JP4951082B2 - Energy saving advice generator - Google Patents

Description

  The present invention relates to an energy saving advice generating apparatus that provides energy saving advice that is highly acceptable to consumers, for example.

  In order to reduce the peak of power demand, technology (demand side management) for controlling consumer devices has been developed. For example, when the power demand is likely to exceed the allowable amount, the power company remotely controls the temperature setting of the air conditioner for the home so that it does not fall below 28 ° C. Alternatively, a technology for stopping the air conditioner in a room where no one is present has been developed. As a result, the peak of power demand is suppressed, electric power companies do not have to make excessive capital investment, and power generation using fossil fuel that must be operated during peak hours can be suppressed, reducing CO2 emissions. There is.

Japanese Patent Laid-Open No. 11-178247

  Changing the uniform temperature setting and stopping the air conditioner in an absent room are not necessarily energy-saving and satisfying measures for the home. For example, if multiple people use air conditioners in each room, the energy saving effect should be improved if they gather in one place. In such a case, it may be possible to rule in advance to generate advice to gather in one place, but in this case, the advice can be matched to the preference of each household. It is desirable that advice is automatically generated for each household.

  Japanese Patent Application Laid-Open No. 11-178247 discloses a method in which a consumer gives an operation pattern of an allowable device, and controls an excessively operated device based on the pattern. However, this method requires time and effort for a consumer to input a life pattern. In addition, it only controls the equipment and does not propose energy-saving methods that involve the movement of consumers.

  The present invention provides an energy saving advice generating apparatus capable of presenting energy saving advice that is easy for consumers to accept.

The energy-saving advice generating apparatus as one aspect of the present invention is
A first receiving unit for receiving occupancy information from a plurality of occupancy sensors for detecting a person's occupancy state in each of a plurality of rooms in the home;
Detecting means for detecting an operating state of each of the plurality of home appliances arranged in the household;
Obtaining means for obtaining power consumption by the plurality of home appliances;
A second receiving unit that receives a reduced power request indicating the amount of power consumption to be reduced from the power company system that performs power management of the household;
A database that stores records including the state of the room, the operating state of the home appliances, and the power consumption by the plurality of home appliances at regular intervals;
Detecting a first record that is a record corresponding to the time when the reduction request is received in the database, for each of a plurality of second records temporally prior to the first record,
(1) Similarity with the first record regarding the occupancy state of the room and the operating state of the home appliance,
(2) Occurrence frequency of the same state as the occupancy state of the room and the operating state of the home appliance shown in the second record,
(3) Frequency of transition from the same state as the occupancy state of the room and the operating state of the home appliance shown in the first record to the same state as the operating state of the room and the home appliance shown in the second record ,
Based on the database, a life preference level calculation unit that calculates a life preference level by weighted and summing the calculated similarity, occurrence frequency, and transition frequency values;
An energy saving degree calculation unit that detects a second record having the power consumption equal to or less than the first target power consumption with respect to a first target power consumption obtained by subtracting the value of the reduced power requirement from the power consumption of the first record. When,
The record having a high preference for life is selected from the second record detected by the energy saving degree calculation unit, and the occupancy state of the room and the operating state of the home appliance shown in the first record are selected records. An advice generation unit that generates advice to prompt the user to make an action to match,
A presentation unit for presenting the advice to the user;
Is provided.

  The present invention makes it possible to present energy saving advice that is easy for consumers to accept.

The structure of the energy consumption reduction apparatus containing the energy-saving advice production | generation apparatus as one Embodiment of this invention is shown. An example of the hardware constitutions of an energy saving advice production | generation apparatus is shown. An example of a life information DB is shown. An operation flow of the control unit 470 is shown. The flow of advice generation processing is shown. The flow of the life preference calculation process performed by the life preference calculation part is shown. The flow of a similarity calculation process is shown. A detailed flow of the seasonality / temporal similarity calculation processing is shown. The detailed flow of an environmental similarity calculation process is shown. The detailed flow of a number of people similarity calculation process is shown. The detailed flow of a household appliance utilization similarity calculation process is shown. The detailed flow of a similarity total process is shown. The detailed flow of the frequency calculation process by a frequency calculation part is shown. The detailed flow of the transition ease calculation process by a transition ease calculation part is shown. The detailed flow of the life preference total process by a life preference total part is shown. The detailed flow of the energy-saving degree calculation process by an energy-saving degree calculation part is shown. The detailed flow of the evaluation value calculation process by an evaluation value calculation part is shown. The detailed flow of the advice selection process by an advice selection part is shown. The detailed flow of the display process by a display process part is shown. An example of a result table for the life information DB of FIG. 3 is shown. An example of advice generation for the life information DB of FIG. 3 is shown. Another example of the life information DB is shown. An example of a result table for the life information DB of FIG. 22 is shown. An example of advice generation for the life information DB of FIG. 22 is shown.

(First embodiment)
FIG. 1 shows a configuration of an energy consumption reduction apparatus provided with an energy saving advice generation apparatus as one embodiment of the present invention.

  1 includes a plurality of in-room sensors 100, a plurality of home appliances 200, a plurality of environmental information sensors 300, an energy saving advice generating device 400, and a display device 500.

  The occupancy sensor 100 is arranged at the entrance of each room, and senses the occupancy state in the room. The occupancy state includes the number of people in the room or the presence or absence of a room. The occupancy sensor 100 arranged for each room senses the occupancy state in the room at regular intervals, and transmits the occupancy information indicating the occupancy state to the energy saving advice generation device 400. In this embodiment, the case where the number of people in a room is counted as an in-room state is shown.

  As the occupancy sensor 100, for example, a pyroelectric infrared sensor (pass sensor) including a set of a transmission unit and a reception unit can be used. Two pyroelectric infrared sensors (A and B) are arranged along the passageway at the entrance of the room. The transmitting unit and the receiving unit of each pyroelectric infrared A and B are arranged so as to face each other across the passage. When a consumer passes through the sensing area between the transmitter and the receiver, the corresponding pyroelectric infrared sensor detects the passage. When the pyroelectric infrared A is placed outside the room and the pyroelectric infrared B is placed inside the room, the consumer detects each pyroelectric infrared sensing area in the order of pyroelectric infrared A → pyroelectric infrared B When a person passes through each pyroelectric infrared sensing area in the order of pyroelectric infrared B → pyroelectric infrared A, the person leaving the room is detected. Use this to count the number of people in the room.

  As another example of the presence sensor 100, an IC tag sensor can be used. IC tags are attached to consumers and tag readers are installed in each room. The number of people in the room is counted according to the number of IC tags read by the reader.

  The environmental information sensor 300 is a sensor that measures environmental information of representative locations indoors and outdoors. The environmental information sensor 300 includes a temperature sensor, a humidity sensor, and an illuminance sensor. The temperature sensor measures the temperature (temperature) of the representative location. The humidity sensor measures the humidity at the representative location. The illuminance sensor measures the illuminance at the representative location. The representative portions of the temperature sensor, the representative sensor, and the humidity sensor may be the same or different. The environmental information sensor 300 measures environmental information (temperature, humidity, illuminance) at regular intervals, and transmits the measured values to the energy saving advice generation device.

  The household electrical appliance 200 is a household electrical appliance such as an air conditioner, a lighting, and a television set disposed in the home. Each household electrical appliance notifies the energy-saving advice generating apparatus 400 of operation information including its operating state and power consumption at regular time intervals.

  The operating state includes presence / absence of operation. The energy saving advice generation device 400 determines that the signal is not sent from the home appliance when it is not in operation. The power consumption during non-operation may be regarded as zero, or may be regarded as a predetermined value of power consumption in consideration of leakage power. Further, the operating state may be further classified. For example, the temperature setting ranges of the air conditioner may be defined as different states. In addition to the operating state and the non-operating state, a standby state may be defined. In the present embodiment, a case where the operation state is operation / non-operation will be described.

  The energy saving advice generation device 400 includes a life information DB 410, a life preference level calculation unit 420, an energy saving level calculation unit 430, an evaluation value calculation unit 440, an advice selection unit 450, a display processing unit (presentation unit) 460, and a control unit 470.

  The control unit 470 includes a first reception unit, a detection unit, and an acquisition unit includes a second reception unit and a third reception unit.

  The first receiving unit receives occupancy information (sensor data) from a plurality of occupancy sensors 100 that detect a person's occupancy state in each of a plurality of rooms in the home.

  The detection means detects the operating state of each of the plurality of home appliances 200 arranged in the home.

  The acquisition unit acquires power consumption (total power consumption) by the plurality of home appliances 200. The total power consumption may be calculated by totaling the power consumption of each device by the control unit 470, or may be obtained by receiving from a meter that measures the power usage of each device. In the present embodiment, the former case will be described.

  The second receiving unit receives a power consumption reduction request (reduction power request) from a power company system that performs household power management.

  The third receiving unit receives environmental information (sensor data) from an environmental information sensor that detects at least one of temperature, humidity, and illuminance.

  FIG. 4 is a flowchart showing an operation flow of the control unit 470.

  The control unit 470 performs the occupancy information for each room from the occupancy sensor 100 (in this example, the number of people in the room) from the occupancy sensor 100 and the operation information of the household appliance 200 from the household appliance 200 (in this example, the operation Presence / absence, power consumption) and environmental information (temperature, humidity, illuminance in this example) are collected from the environmental sensor 300 (S12).

  In step S12, control unit 470 calculates the total number of people for each room based on the collected information. Control unit 470 also calculates the total power consumption of each home appliance. The control unit 470 stores the information received from each sensor 100, 200, 300, the calculated total number of people, and the total power consumption in the living information database (living information DB) 410 together with the collection time (date and time). This stored information is called life information.

  In addition, the control unit 470 receives a reduction power request transmitted from the system of the power company that manages the power used in the home (S13). The power reduction request in this example indicates the power value to be reduced from the power used in the home (total power consumption). For example, if the power reduction request is “300”, this request indicates that power consumption in the home should be reduced by 300. The control unit 470 confirms the power saving request at regular intervals. When there is no request, the control unit 470 indicates that fact (0 in this example), and when the request is received, the value indicated in the request is displayed together with the above life information in the life information DB 410. Save to. In this example, the power reduction request indicates the amount of power to be reduced at the present time, but as another form, the target power consumption (second target power consumption) may be directly indicated. In this case, if the request is “500”, it means that the power consumption in the home is suppressed to 500 or less.

  FIG. 3 shows an example of the life information DB 410. Here, information is acquired every other minute. “Day, time” represents the time at which information was collected. “Room 1, Room 2, Living” represents the number of people in each room collected by the presence sensor. “Air temperature, humidity, and illuminance” represent the air temperature, humidity, and illuminance acquired by the environmental information sensor. “Air conditioner 1, air conditioner 2, air conditioner 3, TV, IH, washing machine” represents an operating state obtained from the home appliance. Here, “●” indicates that the device is being used. Assume that air conditioner 1 is located in room 1, air conditioner 2 is located in room 2, and air conditioner 3 is located in the living room. The power consumption represents the total power consumed at home in the living state. The value of the reduced power request is 0 until the record ID is 1999. However, when the record ID is 2000, the reduced power request having a value of 300 is received. The record (life information) when the reduced power request is received is referred to as a focus record (focus life information). The record of interest corresponds to the first record of the present invention. A record past the record of interest (first record) corresponds to the second record of the present invention.

  When the power reduction request is received (YES in S13), the control unit 470 generates advice by the life preference level calculation unit 420, the energy saving level calculation unit 430, the evaluation value calculation unit 440, the advice selection unit 450, and the display processing unit 460. The process is started (S14). When the reduced power request is not received (NO in S13), the process returns to step S11.

  FIG. 5 is a flowchart showing the flow of advice generation processing.

  The life preference level calculation unit 420 reads the life information from the life information DB 410, calculates the preference level and sends it to the evaluation value calculation unit 440 for each piece of life information that is temporally previous (previous) to the above life information of interest. (S21: Life preference calculation processing). The degree of preference indicates the degree to which the user likes the state indicated in the life information. The life preference level calculation unit 420 includes a similarity level calculation unit 421, a frequency calculation unit 422, a transition ease calculation unit 423, and a counting unit 424 for calculating the preference level. Details of the elements 421 to 424 will be described later.

  The energy saving degree calculation unit 430 reads out living information from the living information DB 410 and calculates an evaluation value by calculating an energy saving degree at least for each piece of living information (second record) temporally prior to the noted living information (focused record). (S22: Energy saving degree calculation process). In this example, the degree of energy saving is the value of the power consumption indicated in the life information (second record) relative to the value when the power consumption of the subject life information is reduced by the value indicated in the power reduction requirement. It is an index indicating whether it is less than that.

  The evaluation value calculation unit 440 calculates an evaluation value of living information from the received preference level and energy saving level and sends it to the advice selection unit 450 (S23: evaluation value calculation process).

  The advice selection unit 450 selects life information based on the received evaluation value (S24: advice selection processing). For example, life information is selected in descending order of evaluation. The advice selection unit 450 obtains a difference between the selected life information and the focused life information with respect to the occupancy state of the room and the operating state of the home appliance, and generates advice based on the obtained difference. Specifically, advice is generated so as to prompt the consumer to make the occupancy state of each room indicated by the life information of interest and the operating state of the home appliance match those indicated by the selected life information. The advice generation is performed, for example, until a predetermined number of different advices are generated. The advice selection unit 450 sends the generated advice to the display processing unit 460.

  The display processing unit 460 displays the received advice on the surface device 500 (S25: display processing). This presents advice to consumers. The advice may be presented to the user by announcing the content of the advice together with the display on the display unit 500 or instead of the display. The display device 500 may be an arbitrary display device such as a liquid crystal display device, a CRT display device, an organic EL display device, a plasma display device, or the like.

  FIG. 2 shows an example of a hardware configuration of the energy saving advice generating apparatus 400. The energy saving advice generating apparatus 400 can be configured by a personal computer (PC) 600 as shown in FIG.

  The PC 600 has a configuration in which a CPU 630, a memory 640, a hard disk 650, and interfaces (IF) 610a and 610b are connected via a bus 620.

  The PC 600 obtains information from the plurality of occupancy sensors 100, the plurality of home appliances 200, and the environment information sensor 300, generates energy saving advice, and sends it to the display device 500. The life information DB 410 in the energy saving advice generating apparatus 400 is constructed on the hard disk 650 in the PC 600. A result table described later is also constructed on the hard disk 650 (or memory 640).

  In the main process executed by the control unit 470 (see FIG. 4), as described above, data collection is performed periodically to obtain the values of occupancy sensors, home appliances, and environmental information sensors, and the total number of people and power consumption are calculated. These are stored in the life information DB 410 (see FIG. 3) together with the time information. The time information may be acquired from the system timer.

  The similarity calculation unit 421, frequency calculation unit 422, transition ease calculation unit 423, life preference level calculation unit 424, energy saving level calculation unit 430, evaluation value calculation unit 440, advice selection unit 450 in the energy saving advice generation device 400 are: It is implemented as a PC600 computer program. The program is stored in advance in the hard disk 650, read into the memory 640, and then executed by the CPU 630 according to the procedure.

  In the main process (see FIG. 4) performed by the control unit 470, as described above, for example, information is collected from the occupancy sensor 100, the environmental information sensor 300, and the home appliance 200 every minute, and the total number of people and power consumption are collected. Calculation is performed, and the collected information and the calculated value are stored in the living information DB 410 together with the time information (S12). When a power reduction request is received, advice generation processing (S14) is executed.

  In the advice generation process (S14), as shown in FIG. 5, life preference level calculation process (S21), energy saving level calculation process (S22), evaluation value calculation process (S23), advice selection process (S24), display Processing is performed in the order of processing (S25).

  FIG. 6 is a flowchart showing the flow of the life preference level calculation process (S21) performed by the life preference level calculation unit 420.

  In the life preference calculation processing, the similarity calculation processing by the similarity calculation unit 421 (S31), the frequency calculation processing by the frequency calculation unit 422 (S32), the transition ease calculation processing by the transition ease calculation unit 423 (S33), the life Life preference level totaling processing (S34) by the preference level totaling unit 424 is performed.

  The similarity calculating unit 421 calculates the similarity between the target life information (first record) and each of the past life information (second record) temporally before the target life information. A detailed processing flow of the similarity calculation processing is shown in FIG.

  As shown in FIG. 7, in the similarity calculation processing, seasonality, temporal similarity calculation processing (S41), environmental similarity calculation processing (S42), number of people similarity calculation processing (S43), home appliance use similarity calculation processing (S44) A similarity count process (S45) is performed.

  FIG. 8 shows a detailed flow of the seasonality / temporal similarity calculation process (S41).

  The following processing is performed for each i while sequentially incrementing i from 1 to N−1 (S51, S54, S55). The record with index N indicates the life information of interest, and i = 1 to N−1 indicates the past life information.

  Difference DiffDay (Day [N], Day [i]) between Day [i] (where i = {1, ..., N-1}) and Day [N] of the life information DB (number of records is N) Is stored in the seasonal similarity degree DayS [i] (S52). Day is the “day” of the life information DB.

  Also, the difference DiffTime (Time [N], Time [i]) between Time [i] (where i = {1,..., N-1}) and Time [N] is calculated, and the time similarity similarity TimeS [i [S52]. Time is the “time” of the life information DB.

  Further, DayS [i] and TimeS [i] (where i = {1,..., N-1}) are weighted and summed to calculate the seasonal temporal similarity degree DayTimeS [i] (S53). Write DayTimeS [i] to the result table. An example of the result table is shown in FIG.

here
DiffDay (x, y) = 182.5- | (| xy | mod 365)-182.5 |
DiffTime (x, y) = 7200- | (| xy | mod 14400)-7200 | (where x and y are minutes)
DayTimeS [i] = C11 ・ DayS [i] + C12 ・ TimeS [i] (C11 and C12 are weighting factors)
And

DiffDay (x, y) is smaller as the day is closer and larger as it is farther away. The value is highest when half a year away. DiffTime (x, y) is also smaller as the time is closer and larger as it is farther away.

  FIG. 9 shows a detailed flow of the environment similarity calculation process (S42).

  The following processing is performed for each i while sequentially incrementing i from 1 to N-1 (S61, S64, S65).

  Difference DiffTemp (Temp [N], Temp [i]) between Temp [i] (where i = {1, ..., N-1}) and Temp [N] in the life information DB (with N records) Is stored in the temperature similarity TempS [i] (S62).

  Also, the difference DiffHumi (Humi [N], Humi [i]) between Humi [i] (where i = {1, ..., N-1}) and Humi [N] is calculated and the humidity similarity HumiS [i] (S62).

  Also, the difference DiffIllu (Illu [N], Illu [i]) between Illu [i] (where i = {1, ..., N-1}) and Illu [N] is calculated, and the illuminance similarity IlluS [i] (S62).

  Further, TempS [i], HumiS [i], and IlluS [i] (where i = {1,..., N-1}) are weighted and summed to calculate the environmental similarity EnvS [i] (S63). Write EnvS [i] to the result table (see FIG. 20).

here
DiffTemp (x, y) = max (1- | xy | / 3, 0)
DiffHumi (x, y) = max (1- | xy | / 20, 0)
DiffIllu (x, y) = max (1-| xy | / 50, 0)
EnvS [i] = C21 ・ TempS [i] + C22 ・ Humi [i] + C23 ・ Illu [i]
And

For example, in the case of DiffTemp (x, y), the value is 0 if the temperature differs by 3 degrees or more. If it is less than 3 degrees, the closer the temperature is, the larger the value becomes, and 1 when the values match. DiffHumi (x, y)
, DiffIllu (x, y) is interpreted in the same way.

  FIG. 10 shows a detailed flow of the number-of-persons similarity calculation process (S43).

  The following processing is performed for each i while sequentially incrementing i from 1 to N-1 (S71, S79, S80).

  Room [r, i] (where i = {1,…, N-1}, r = {1,…, NR}) and Room [r, N] The difference DiffRoom (Room [N], Room [i]) is calculated and stored in the room number similarity RoomS [r, i].

  Here, NR represents the number of rooms. Room [r, i] represents the number of people in room r of record i. Room [r, N] represents the number of people in room r of record N. The definitions of DiffRoom (x, y) and DiffNumP (x, y) are as follows.

DiffRoom (x, y) = 1 (x = y), 0 (x! = Y)
DiffNumP (x, y) = 1 (x = y), 0 (x! = Y)
RoomS [r, i] multiplied by coefficient C3 [r] (room weighting coefficient) and added for r = {1,..., NR} is defined as RoomSAll [i] (S74, S75, S76) ).

  The difference DiffNumP (NumP [i], NumP [N]) between NumP [i] (where i = {1, ..., N-1}) and NumP [N] is calculated, and the number of people similarity to NumPS [i] Is stored (S77). NumP [i] is the total number of people in the room of record i. NumP [N] is the total number of people in record N's room.

  Furthermore, NumPS [i] is weighted, added to RoomSAll [i] (where i = {1,..., N-1}), and stored in NumPRoomS [i].

Ie
NumPRoomS [i] = RoomSAll [i] + C3N ・ NumPS [i]
It is.

Write NumPRoomS [i] to the result table (see Figure 20)
FIG. 11 shows a detailed flow of home appliance use similarity calculation processing (S44).

  The following processing is performed for each i while sequentially incrementing i from 1 to N-1 (S91, S97, S98).

  HA [p, i] (where i = {1,…, N-1}, p = {1,…, NP}) and HA [p, N] The difference DiffHA (HA [p, N], HA [p, i]) is calculated and stored in the home appliance utilization similarity HAS [p, i] (S93). Here, NP represents the number of home appliances. Also, DiffHA (x, y) = 1 (x = y), 0 (x! = Y).

  Multiply the calculated HAS [r, i] by the coefficient C4 [p] (home appliance weighting factor) and add r = {1, ..., NP} to HASAll [i] (S94, S95, S96) ). Write HASAll [i] to the result table (see Figure 20).

  FIG. 12 shows a detailed flow of the similarity count processing (S45).

  The following processing is performed for each i while sequentially incrementing i from 1 to N-1 (S101, S103, S104).

  DayTimeS [i] (seasonality, temporal similarity), EnvS [i] (environmental similarity), NumPRoomS [i] (number of people similarity), HASAll [i] (home appliance similarity) (however i = {1,..., N-1}) is added and substituted for similarity Similar [i]. Write Similar [i] to the result table (see Figure 20).

  As described above, the similarity between the target life information (N) and each of the past life information (1 to N-1) that is temporally prior to the target life information is calculated.

  The frequency calculation unit 422 adds the frequency of occurrence of individual life information to the life information history.

  FIG. 13 shows a detailed flow of frequency calculation processing (S32 in FIG. 6) by the frequency calculation unit 422.

  The following processing is performed for each i while sequentially incrementing i from 1 to N-1 (S111, S117, S118).

For record i, the number of records that have the same Room [r, i] (where r = {1,…, NR}) and HA [p, i] (where p = {1,…, NP}) as record i Are substituted into Freq [i] (S112 to S116). That is, the number of records in which the number of people in each room and the operating state of each household electrical appliance are exactly the same as the record i are counted. Write Freq [i] to the result table. When counting the number of records, record N may or may not be included in the count target (in this example, see S116).
.

  Instead of the number of records, the ratio of the same number of records to the total number of records N-1 may be substituted into Freq [i] as a frequency. That is, the frequency of the present invention may be the number of matching cases or the ratio of the number of matching cases to the total number. The result table of FIG. 20 shows the case of ratio.

  The transition ease calculation unit 423 calculates a transition frequency representing the ease of transition from the current life information (focused life information) to the past life information (transition ease) in the life information history. In the following, a transition probability is shown as the transition frequency, but the number of transitions may be used.

  FIG. 14 shows a detailed flow of the transition ease calculation process (S33) by the transition ease calculator 423.

  In step S121, Room [r, i] matches for all r = {1, ..., NR} and HA [p, i] matches for all p = {1, ..., NP} (the pattern is Create a continuous set of i starting from the first record with i = 1 in the life information DB. Room [r, i] matches for all r = {1,…, NR} and HA [p, i] matches for all p = {1,…, NP} The number of people is the same, and the operating status of each home appliance is the same. At this time, a set whose continuous number is equal to or smaller than THcont is deleted (a state where the number of continuous records is small is not regarded as a pattern).

  A pattern identifier related to the number of people in the room and the operating state of the home appliance is set for the created set. For example, when M sets are created, pattern identifiers (names) P1, P2,. However, the same pattern identifier is assigned to a set having the same state (the number of people in each room and the operating state of home appliances).

  In steps S122 and S123, assuming that the number of persons in each room of record N and the operation pattern of the home appliance are Pdsm, the transition probability from Pdsm to each pattern P1, P2,.

More specifically, first, a transition table describing the number of transitions between pattern identifiers as described below is created by tracing the transition of pattern identifiers in the order in which the sets are generated. The vertical axis is the transition source, and the horizontal axis is the transition destination. For example, if the pattern identifier transition is P3 → P2 → P8 → P4 → P1 → P2 → P4 → P9 → P2 → P1 → P2 ... If two `` P1 → P2 '' are included, The number of transitions from pattern P1 to P2 is two. Transition from Pdsm to P1 to PM (excluding the same pattern as Pdsm) by finding the same pattern as Pdsm from P1 to PM and calculating the probability of transition from the found pattern to another pattern Probability is obtained. The transition probability can be calculated by the number of transitions of interest with respect to the total number of transitions (the sum of all cell values in the table below). For example, if Pdsm matches P2, the transition probability from Pdsm to P1 is “4 / total number”.

  In step S124, for the record i (where i = {1,..., N-1}) in the life information DB, the transition probability corresponding to the pattern of the record is substituted into Move [i]. Write Move [i] to the result table. Further, the record i of the living information DB (where i = {1,..., N-1}) is substituted into the pattern identifier P [i] of the record. Write P [i] to the result table. Note that transition probabilities need not be assigned to records having the same pattern as record N. When using the transition frequency instead of the transition probability as the transition frequency, the value of the corresponding cell in the above table may be substituted.

  In this example, a pattern is created that matches Room [r, i] for all r = {1, ..., NR} and HA [p, i] for all p = {1, ..., NP}. However, as another example, a pattern with matching Room [r, i] may be created for all r = {1,..., NR}, and the transition probability may be calculated in the same manner as described above.

  FIG. 15 shows a detailed flow of the life preference level totaling process (S34) by the life preference level totaling unit 424.

  The following processing is performed for each i while sequentially incrementing i from 1 to N-1 (S131, S133, S134).

  That is, C51 · Similar [i] (similarity), C52 · Freq [i] (frequency) and C53 · Move [i] (ease of transition) are added and substituted into the life preference similarity similarity [i]. . Here, C51, C52, and C53 are coefficients for calculating life preference. Write Similar [i] to the result table (see Figure 20).

  FIG. 16 shows a detailed flow of the energy saving degree calculation process (S22 in FIG. 5) by the energy saving degree calculation unit 430.

  The following processing is performed for each i while sequentially incrementing i from 1 to N (S141, S143, S144).

  The value “W [N] −Wdsm” (first target power consumption) obtained by subtracting the value of the power consumption requirement Wdsm from the power consumption W [N] of the life information of interest (record N) and the power consumption W [i of each record ] Is used to calculate DiffW (W [N] −Wdsm, W [i]), and is assigned to the energy saving degree Echo [i]. Echo [i] is written to the result table (see FIG. 20).

here,
DiffW (a, b) = 0 (a ≧ b), -∞ (a <b)
And

  In other words, the power consumption of each record is compared with the power consumption of each record by comparing the power consumption of each record with the value (first target power consumption) after subtracting the value of the reduction power requirement from the power consumption of the focus life information. Set to 0 if the power is less than or equal to -∞ (minus infinity) otherwise. That is, a record in which 0 is set is a record that satisfies the reduced power requirement.

  In addition, when the reduced power requirement indicates the second target power value in another embodiment of the present invention, the second target power consumption is set instead of “W [N] −Wdsm” (first target power consumption). Use it.

  FIG. 17 shows a detailed flow of an evaluation value calculation process (S23 in FIG. 5) by the evaluation value calculation unit 440.

  The following processing is performed for each i while sequentially incrementing i from 1 to N-1 (S151, S153, S154).

  The evaluation value Eval [i] is calculated by adding C61 · Favor [i] and C62 · Echo [i] for i = {1,..., N−1}. The evaluation value Eval [i] is written into the result table (see FIG. 20).

  FIG. 18 shows a detailed flow of advice selection processing (S24 in FIG. 5) by the advice selection unit 450. In this flow, for i = {1, ..., N-1}, life information (records) is sorted in descending order of evaluation value Eval [i], and advice is generated for life information in descending order of Eval [i]. To go.

  More specifically, the life information DB is first sorted in descending order of Eval [i] (S161).

  Initial settings a = 0, i = 1, and ADV = φ are performed (S162). a is a variable for storing the number of generated advices, i is a variable for storing the sorted record IDs, and ADV is a list for storing the generated advices.

  In the life information DB after sorting, the following processing is performed for each i while i is sequentially incremented. That is, the following processing is performed in descending order of evaluation value.

  Advice is generated based on the difference between the record i and the record of interest N (difference between the number of people in each room and the operating state of the home appliance) (S163).

  For example, if the air conditioner 1 of the record of interest is ON and the air conditioner 1 of the record i is OFF, the advice “Please turn off the air conditioner 1” is generated. If the number of people in room 1 and room 2 in the record of interest is 1, respectively, the number of people in room 1 in record i is 0 and the number of people in room 2 is 2, for example, “Please move from room 1 to room 2.” Generate advice. It should be noted that regarding the state common between the record of interest and the record i, an advice that the state may be kept may be added. For example, if the air conditioner 1 is ON for both the record i and the record of interest, advice such as “the air conditioner 1 may remain ON” may be added.

  In step S164, when the same advice as the generated advice has already been generated (YES), the process proceeds to the next record (S167). On the other hand, when it is generated for the first time, the generated advice adv is added to the list ADV (S165).

  a is incremented (S166), and the process proceeds to the next record (S167).

  After proceeding to the next record, it is checked whether or not the generated advice number a has reached the number THadv specified in advance (S168), and if not, the process returns to step S163. When it reaches or reaches the last record N-1, this processing is terminated.

  Note that the record for which no pattern identifier is set in step S163 is in a state that is not familiar to the user, so that no advice may be generated. Note that the generation of advice may be omitted for records having the same pattern as the record of interest.

  FIG. 19 shows a detailed flow of the display process (S25 in FIG. 5) by the display processing unit 460.

In this flow, the generated advice is sent to the display device 500 and displayed.
In more detail, first, the advice adv stored in the ADV is extracted from the head (S172), and the advice adv is transmitted to the display device 500. The display device 500 presents the sent advice to the user. The reason for taking out from the top is that the advice generated earlier is displayed earlier. Further, the advice generated earlier (the advice generated from the record having a higher evaluation value) is preferentially displayed on the screen. For example, each advice may be displayed with a priority rank (advice rank) (see FIG. 20). Alternatively, priority advice may be displayed on the screen. The advice rank is set higher as it is stored at the head of the ADV. The advice rank is set by the advice selection unit 450 and stored in the result table (see FIG. 20).

  Until all the advices are extracted from the ADV, the advices are sequentially extracted from the head side and transmitted to the display device (S171, S174, S175). The display device displays the advice sent sequentially, for example, from top to bottom. Note that | ADV | in the block of S175 shown in the figure indicates the number of advices. s is a parameter that controls the number of repetitions.

  FIG. 21 shows an example of advice generation for the life information DB of FIG. The reduction power requirement is 300 W, and the series of calculations described so far are performed, and finally two advices are generated.

(1) Let's move from room 1 to the living room. You can turn on the air conditioner in your living room. Turn off the air conditioners in rooms 1 and 2. Let's stop the washing machine.
(2) Let's move from room 1 to the living room. You can turn on the air conditioner in your living room. You can watch TV. Turn off the air conditioners in rooms 1 and 2. Let's stop the washing machine.

The advice of (1) (advice rank 1) is obtained by comparing, for example, the record N (= 2000) of the life information DB in FIG. 3 and any one of the records 1013 to 1022.
The advice of (2) (advice rank 2) is obtained by comparing the record N (= 2000) of the life information DB in FIG. 3 with any of the records 1023 to 1027.

  FIG. 22 shows an example of the life information DB when the reduction power request is 200 W.

  FIG. 23 shows an example of a result table for the life information DB of FIG.

  FIG. 24 shows an example of advice generation for the life information DB of FIG.

  At this time, the same calculation as above is performed, and finally three advices are generated.

(1) Move from the living room to room 2. Let's stop the washing machine
(2) Let's move from room 1 to the living room. You can turn on the air conditioner in your living room. Turn off the air conditioners in rooms 1 and 2. Let's stop the washing machine.
(3) Let's move from room 1 to the living room. You can turn on the air conditioner in your living room. You can watch TV. Turn off the air conditioners in rooms 1 and 2. Let's stop the washing machine.

As described above, in the present embodiment, from the history of occupancy information, environmental information, home appliance operation information, power consumption collected in the home, the reduced power requirement is satisfied, and similar to the current situation, In addition, energy saving advice that prompts the user to make an action so as to make a transition to a state where the transition is easy is presented to the user. This makes it possible to create energy-saving advice closer to the current living state and present it to the user according to the amount of power demand that is occasionally reduced.
In addition to simple temperature setting operations and air conditioning in absent rooms, it is possible to provide advice on energy-saving actions that involve movement of people.The priority of advice is energy-saving and close to current actions. Because it is frequently performed in the household, it has the effect of being easily accepted by consumers.

Claims (9)

A first receiving unit for receiving occupancy information from a plurality of occupancy sensors for detecting a person's occupancy state in each of a plurality of rooms in the home;
Detecting means for detecting an operating state of each of the plurality of home appliances arranged in the household;
Obtaining means for obtaining power consumption by the plurality of home appliances;
A second receiving unit that receives a reduced power request indicating the amount of power consumption to be reduced from the power company system that performs power management of the household;
A database that stores records including the state of the room, the operating state of the home appliances, and the power consumption by the plurality of home appliances at regular intervals;
Detecting a first record that is a record corresponding to the time when the reduction request is received in the database, for each of a plurality of second records temporally prior to the first record,
(1) Similarity with the first record regarding the occupancy state of the room and the operating state of the home appliance,
(2) Occurrence frequency of the same state as the occupancy state of the room and the operating state of the home appliance shown in the second record,
(3) Frequency of transition from the same state as the occupancy state of the room and the operating state of the home appliance shown in the first record to the same state as the operating state of the room and the home appliance shown in the second record ,
Based on the database, a life preference level calculation unit for calculating a life preference level by weighted sum of the calculated similarity, occurrence frequency, and transition frequency,
An energy saving degree calculation unit for detecting a second record having the power consumption of a value equal to or less than a target power consumption obtained by subtracting the value of the power reduction requirement from the power consumption of the first record;
The record having a high preference for life is selected from the second record detected by the energy saving degree calculation unit, and the occupancy state of the room and the operating state of the home appliance shown in the first record are selected records. An advice generation unit that generates advice to prompt the user to make an action to match,
A presentation unit for presenting the advice to the user;
Energy-saving advice generator with The advice generation unit repeatedly generates the advice by selecting the second record in descending order of the life preference until a predetermined number of different advices are generated,
2. The energy saving advice generation apparatus according to claim 1, wherein the presenting unit performs advice display so that advice generated from a record having a high life preference level is preferentially displayed. A third receiver that receives environmental information from an environmental information sensor that detects at least one of temperature, humidity, and illuminance;
The database includes the environmental information in the record,
3. The lifestyle preference degree calculation unit calculates the degree of residing in the room, the operating state of the home appliance, and the degree of similarity related to the environment information as the degree of similarity in (1). Energy saving advice generator described in 1. The database includes the time information for each predetermined time in the record,
3. The life preference level calculation unit calculates a similarity level regarding the presence status of the room, the operating status of the home electrical appliance, and the time information as the similarity level of (1). Energy saving advice generator described in 1.   5. The occupancy sensor detects the presence / absence of an occupancy in the room or the number of occupants in the room as the occupancy state of the room. Energy saving advice generator.   6. The energy saving advice generation apparatus according to claim 1, wherein the detection unit detects whether or not the operation is performed as the operation state. The reduced power demand indicates the target power consumption of the household,
The energy saving degree calculation unit compares the target power consumption indicated in the reduction power requirement with the power consumption of the second record, and detects a second record having the power consumption that is equal to or less than the target power consumption. The energy-saving advice generation device according to any one of claims 1 to 6, wherein A first receiving step of receiving occupancy information from a plurality of occupancy sensors for detecting a person's occupancy state in each of a plurality of rooms in the home;
A detection step of detecting an operating state of each of the plurality of home appliances arranged in the home;
An acquisition step of acquiring power consumption by the plurality of home appliances;
A second receiving step of receiving a reduced power request indicating an amount of power consumption to be reduced from the power company system that performs power management of the household;
A record corresponding to the time when the reduction request is received in a database that stores records including the occupancy state of the room, the operating state of the home appliances, and the power consumption of the plurality of home appliances at regular intervals. For each of a plurality of second records temporally prior to the first record,
(1) Similarity with the first record regarding the occupancy state of the room and the operating state of the home appliance,
(2) Occurrence frequency of the same state as the occupancy state of the room and the operating state of the home appliance shown in the second record,
(3) Frequency of transition from the same state as the occupancy state of the room and the operating state of the home appliance shown in the first record to the same state as the operating state of the room and the home appliance shown in the second record ,
A life preference calculating step for calculating a life preference by calculating a weighted sum of the calculated similarity, occurrence frequency, and transition frequency based on the database;
An energy saving degree calculating step of detecting a second record having the power consumption equal to or lower than a target power consumption obtained by subtracting the value of the reduced power requirement from the power consumption of the first record;
The record having a high life preference level is selected from the second record detected in the energy saving level calculating step, and the occupancy status of the room and the operating status of the home appliance shown in the first record are selected records. An advice generation step for generating advice to prompt the user to make an action to match
A presenting step of presenting the advice to the user;
A computer-generated energy saving advice generation method. The reduced power demand indicates the target power consumption of the household,
The energy saving degree calculating step compares the target power consumption indicated in the reduction power requirement with the power consumption of the second record, and detects a second record having the power consumption that is equal to or less than the target power consumption. The energy saving advice generation method according to claim 8.

Images