring results to the tracking system 3. The configuration of the AI ​​edge device 10 will be described in detail later.

[0029] The tracking system 3 is a platform configured to be able to communicate with each electricity consumer's AI edge device 10 and the P2P system 2.

[0030] When the tracking system 3 receives the specification information and trace information of the PV4 from each AI edge device 10, it stores the received specification information and trace information using a highly tamper-resistant data storage method such as a blockchain. Also, when the tracking system 3 receives the monitoring result of the PV4 from each AI edge device 10, it associates the received monitoring result with the corresponding specification information and trace information, and then manages it using a highly tamper-resistant data storage method such as a blockchain.

[0031] When the tracking system 3 receives a collation request for a certain PV4's specification information from the P2P system 2, it collates whether the specification information and trace information of the PV4 for which the collation request was made are regularly registered based on the information it stores itself, and also returns the monitoring result of the PV4 for which the collation request was made to the P2P system 2.

[0032] <Configuration of AI Edge Device 10> FIG. 5 is a diagram schematically showing an example of the configuration of the AI edge device 10. The AI edge device 10 includes a specification information registration unit (first registration unit) 11, a trace information registration unit (second registration unit) 12, a sensing unit 13, a PV output determination unit 14, and a communication unit 15. The communication unit 15 is configured to be able to communicate with the tracking system 3.

[0033] The specification information registration unit 11 is constituted by, for example, a non-volatile memory. The specification information registration unit 11 stores the specification information (model type, serial number, etc.) of the PV4 to be monitored. The specification information of the PV4 is automatically registered in the specification information registration unit 11, for example, by an operation in which the installer of the PV4 reads the QR code (registered trademark) described on the PV4.

[0034] The trace information registration unit 12 is also configured, for example, by a non-volatile memory, like the specification information registration unit 11. Trace information (information such as factory shipment, installation contractor name, and installation work handover) of the PV4 to be monitored is registered in the trace information registration unit 12. The trace information of the PV4 is registered in the trace information registration unit 12 by the PV4 installation contractor when the PV4 installation work is handed over, for example.

[0035] When the specification information is registered in the specification information registration unit 11, the registered specification information is transmitted to the tracking system 3 via the communication unit 15. Similarly, when the trace information is registered in the trace information registration unit 12, the registered trace information is transmitted to the tracking system 3 via the communication unit 15.

[0036] In addition, the PV4 installation contractor will take a photo of the plate showing the model and serial number of the installed PV4 using a device such as a smartphone, and send the photo along with GPS data to the tracking system 3.

[0037] The tracking system 3 manages the specification information and trace information received from the AI ​​edge device 10, and the photos with GPS data received from the smartphone of the PV4 installation contractor, using a highly tamper-resistant data storage method such as blockchain. The tracking system 3 analyzes the photos with GPS data received from the smartphone of the installation contractor to identify the model and serial number of the PV4 shown in the photo, the installation location, etc., and if the identified information matches the information received from the AI ​​edge device 10, the specification information and trace information of the PV4 are registered in the blockchain as genuine information that has not been tampered with. The PV4 registered in the blockchain by the tracking system 3 becomes available for P2P energy trading in the P2P system 2.

[0038] The sensing unit 13 of the AI ​​edge device 10 measures the output (power generation amount) of the PV 4 and transmits the measurement result to the PV output determination unit 14.

[0039] The PV output determination unit 14 has a function of determining whether the output of the PV4 is abnormal, more specifically, whether the output of the PV4 is actually power generated using sunlight (renewable energy). The PV output determination unit 14 includes an AI learning unit 14a, a collation unit 14b, and an output unit 14c.

[0040] The AI ​​learning unit 14a learns the correspondence between the environment when the PV4 generates electricity and the amount of electricity generated by the PV4 using artificial intelligence (AI). Specifically, the AI ​​learning unit 14a uses a plurality of combinations of the environment when the PV4 generates electricity and the amount of electricity generated by the PV4 as learning data to generate a learning model in which the environment when the PV4 generates electricity is input and the amount of electricity generated by the PV4 is output. The "environment when the PV4 generates electricity" used in the learning data desirably includes an environment that affects the amount of electricity generated by the PV4. Since the PV4 generates electricity using sunlight, it desirably includes an environment related to the amount and intensity of sunlight received by the PV4, such as the date and time and weather. The environment when the PV4 generates electricity is input to the AI ​​learning unit 14a from a device not shown, for example. The amount of electricity generated (output) by the PV4 is input to the AI ​​learning unit 14a from the sensing unit 13.

[0041] The learning by the AI ​​learning unit 14a is continuously performed. As a result, the learning model generated by the AI ​​learning unit 14a is updated to the latest state that takes into account the influence of aging deterioration of the PV4.

[0042] The collation unit 14b estimates the output of PV4 for the environment when PV4 generates power by inputting the environment when PV4 generates power into a learning model that is the result of learning by the AI ​​learning unit 14a. The collation unit 14b calculates the change over time of the estimated output of PV4 (hereinafter also referred to as the "first output pattern"). The collation unit 14b also calculates the change over time of the output of PV4 actually measured by the sensing unit 13 (hereinafter also referred to as the "second output pattern"). The collation unit 14b then compares the first output pattern with the second output pattern.

[0043] For example, if PV4 is used continuously, it is expected that the first output pattern estimated from the learning result and the second output pattern actually measured will be almost the same. On the other hand, if PV4 is illegally replaced with a power generation device (also called an "engine power generation device") that generates electricity using the power of an engine driven by a fossil fuel that is not renewable energy, it is expected that the output pattern of the engine power generation device actually measured by the sensing unit 13 (= the second output pattern) will deviate from the output pattern of PV4 estimated from the learning result (= the first output pattern).

[0044] In view of this, the collation unit 14b determines whether or not the first output pattern and the second output pattern diverge. For example, the collation unit 14b calculates the degree of deviation between the first output pattern and the second output pattern according to a predetermined calculation method, and determines that the first output pattern and the second output pattern diverge when the calculated degree of deviation exceeds a predetermined reference value.

[0045] If the comparison unit 14b determines that the first output pattern and the second output pattern diverge, it is assumed that the PV4 has been illegally replaced with another power generation device, and the output unit 14c outputs abnormality information to the tracking system 3 via the communication unit 15. This abnormality information includes the specification information of the PV4 registered in the specification information registration unit 11, the trace information registered in the trace information registration unit 12, and information indicating that the PV4 identified by the specification information and the trace information has not been used properly.

[0046] When the tracking system 3 receives anomaly information from the AI ​​edge device 10, it registers the PV4 identified by the specification information and trace information contained in the anomaly information in the blockchain as an unauthorized device.

[0047] A PV4 registered as an unauthorized device in the tracing system 3 cannot engage in P2P energy trading in the P2P system 2. Specifically, when the tracing system 3 receives a matching request for a PV4 of certain specification information from the P2P system 2, it checks, based on the information it stores, whether the trace information of the PV4 for which matching is requested has been properly registered and whether the PV4 for which matching is requested corresponds to an unauthorized device.

[0048] Then, if the specification information and trace information of the PV4 for which a matching request has been made have been properly registered and the PV4 for which a matching request has been made is not an unauthorized device, the tracking system 3 returns a response to the P2P system 2 to the effect that electricity trading for the PV4 for which a matching request has been made is permitted.

[0049] On the other hand, if the specification information and trace information of the PV4 for which a matching request has been made have not been properly registered, or if the PV4 for which a matching request has been made is an unauthorized device, the tracking system 3 returns a response to the P2P system 2 to the effect that energy trading for the PV4 for which a matching request has been made is not permitted.

[0050] The P2P system 2 determines whether or not to carry out the energy trade for PV4 according to the response from the tracking system 3.

[0051] As described above, the AI ​​edge device 10 according to this embodiment comprises a specification information registration unit 11 in which specification information of PV4 is registered, a trace information registration unit 12 in which trace information up to the time of installation of PV4 is registered, a sensing unit 13 that measures the output of PV4, a PV output determination unit 14 that determines whether the output of PV4 is abnormal, and a communication unit 15 that communicates with the tracking system 3.

[0052] The PV output determination unit 14 includes an AI learning unit 14a that learns the correspondence between the environment (date, time, weather, etc.) during power generation by PV4 and the output of PV4, a comparison unit 14b that compares the output pattern of PV4 (first output pattern) estimated using the learning results by the AI ​​learning unit 14a with the output pattern of PV4 (second output pattern) measured by the sensing unit 13, and an output unit 14c that outputs abnormality information to the external tracking system 3 via the communication unit 15 when the comparison unit 14b determines that the first output pattern and the second output pattern diverge.

[0053] According to the above configuration, the AI ​​edge device 10 learns the correspondence between the environment when the PV4 generates power and the output of the PV4, and if the output pattern of the PV4 (first output pattern) estimated using the learning result deviates from the output pattern of the PV4 actually measured (second output pattern), it is assumed that the PV4 has been illegally replaced with another power generation device, and the AI ​​edge device 10 outputs abnormality information to the tracking system 3. Therefore, the tracking system 3 that receives the abnormality information can recognize that the PV4 identified by the abnormality information is being used illegally. As a result, the tracking system 3 can monitor whether the power generated by the PV4 is truly power generated using sunlight (renewable energy) depending on whether the tracking system 3 has received abnormality information from the AI ​​edge device 10.

[0054] In particular, the anomaly information transmitted from the AI ​​edge device 10 to the tracking system 3 includes the specification information registered in the specification information registration unit 11, the trace information registered in the trace information registration unit, and information indicating that the PV4 has not been used legitimately. Therefore, the tracking system 3 can easily identify an unauthorized PV4 based on the anomaly information from the AI ​​edge device 10.

[0055] The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is defined by the claims, not by the description of the embodiments described above, and is intended to include all modifications within the meaning and scope of the claims. [Explanation of symbols]

[0056] 1 Power transmission and distribution system, 2 P2P system, 3 Tracking system, 10 AI edge device, 11 Specifications information registration unit, 12 Trace information registration unit, 13 Sensing unit, 14 Output judgment unit, 14a AI learning unit, 14b Collation unit, 14c Output unit, 15 Communication unit, EC Charger / discharger equipment, EV Electric vehicle, PL Power transmission line network.

Claims

[Claim 1] A power monitoring device for a power generation device that generates power using renewable energy, a first registration unit in which specification information of the power generation device is registered; a second registration unit in which distribution information including information capable of identifying a distribution route up to the installation of the power generation device is registered; A sensing unit that measures the output of the power generation device; a determination unit that determines whether an output of the power generation device is abnormal; A communication unit that communicates with an external system, The determination unit is a learning unit that learns a correspondence relationship between an environment that affects an amount of power generated by the power generating device during power generation and an output of the power generating device, and generates a learning model that estimates the output of the power generating device by inputting the environment; a comparison unit that compares a change over time in a first output, which is an output of the power generation device, estimated by inputting the environment into the learning model generated by the learning unit, with a change over time in a second output, which is an output of the power generation device, measured by the sensing unit; an output unit that outputs anomaly information to the external system via the communication unit when the comparison unit determines that a discrepancy occurs between the first output and the second output, A power monitoring device for a power generation unit, wherein the abnormality information includes the specification information registered in the first registration unit, the distribution information registered in the second registration unit, and information indicating that the power generation unit has been fraudulently replaced with another power generation unit.

Citation Information

Patent Citations

  • Anti-cheating system for photovoltaic power generation

    CN103389397A

  • Photovoltaic power generation subsidy cheating prevention device based on neural network algorithm

    CN104345192A

  • RPS institution application / notification information check support method, RPS institution application / notification information check support program, and storage medium

    JP2004334354A

  • Power management device

    JP2015230544A

  • Performance evaluation method of photovoltaic power generation

    JP2018007347A