JP2014187839A - Power management device and power management system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a power management device capable of estimating unknown electric equipment even when the unknown electric equipment which is not connected to a measuring instrument is connected to an electric power line.SOLUTION: A power management device is configured to manage (n) pieces of electric equipment provided in a home, and includes: a storage part which stores an electric power pattern database in which symbol information related to waveform patterns of electric power consumptions of the respective pieces of electric equipment are registered corresponding to the respective pieces of electric equipment; an acquisition part which acquires the total electric power consumption in the home and electric power consumptions from (m) measuring instruments provided corresponding to (m) (m<n) pieces of electric equipment; a calculation part which calculates an electric power consumption of electric equipment provided with no measuring instrument based upon the total electric power consumption in the home and the electric power consumptions from the (m) measuring instruments that the acquisition part acquires; and an estimation part which estimates electric equipment calculated based upon the calculation result calculated by the calculation part and the electric power pattern database stored in the storage part.

Description

  The present invention relates to a power management apparatus and a power management system that manage power information related to power consumption in electrical equipment.

  In recent years, from the viewpoint of environmental protection and energy problems, efforts have been made to raise awareness of energy saving (hereinafter also referred to as “energy saving”) of electrical equipment (mainly home appliances) used in the home. Energy saving is intended to reduce energy consumption by efficiently using energy or suppressing the consumption of excess energy. As a system for supporting such efforts, HEMS (Home Energy Management System) has been put into practical use. By using such HEMS, home power management and power control can be performed more easily.

  In order to realize energy saving, first, it is necessary to measure the state of power consumption in electrical equipment. In this respect, it is easy to measure the power consumption state of an electric device by associating the electric device with a measuring device, but the electric power consumption of the electric device can be reduced without using the measuring device. An estimation method has also been devised (Patent Document 1). However, Patent Document 1 requires information related to the energy of the device to be estimated. For example, when there is a personal computer, it is necessary to estimate the power consumption of the peripheral device (for example, printer) of the personal computer from the operation information of the personal computer. That is, it is necessary that the status of the peripheral device is included in the personal computer and the operation information of the personal computer.

Japanese Patent Laid-Open No. 2004-280618

  In this regard, the present invention provides a power management apparatus and power management capable of estimating an unknown electrical device that is not connected to a measuring instrument even when the unknown electrical device is connected to a power line. The purpose is to provide a system.

  A power management apparatus according to an aspect of the present invention is a power management apparatus that manages n electrical devices provided in a home, and is associated with a waveform pattern of power consumption of each electrical device corresponding to each electrical device. Storage unit for storing a power pattern database in which the registered symbol information is registered, power consumption of the entire home, and m measuring devices provided corresponding to m (m <n) electrical devices, respectively. The power consumption of an electrical device that does not have a measuring device based on the acquisition unit that acquires power consumption, and the power consumption of the entire home acquired by the acquisition unit and the power consumption from m measuring devices A calculation unit that calculates the electrical device calculated based on a calculation result calculated by the calculation unit and a power pattern database stored in the storage unit.

  Preferably, when symbol information corresponding to the waveform pattern of the power consumption acquired by the acquisition unit is not registered in the power pattern database, symbol information associated with the waveform pattern is generated, and the generated symbol information Is further provided with a pattern registering unit for registering in the power pattern database.

  Preferably, the estimation unit analyzes the waveform pattern of the power consumption that is the calculation result calculated by the calculation unit, generates symbol information, and the generated symbol information is registered in the power pattern database. It is determined whether or not the condition of the hidden Markov model according to is satisfied.

  A power management system according to an aspect of the present invention is provided in common to a plurality of power management devices that respectively manage power consumption of a plurality of electrical devices respectively provided in a plurality of homes, and a plurality of power management devices, A storage device that stores a power pattern database in which symbol information associated with the waveform pattern of the power consumption of each electrical device is registered corresponding to each electrical device, and each power management device An acquisition unit that acquires the amount and power consumption from m measuring devices provided corresponding to m (m <n) electrical devices out of n electrical devices provided in the house, respectively A calculation unit for calculating the power consumption of an electrical device not provided with a measuring device based on the power consumption of the entire home acquired by the acquisition unit and the power consumption from the m measuring devices; Calculation The including calculation results, the estimation unit that estimates an electrical equipment which is calculated on the basis of the power pattern database stored in the storage device.

  Even when an unknown electrical device that is not connected to a measuring instrument is connected to a power line, the unknown electrical device can be estimated.

It is a schematic diagram which shows the whole structure of the power management system 1 according to embodiment of this invention. It is a schematic diagram which shows the hardware constitutions of the home controller 100 according to embodiment of this invention. It is an external view of the measuring device 400 according to embodiment of this invention. It is a schematic diagram which shows the hardware constitutions of the measuring device 400 according to embodiment of this invention. It is a functional block diagram of the home controller 100 according to the embodiment of the present invention. It is a figure explaining the transition of a hidden Markov model. It is a figure explaining the information regarding the power consumption of refrigerator 200D according to this Embodiment. It is a figure explaining the waveform pattern used as modeling at the time of classifying the waveform pattern according to this Embodiment. It is a figure explaining the case where the waveform pattern used as modeling is produced | generated based on the waveform pattern according to FIG. It is a state transition diagram using the hidden Markov model for estimating the information regarding the power consumption of the refrigerator according to embodiment of this invention. It is a figure explaining 1 section according to the modification 1 of this Embodiment. It is a schematic diagram which shows the whole structure of the power management system according to another form of this Embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the following description, the same parts are denoted by the same reference numerals. Their names and functions are also the same. Therefore, detailed description thereof will not be repeated.

<Configuration of power management system>
Hereinafter, a power management system including one or a plurality of electric devices used in a house will be described as an example. However, the present invention is not applied only to such a power management system. Houses include houses and offices.

  FIG. 1 is a schematic diagram showing an overall configuration of a power management system 1 according to the embodiment of the present invention.

  Referring to FIG. 1, power management system 1 according to the present embodiment is installed in a house such as a house or an office. More specifically, the power management system 1 includes a plurality of home appliances as electric devices that consume power.

  Although not limited to these, FIG. 1 illustrates an air conditioner (air conditioner) 200A, a television 200B, a microwave oven 200C, a refrigerator 200D, a lighting fixture 200E, and the like installed in a house as home appliances. Has been.

  Furthermore, the power management system 1 includes a home controller 100 which is a kind of power management apparatus for monitoring and controlling the electric device 200 via the relay 150, a display unit 102 connected to the home controller 100, and a smart meter 160. Including.

  The home controller 100 can perform data communication with the measuring devices 400A, 400B, 400C, 400E and the like associated with the electric device 200 via the wired or wireless network 401 in cooperation with the repeater 150. In this example, the case where the home controller 100 manages n (5 in this example) electrical devices is shown, and there are m (m <n) measuring instruments 400 (4 in this example). The case where it is provided is shown.

  The smart meter 160 is installed outside the house (outside the house). The smart meter 160 is a watt-hour meter having a communication function, and is connected to the home controller 100 so as to be communicable. The smart meter 160 is provided so that it can receive electric power generated by a power plant of an electric power company via an electric power system and can supply power to each electric device 200 through each electric device 200 and a power line. Then, the smart meter 160 measures the power consumption of the entire home and transmits it to the home controller 100. In this example, a method is described in which the home controller 100 acquires the power consumption of the entire home transmitted from the smart meter 160. However, the home controller 100 is not limited to the smart meter 160, and other measuring devices are used for the entire home. The home controller 100 may be configured to measure the power consumption and acquire the power consumption. For example, the power consumption of the entire house may be measured with a power conditioner, and the home controller 100 may acquire the measured power consumption.

  An arbitrary network can be used as the network 401. For example, Ethernet (registered trademark), PLC (Power Line Communications), or the like can be used as long as it is a wired network. In addition, for a wireless network, for example, a wireless local area network (LAN), ZigBee (registered trademark), Bluetooth (registered trademark), an infrared communication method, or the like conforming to the IEEE 802.11 standard can be used. Further, a plurality of communication methods may be combined. In addition, in this example, although the structure which provided the repeater 150 is shown, it is also possible to set it as the structure which the home controller 100 directly controls as a power management apparatus, without providing the repeater 150. Further, when the repeater 150 is provided, the repeater 150 may have a function as a power management device, and application software for controlling the repeater 150 may be operated on the home controller 100.

  The measuring device 400 is associated with one of the electric devices 200, measures information regarding power consumption in the associated electric device 200, and transmits the measurement information to the home controller 100. Typically, as the measuring instrument 400, a power consumption measuring device that is disposed between the power line 402 and the plug of the electric device 200 and measures the state of power consumption is used.

  In the example shown in FIG. 1, four measuring devices 400A, 400B, 400C, and 400E are electrically connected to the power line 402. Plug 250A of air conditioner 200A is connected to measuring instrument 400A, plug 250B of television 200B is connected to measuring instrument 400B, and plug 250C of microwave oven 200C is connected to measuring instrument 400C. The measuring instrument 400E is connected to the plug 250E of the lighting apparatus 200E. Therefore, measuring instruments 400A, 400B, 400C, and 400E measure information related to power consumption in air conditioner 200A, television 200B, microwave oven 200C, and lighting fixture 200E, respectively. In addition, about the refrigerator 200D, since the measuring device 400 is not matched, the state of power consumption cannot be measured directly.

  The home controller 100 stores, in the hard disk 109, information related to power consumption respectively transmitted from the measuring device 400 or the like associated with each electric device 200 acquired via the repeater 150.

  The home controller 100 provides a user interface that presents a user with a power consumption state in the power management system 1 and receives an instruction regarding power management in the power management system 1 from the user.

  The home controller 100 can also display a graph related to power consumption based on information related to power consumption stored in the hard disk 109. Home controller 100 may be a portable type, may be detachable from a base arranged on a table, or may be fixed to a wall of a room.

  In this example, the configuration in which the home controller 100 acquires data by performing data communication with each measuring device has been mainly described. However, a server that collects data (not shown) is provided, and the server communicates with each measuring device. Thus, the data can be stored, and the home controller 100 can obtain the data stored in the server.

  Next, the hardware configuration of the main home controller 100 in terms of configuring the power management system 1 shown in FIG. 1 will be described.

<Home controller 100>
FIG. 2 is a schematic diagram showing a hardware configuration of home controller 100 according to the embodiment of the present invention.

  Referring to FIG. 2, home controller 100 includes a CPU (Central Processing Unit) 101 that is a processor, a communication interface 106, an output interface 107, an input interface 108, a hard disk 109, and a memory 110. In this configuration, the output interface 107, the speaker 111 of the display unit 102, and the display 103 are connected, and sound and images can be displayed on the display unit 102. The display unit 102 includes a tablet 104 and operation buttons 105, and the input interface 108 receives operation inputs from the tablet 104 and operation buttons 105.

  The CPU 101 is a processing entity that controls the overall processing in the home controller 100, and provides various functions as described below by executing a program stored in advance in the memory 110 or the like. In response to a user operation input from the tablet 104 or the operation button 105 via the input interface 108, the CPU 101 executes processing instructed by the user operation. Such an instruction includes an instruction regarding operation / stop of the electric device 200, an instruction to control power supply to the measuring device 400, an instruction to display a current or past power management state, and the like.

  The display unit 102 is a device that provides a user interface, presents various types of information to the user in accordance with instructions from the CPU 101, and outputs instructions input from the user to the CPU 101. More specifically, the display 103 includes, for example, an LCD (Liquid Crystal Display) or an organic EL (Electro Luminescence) display, and displays an image on its display surface. The tablet 104 detects a touch operation with a user's finger or the like, and outputs a coordinate value indicating a position where the touch operation is performed to the CPU 101. In the present embodiment, a tablet 104 is provided in association with the display surface of the display unit 102. However, the home controller 100 does not necessarily need to include a touch panel, and only needs to be able to present various types of information to the user.

  The operation buttons 105 are input means for accepting user operations, and typically one or more are arranged on the surface of the display unit 102. Typically, the operation button 105 includes a plurality of buttons and keys such as a determination button, a return button, a direction button, and a numeric keypad. When the operation button 105 receives a user operation, the operation button 105 outputs information indicating the user operation to the CPU 101.

  The communication interface 106 performs data communication with the measuring device 400 or the like according to a command from the CPU 101. More specifically, the communication interface 106 includes the Ethernet (registered trademark), the PLC (Power Line Communications), the wireless LAN (Local Area Network) conforming to the IEEE 802.11 standard, the ZigBee (registered trademark), as described above. Bluetooth (registered trademark), an infrared communication system, or the like is used.

  The output interface 107 mediates exchange of internal commands between the CPU 101 and the display unit 102. The input interface 108 mediates exchange of internal commands between the tablet 104 and / or the operation buttons 105 and the CPU 101.

  The hard disk 109 stores various data necessary for information processing in the home controller 100 as a kind of storage means.

  The memory 110 is realized by a RAM (Random Access Memory) that is a volatile storage device, a ROM (Read-Only Memory) that is a nonvolatile storage device, and the like. Store the necessary work data.

  The speaker 111 is an audio device, and outputs audio according to a command from the CPU 101. The clock 112 is a time measuring means and responds to the CPU 101 with the current date and time according to a command from the CPU 101.

  The hard disk (HDD) 109 and / or the memory 110 functioning as storage means may be realized using a storage medium connected via a communication interface. As such storage media, semiconductor memories such as flash memory, mask ROM, EPROM (Electronically Programmable Read-Only Memory), EEPROM (Electronically Erasable Programmable Read-Only Memory), IC (Integrated Circuit) card, CD-ROM (Compact Disc-Read Only Memory) and optical disc storage media such as DVD-ROM (Digital Versatile Disk-Read Only Memory), magneto-optical disc storage media such as MO (Magnetic Optical Disc) and MD (Mini Disc), FD ( Magnetic storage media such as Flexible Disk), magnetic tape, and cassette tape can be used.

  Information processing in the home controller 100 is realized by the CPU 101 executing a program in cooperation with peripheral hardware components. Generally, such a program is installed in advance in the memory 110 or the like.

  Such a program can be provided by being stored and distributed in an arbitrary storage medium. Alternatively, such a program can be provided by downloading from a server device (or other device) connected to the Internet or the like. That is, the program stored in the storage medium is read out, or the program is acquired by downloading from the server device and temporarily stored in the memory 110 or the like. The CPU 101 expands the program stored in the memory 110 into an executable format and then executes the program. Storage media for storing such programs include semiconductor memory media such as flash memory, mask ROM, EPROM, EEPROM, and IC card, optical disk storage media such as CD-ROM and DVD-ROM, and optical media such as MO and MD. Magnetic storage media such as magnetic disk storage media, FD, magnetic tape, cassette tape, etc. can be used.

  Furthermore, instead of installing the program in advance in the memory 110 or the like, the CPU 101 may read and execute a program stored in another system or apparatus.

  Furthermore, after a program read from a storage medium or the like is written to a memory or the like mounted on a function expansion board or function expansion unit mounted on the computer, it is stored in the function expansion board or function expansion unit according to the program. The function according to the present embodiment may be realized by performing all or a part of the necessary processing by a mounted arithmetic unit (CPU or the like).

  Furthermore, not only the functions according to the present embodiment are realized by the CPU 101 executing the program, but also all or part of the processing required by the OS (operating system) executed on the computer according to the program. By performing the above, the function according to the present embodiment may be realized.

  When the function according to the present embodiment is realized by the software as described above, the program itself read from the storage medium or the like, or the storage medium storing the program constitutes one embodiment of the present invention. Become.

  In this specification, the program includes not only a program that can be directly executed by the CPU 101 but also a program in a source program format, a compressed program, and an encrypted program.

  Further, a part of the functions of the home controller 100 described above may be provided in the repeater 150, and the function according to the present embodiment may be realized by controlling the repeater from the home controller using a browsing tool.

<Measurement device 400>
FIG. 3 is an external view of measuring instrument 400 according to the embodiment of the present invention. Here, FIG. 3A shows a perspective view including the socket 4001 of the measuring instrument 400, FIG. 3B shows a side view of the measuring instrument 400, and FIG. A perspective view including a plug 4002 of the container 400 is shown.

  Referring to FIGS. 3A to 3C, measuring instrument 400 is arranged to be inserted between a socket for supplying power flowing through power line 402 and a plug of electric device 200. . More specifically, referring to FIG. 3A, a plug insertion socket 4001 is provided on the surface 4051 of the measuring instrument 400. On the other hand, referring to FIG. 3B and FIG. 3C, a plug 4002 is provided on a surface 4053 which is a surface opposite to the surface 4051 of the measuring device 400. The socket 4001 is plugged into the electric device 200 and the plug 4002 is plugged into a socket (outlet / outlet) for supplying power via a power line 402 provided in the house.

  Since the measuring device 400 is preferably as thin as possible, the thickness of the side surface 4052 is designed to be as small as possible.

  On the surface 4051 of the measuring instrument 400, an LED 4041 and a setting button 4042 are further provided. The LED 4041 displays a data processing state in the measuring device 400. More specifically, the LED 4041 varies the presence / absence of lighting, the presence / absence / cycle of blinking, the emission color, and the like according to the data processing state. The setting button 4042 is an input means for accepting a user operation. When the setting button 4042 is operated by the user, initial setting in the measuring instrument 400 is started.

  FIG. 4 is a schematic diagram showing a hardware configuration of measuring instrument 400 according to the embodiment of the present invention.

  Referring to FIG. 4, measuring instrument 400 includes, in addition to socket 4001, plug 4002, LED 4041, and setting button 4042, a pair of main wirings 4004 and 4005 that electrically connect socket 4001 and plug 4002, A shunt resistor 4003 inserted in the wiring 4005, a power supply unit 4007, a power detection unit 4010, a communication module 4020, and an antenna 4030 are included.

  The power detection unit 4010 detects power flowing from the plug 4002 to the socket 4001. More specifically, the power detection unit 4010 includes a voltage input ADC (Analog to Digital Converter) 4011, a current input ADC 4012, a multiplier 4013, and a digital / frequency conversion unit 4014.

  Voltage input ADC 4011 is connected to main wirings 4004 and 4005 via wirings V1P and V1N, respectively. The voltage input ADC 4011 outputs a digital signal indicating a voltage (potential difference) generated between the main wirings to the multiplier 4013.

  The current input ADC 4012 is electrically connected to both ends of the shunt resistor 4003 inserted in the main wiring 4005 via the wirings V2P and V2N. The shunt resistor 4003 is a minute (several hundred micro Ω) resistor used for measuring a flowing current value. The current input ADC 4012 outputs a digital signal indicating the current value of the current flowing through the shunt resistor 4003 to the multiplier 4013.

  The multiplier 4013 multiplies the digital signal (voltage value) from the voltage input ADC 4011 by the digital signal (current value) from the current input ADC 4012, and a value obtained as a result (power consumption / unit: W or kW). Is output to the digital / frequency converter 4014.

  The digital / frequency converter 4014 converts the digital signal from the multiplier 4013 into a frequency signal, and outputs the resulting frequency signal to the communication module 4020.

  The power supply unit 4007 supplies power to each component of the measuring device 400. The power supply unit 4007 is connected to the main wirings 4004 and 4005, and uses part of the power flowing from the plug 4002 to the socket 4001 as power for operation of the measuring device 400. The power supply unit 4007 converts the AC power into DC power, and then supplies the DC power to the power detection unit 4010 and the communication module 4020.

  The communication module 4020 transmits a radio signal indicating the power consumption in the electric device connected to the socket 4001 calculated by the power detection unit 4010 via the antenna 4030. More specifically, the communication module 4020 includes a CPU 4021, a ROM 4022, a RAM 4023, a GPIO (General Purpose Input / Output) 4024, and a wireless RF (Radio Frequency) unit 4025.

  The GPIO 4024 receives the frequency signal input from the digital / frequency conversion unit 4014 and outputs information on the frequency signal to the CPU 4021.

  The CPU 4021 converts the frequency signal information from the GPIO 4024 according to a predetermined logic, and outputs the result to the wireless RF unit 4025. The wireless RF unit 4025 generates a wireless signal by modulating a carrier wave based on a data conversion result from the CPU 4021. A wireless signal generated by the wireless RF unit 4025 is transmitted via the antenna 4030.

  The CPU 4021 implements the processing as described above by executing a program stored in the ROM 4022 in advance. The RAM 4023 stores work data necessary for the CPU 4021 to execute the program.

<Measurement information>
The measuring instrument 400 basically measures the power consumption (unit: W or kW) consumed by the connected electrical device as information regarding power consumption. By integrating the power consumption over a predetermined time, the power consumption amount (unit: Wh or kWh) of the electric device is calculated.

  The home controller 100 can display both the power consumption and the power consumption of each electric device. Furthermore, it is also possible to group a plurality of electric devices and display the power consumption and power consumption for the entire group.

  Therefore, an example of mounting measurement information about the corresponding electrical device 200 transmitted from the measuring device 400 is as follows (however, it is not limited to the following example).

  (1) The measuring device 400 measures the power consumption in the electric device 200 every predetermined cycle (for example, every second), and transmits the measured power consumption as measurement information in the same transmission cycle as the measurement cycle.

  (2) The measuring device 400 measures the power consumption in the electric device 200 every predetermined cycle (for example, every second), and is measured between the previous transmission and the current transmission at a transmission cycle longer than the measurement cycle. The plurality of power consumptions are transmitted as measurement information.

  (3) The measuring device 400 measures the power consumption in the electric device 200 at predetermined intervals (for example, every second), and is measured between the previous transmission and the current transmission at a transmission cycle longer than the measurement cycle. The average power consumption obtained by averaging the plurality of power consumptions is transmitted as measurement information.

  (4) In (2) or (3), transmission is performed after adding to the measurement information a power consumption amount obtained by integrating a plurality of power consumptions measured from the previous transmission to the current transmission. Furthermore, you may add the average value, minimum value, maximum value, etc. of some power consumption.

  Although time information acquired by the measuring device 400 by some means may be added to the measurement information, generally, when the home controller 100 receives the measurement information, the time at that time is acquired from the clock 112, The data is stored in the hard disk 109 in association with the received measurement information.

  An arbitrary protocol can be adopted for exchanging measurement information between the home controller 100 and the measuring device 400. Typically, a configuration is adopted in which the measuring device 400 broadcasts a packet including its own measurement result, and the home controller 100 receives this packet. However, the home controller 100 may periodically poll each measuring device 400.

<Functional configuration>
FIG. 5 is a functional block diagram of home controller 100 according to the embodiment of the present invention.

  Referring to FIG. 5, here, home controller 100 shows a case where CPU 101 implements various functional blocks in cooperation with other hardware or the like based on a program or the like stored in memory 110. Has been.

  Specifically, home controller 100 includes an information acquisition unit 300, an information database 301, a power pattern extraction unit 302, a symbol information generation unit 303, and a power pattern database 304. The information database 301 and the power pattern database 304 are stored in the HDD 109.

  The information acquisition unit 300 acquires information related to the total power consumption of the electrical devices in the house from the smart meter 160. Further, information on power consumption related to each corresponding electric device is acquired from each measuring device 400. Then, the information acquisition unit 300 stores the acquired information regarding power consumption in the information database 301.

  The power pattern extraction unit 302 extracts a waveform pattern of power consumption according to a certain rule for the electrical device based on information regarding power consumption for each electrical device stored in the information database. And it outputs to the symbol information generation part 303.

  The symbol information generation unit 303 generates symbol information based on the extracted waveform pattern. In this example, the symbol information generation unit 303 generates symbol information used in a hidden Markov model, which will be described later, based on the extracted waveform pattern. The hidden Markov model will be described later. Then, the symbol information generation unit 303 registers the generated symbol information in the power pattern database 304 as a waveform pattern of the power consumption of the electric device. In other words, the power pattern extraction unit 302 and the symbol information generation unit 303 have a function as a pattern registration unit that generates symbol information associated with the waveform pattern of the power consumption of the electrical equipment and registers it in the power pattern database 304.

  Furthermore, home controller 100 further includes a difference calculation unit 306 and a pattern estimation unit 308.

  The difference calculation unit 306 performs a process of subtracting the information regarding the electric device measured by each measuring instrument 400 from the information regarding the total power consumption of the electric devices in the house stored in the information database 301. By this processing, information related to the power consumption of the electrical device that is not measured by the measuring device 400 can be acquired.

  The pattern estimation unit 308 estimates an electrical device based on the information regarding power consumption acquired by the difference calculation unit 306 and the symbol information registered in the power pattern database 304. Specifically, the pattern estimation unit 308 outputs, to the power pattern extraction unit 302, information related to the power consumption of the electrical equipment that has not been measured by the measuring device 400 acquired by the difference calculation unit 306. Then, the power pattern extraction unit 302 extracts a power consumption waveform pattern according to a certain rule based on the information regarding the electrical equipment output from the pattern estimation unit 308. And it outputs to the symbol information generation part 303. The symbol information generation unit 303 generates symbol information based on the extracted waveform pattern, and outputs the generated symbol information to the pattern estimation unit 308. The pattern estimation unit 308 determines whether the symbol information generated by the symbol information generation unit 303 satisfies the condition of the state transition diagram according to the symbol information registered in the power pattern database 304.

  If the pattern estimation unit 308 determines that the symbol information output from the symbol information generation unit 303 satisfies the condition of the state transition diagram according to the symbol information registered in the power pattern database 304, the pattern estimation unit 308 includes the symbol information. Estimate the associated electrical device and output it. Thereby, even if it is an electric device which is not measured by the measuring instrument 400, that is, an unknown electric device, it is possible to estimate the electric device and manage information regarding power consumption as the estimated electric device.

Next, the hidden Markov model will be described.
A hidden Markov model is one of probabilistic models and is an algorithm that estimates an unknown parameter from observable information on the assumption that the system is a Markov process with an unknown parameter.

FIG. 6 is a diagram for explaining the transition of the hidden Markov model.
Referring to FIG. 6, in speech pattern recognition, here, a transition of a hidden Markov model that distinguishes between alphanumeric 1 (One) and 2 (Two) is shown. Here, the syllables of One are in the order of w → ah → n. On the other hand, Two is in the order of time t → uw.

  The transition diagram has a plurality of states, and the state transitions according to the input voice pattern, and recognizes One or Two and ends.

For each state, the transition criterion is statistically learned in advance.
For example, in the case of One, when a voice pattern of w is input after the start, the state transitions to the w state. For example, although the sound of w varies depending on the person, since the transition to the w state is statistically determined, it is possible to transition correctly even if there are individual differences. In the w state, input of the next ah speech pattern is awaited. Depending on the case, the length of One is different, and therefore, the waiting state is entered. Next, in the ah state, input of n sound patterns is awaited. Next, the state transitions to the n state according to the input of the n speech patterns, and One is recognized. The same applies to the case of Two.

  Here, options other than One and Two are conceivable, but in that case, it is possible to add a new option or to transition from all states to the garbage state. In this case, when an audio pattern that is not expected in each state is input, the state can be changed to the garbage state, for example, “disagreement”.

Here, the syllable and the predetermined voice pattern are associated with each other.
In the present embodiment, symbol information and a waveform pattern of a predetermined power consumption are associated with each other using the principle of the hidden Markov model.

  For example, when a waveform pattern of a certain amount of power consumption is input, certain symbol information is associated and registered. When a plurality of waveform patterns having different power consumptions are input, they are registered as different symbol information. A syllable may be registered as symbol information.

<Registering electrical equipment>
Next, registration of symbol information in power pattern database 304 according to the present embodiment will be described.

  FIG. 7 is a diagram illustrating information regarding power consumption of refrigerator 200D according to the present embodiment.

  With reference to FIG. 7, here, for example, a case where measuring device 400 and refrigerator 200D are connected and information regarding power consumption of the refrigerator is acquired will be described. Note that the information regarding the power consumption of the refrigerator is measured by the measuring device 400 at least for the time necessary to generate the symbol information. Note that the information to be acquired may be measured for a certain period (for example, one week), and only the period in which the measurement result is analyzed to be a regular pattern may be acquired.

  The power pattern extraction unit 302 extracts a power consumption waveform pattern based on the acquired information on power consumption. In this example, as an example, the divided waveform patterns are divided every predetermined time, the divided waveform patterns of power consumption are analyzed, and the divided waveform patterns are classified. In the analysis and classification, for example, it is possible to classify the waveform pattern based on the similarity to the compared waveform pattern by comparing with the waveform pattern that is modeled according to a predetermined algorithm.

  In this example, information on power consumption measured for 25 seconds as a whole in time series is shown, and a waveform pattern divided every 1.6 seconds as a predetermined time is shown.

  In the divided waveform patterns, waveform patterns with high similarity are classified, and in this example, the waveform patterns PA, PB, and PC are classified.

  FIG. 8 is a diagram illustrating a waveform pattern that is used for modeling when the waveform pattern according to the present embodiment is classified.

  Referring to FIG. 8, here, a case where a plurality of waveform patterns divided every predetermined time is superimposed is shown.

  Specifically, four waveform patterns PB1 to PB4 that are approximate in outline are shown. Here, four waveform patterns are shown, but the number is not particularly limited to four, and may be generated based on a plurality of waveform patterns. Note that whether or not the outer shape is approximated is statistically determined by acquiring a plurality of sampling numbers. Specifically, for example, the feature amount of the waveform pattern may be calculated and determined.

  FIG. 9 is a diagram for explaining a case where a waveform pattern to be modeled is generated based on the waveform pattern according to FIG.

  Referring to FIG. 9, for example, a case where a waveform pattern to be modeled is generated by averaging a plurality of waveform patterns in FIG.

  Then, for example, the generated waveform pattern to be modeled is used as a reference, the waveform pattern is compared with another waveform pattern, and if the degree of similarity is high, it can be recognized as a waveform pattern belonging to the same category.

  The waveform pattern PA and the waveform pattern PC can be classified by the same method.

  With this method, referring again to FIG. 7, symbol information generation unit 303 generates symbol information based on the extraction result of power pattern extraction unit 302. In this example, symbol information “PA, PB, PA, PB, PC” is generated in accordance with the order of the waveform patterns.

  FIG. 10 is a state transition diagram using a hidden Markov model for estimating information related to power consumption of the refrigerator according to the embodiment of the present invention.

  Referring to FIG. 10, a state transition diagram of “PA → PB → PA → PB → PC” based on the symbol information generated by symbol information generation section 303 is shown. When the condition for the state transition is satisfied, it can be estimated that the refrigerator is used.

  Then, the symbol information that is the basis of the state transition diagram is registered in the power pattern database 304 in association with the electric device “refrigerator”.

  In addition, in this example, the case where it was known that the electric device connected to the measuring device 400 in advance was the refrigerator 200D was described. That is, the case where the measuring device 400 is associated with the electric device has been described. On the other hand, there may be a case where the measuring device 400 is not associated with the electric device. In that case, for example, when the symbol information is registered in the power pattern database 304, an interface screen for registering the name of the electrical device associated with the symbol information may be displayed. For example, on the interface screen, the user may input “refrigerator” to be associated and registered.

  The electric device is not limited to the “refrigerator”, and other electric devices can be registered according to the same method.

<Estimation of electrical equipment>
For example, in the difference calculation unit 306, it is assumed that the information related to the power consumption of the electrical device that is not measured by the measuring instrument is as described in FIG.

  The pattern estimation unit 308 outputs the information to the power pattern extraction unit 302, a predetermined waveform pattern is extracted, and the symbol information generation unit 303, for example, “PA, PB, PB..., PA, PB, A symbol signal “PB... PC” is generated and output to the pattern estimation unit 308.

  The pattern estimation unit 308 refers to the power pattern database 304 based on the symbol information output from the symbol information generation unit 303 and determines whether or not the condition of the state transition diagram based on the registered symbol information is satisfied. The

  If it is determined by referring to the power pattern database 304 that the condition of the state transition diagram based on the registered symbol information is satisfied, the unknown electrical device is registered in association with the symbol information that satisfies the condition. It is estimated that this is an electrical device.

  Then, the pattern estimation unit 308 outputs the estimation result. In this example, information related to power consumption that is not measured by the measuring device is output as “refrigerator”.

  Thereby, even if an unknown electrical device is simply connected to the power line by home controller 100 according to the present embodiment, the electrical device is estimated and information on the estimated power consumption of the electrical device is managed. Is possible.

  In this example, regarding a refrigerator as an electric device, a power pattern extraction unit 302 extracts a waveform pattern according to information on power consumption, a symbol information generation unit 303 generates symbol information corresponding to the waveform pattern, and power Although the example in the case of registering the symbol information in the pattern database 304 has been described, the symbol information is generated according to the same method for other electric devices such as an air conditioner, a television, a microwave oven, a lighting fixture, etc. Thus, the generated symbol information can be registered in the power pattern database 304.

  Then, when information related to the power consumption of the electrical device is acquired regarding the unknown electrical device, the condition of the state transition diagram according to the symbol information registered in the power pattern database using the hidden Markov model as described above It is possible to estimate an unknown electrical device based on the determination result.

(Modification 1)
In the above example, the case of generating symbol information according to the waveform pattern divided every 1.6 seconds as the period of one waveform pattern has been described. However, for example, by shortening the sampling period, the accuracy is high. Estimation is possible.

FIG. 11 is a diagram illustrating one section according to the first modification of the present embodiment.
Referring to FIG. 11, it is possible to estimate with high accuracy by acquiring a waveform pattern with a sampling frequency of 60 Hz (16.67 ms).

In this example, waveform patterns PD1 and PD2 every 16.67 ms are shown.
Symbol information according to the waveform pattern can be generated and registered in the power pattern database 304 according to the same method as described above.

  On the contrary, it is possible to simplify the calculation process for estimating the electric equipment by increasing the sampling period, and to estimate it at high speed.

(Another form)
In the above configuration, the method of estimating an unknown electrical device in one home has been described. However, an unknown electrical device is estimated by providing a power pattern database or the like on an external server or the like and sharing it in a plurality of homes. It is also possible to adopt a configuration.

FIG. 12 is a schematic diagram showing an overall configuration of a power management system according to another embodiment.
Referring to FIG. 12, here, there is shown a configuration in which a plurality of houses HA and HB are provided and a common server 500 is provided outside the house.

The houses HA and HB are provided with a plurality of electric devices.
In addition, home controllers 100A and 100B, which are types of power management devices for monitoring and controlling electric devices provided in the houses HA and HB, are provided.

  Further, the house HA is provided with a plurality of electrical devices (n) and (n-1) measuring devices are provided. Therefore, no measuring instrument is connected to one electrical device.

  The home controller 100A includes an information acquisition unit 300A, a difference calculation unit 306A, and a pattern estimation unit 308A.

  The house HB has the same configuration as the house HA, and is provided with a home controller 100B for monitoring and controlling a plurality of electrical devices. The configuration of the home controller 100B is the same as that of the home controller 100A.

  The server 500 includes an information database 502, a power pattern extraction unit 504, a symbol information generation unit 506, and a power pattern database 510. The information database 502 and the power pattern database 510 are stored in an HDD that is one of storage devices provided in the server 500.

  That is, the server 500 is provided with the functions that the home controller 100 has in common for each home controller.

The processing of the system will be described.
The information acquisition unit 300A acquires information related to the total power consumption of the electrical devices in the home from a smart meter provided outside the home. Further, information on power consumption related to each corresponding electric device is acquired from each measuring device 400. Then, the information acquisition unit 300A transmits the acquired information regarding power consumption to the server 500, and the information is stored in the information database 502 of the server 500.

  The power pattern extraction unit 504 in the server 500 extracts a waveform pattern of the power consumption according to a certain rule of the electrical device based on the information regarding the power consumption related to each electrical device stored in the information database 502. Then, the information is output to the symbol information generation unit 506.

  The symbol information generation unit 506 generates symbol information used in the hidden Markov model based on the extracted waveform pattern. Then, the symbol information generation unit 506 registers the generated symbol information in the power pattern database 510 as a waveform pattern of the power consumption of the electric device.

  In addition, the difference calculation unit 306A of the home controller 100A performs a process of subtracting information related to the electrical equipment measured by each measuring instrument 400 from information related to the total power consumption of the electrical equipment in the home acquired by the information acquisition unit 300A. By this processing, information related to the power consumption of the electrical device that is not measured by the measuring device 400 can be acquired.

  The pattern estimation unit 308A of the home controller 100A estimates an electrical device based on the information related to power consumption acquired by the difference calculation unit 306A and the symbol information registered in the power pattern database 510 of the server 500. Specifically, the pattern estimation unit 308A outputs information related to the power consumption of the electrical device that is not measured by the measuring device 400 acquired by the difference calculation unit 306A to the power pattern extraction unit 504 of the server 500.

  The power pattern extraction unit 504 extracts a waveform pattern of the power consumption according to a certain rule based on the information regarding the electrical equipment output from the pattern estimation unit 308A of the home controller 100A. Then, the information is output to the symbol information generation unit 506. The symbol information generation unit 506 generates symbol information based on the extracted waveform pattern, and outputs the generated symbol information to the pattern estimation unit 308A. The pattern estimation unit 308A determines whether the symbol information generated by the symbol information generation unit 506 satisfies the condition of the state transition diagram according to the symbol information registered in the power pattern database 510.

  If the pattern estimation unit 308A determines that the symbol information output from the symbol information generation unit 506 satisfies the condition transition diagram according to the symbol information registered in the power pattern database 510, the pattern estimation unit 308A includes the symbol information. Estimate the associated electrical device and output it. Thereby, even if it is an electric device which is not measured by the measuring instrument 400, that is, an unknown electric device, it is possible to estimate the electric device and manage information regarding power consumption as the estimated electric device.

  For home controller 100B, in accordance with the same method, for information related to power consumption of an unknown electrical device that is not measured by a measuring instrument, the electrical device is estimated and information regarding the estimated power consumption of the electrical device is managed. Is possible.

  In the system, since the server 500 includes an information database 502, a power pattern extraction unit 504, a symbol information generation unit 506, and a power pattern database 510, each home controller shares the server 500, so that each home It is possible to simplify the configuration of the controller and improve the efficiency of the entire system. Further, since the database can be shared in the server 500, the efficiency of the entire system can be improved.

  In addition, in this example, although two houses HA and HB were demonstrated, it is possible to apply similarly even if it is the structure in which several houses are provided.

  Moreover, in the above, although the system which estimates an electric equipment using a hidden Markov model was demonstrated, it is not restricted to this in particular, For example, it applies similarly also to the system which estimates an electric equipment using the dynamic time expansion / contraction method. It is possible.

  The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

  100, 100A, 100B Home controller, 101, 4021 CPU, 102 display unit, 103 display, 104 tablet, 105 operation buttons, 106 communication interface, 107 output interface, 108 input interface, 109 hard disk, 110 memory, 111 speaker, 112 clock , 150 repeaters, 160 smart meters, 200 electrical equipment.

Claims (4)

A power management device for managing n electrical devices provided in a house,
A storage unit that stores a power pattern database in which symbol information associated with a waveform pattern of power consumption of each electrical device is registered corresponding to each electrical device;
An acquisition unit for acquiring the power consumption of the entire home and the power consumption from m measuring devices provided corresponding to m (m <n) electrical devices,
A calculation unit that calculates the power consumption of an electrical device not provided with a measuring device based on the power consumption of the entire home acquired by the acquisition unit and the power consumption from the m measuring devices;
A power management apparatus comprising: an estimation unit that estimates an electrical device calculated based on a calculation result calculated by the calculation unit and a power pattern database stored in the storage unit.   When the symbol information corresponding to the waveform pattern of the power consumption acquired by the acquisition unit is not registered in the power pattern database, the symbol information associated with the waveform pattern is generated, and the generated symbol information The power management apparatus according to claim 1, further comprising: a pattern registration unit that registers the power in the power pattern database.   The estimation unit analyzes a waveform pattern of power consumption that is a calculation result calculated by the calculation unit, generates symbol information, and the generated symbol information is symbol information registered in the power pattern database. The power management apparatus according to claim 1, wherein the power management apparatus determines whether or not a condition of a hidden Markov model according to claim 1 is satisfied. A plurality of power management devices that respectively manage the power consumption of a plurality of electrical devices respectively provided in a plurality of homes;
A storage device that is provided in common to the plurality of power management devices and stores a power pattern database in which symbol information associated with the waveform pattern of the power consumption of each electrical device is registered corresponding to each electrical device; With
Each of the power management devices
The power consumption of the entire home and the power consumption from m measuring devices provided corresponding to m (m <n) of the n electrical devices in the home, respectively. An acquisition unit to acquire;
A calculation unit that calculates the power consumption of an electrical device not provided with a measuring device based on the power consumption of the entire home acquired by the acquisition unit and the power consumption from the m measuring devices;
A power management system comprising: an estimation unit that estimates an electrical device calculated based on a calculation result calculated by the calculation unit and a power pattern database stored in the storage device.

Images