JP6786770B2 - A device equipped with a communication function, a system equipped with the device, and a unit search method. - Google Patents

Description

The present invention relates to a device having a communication function and communicating with another device, a system having the device and forming a network by communication, and a method of searching for the number of devices constituting the system. ..

Consensus cooperative control has been developed, which is a method in which the internal values of a plurality of devices are matched by transmitting and receiving internal values by communication between the devices, rather than being matched by the management device that manages all of these devices. .. In the consensus cooperative control, each device sends and receives an internal value to at least one other device, and generates an internal value using a calculation result based on the generated internal value and the received internal value. By performing this process in each device, the internal value of each device converges to the same value. As an example of using this method, Patent Document 1 describes that the internal phase of the inverter device is synchronized, and Patent Document 2 describes the amount of suppression of the output active power by matching the compensation values of the inverter devices. It is stated to adjust. Further, Patent Document 3 describes that each measuring device calculates an overall average value by matching an internal average value based on the measured value, and Patent Document 4 describes that each measuring device is based on the measured value. It is described that the total maximum value (minimum value) is calculated by matching the internal maximum value (internal minimum value). In consensus coordinated control, it is not necessary to provide a management device that manages the entire device.

Japanese Unexamined Patent Publication No. 2015-027155 JP-A-2015-084612 JP-A-2015-166901 JP-A-2015-215204

Reza Olfati-Saber, J. Alex Fax, and Richard M. Murray, "Consensus and Cooperation in Networked Multi-Agent Systems", Proceedings of the IEEE, Vol.95, No.1, (2007) Mehran Mesbahi and Magnus Egerstedt, "Graph Theoretic Methods in Multiagent Networks", Princeton (2010)

In consensus coordinated control, it is difficult to manage the number of devices constituting the control network because the management device for managing the entire device is not provided. However, it is necessary to know the number of devices that make up the network for maintenance. In some cases, a coefficient according to the number of units is used for control.

The present invention was conceived under the above-mentioned circumstances, and when a management device for managing the entire devices constituting the network is not provided, the number of devices constituting the network can be determined. Its purpose is to provide a method of searching, a device and a system that implements the method.

In order to solve the above problems, the following technical measures are taken in the present invention.

The system provided by the first aspect of the present invention is a system in which a plurality of devices constitute a network by communication, and the device includes an internal value generating means for generating an internal value and the internal value generating. The internal value is provided as a communication means for transmitting an internal value generated by the means as a generated internal value to at least one of other devices and receiving an internal value transmitted by at least one of the other devices as a receiving internal value. The value generating means is a calculation means that subtracts the generated internal value from the received internal value and adds all the subtraction results, and an integrating means that integrates the calculation result output by the calculation means to calculate the internal value. After setting the first value or the second value as the initial value of the generated internal value, it converged by repeating the calculation by the calculation means, the integration by the integration means, and the transmission and reception of the internal value by the communication means. It is provided with a number search means for calculating the number of devices constituting the network based on the generated internal value, and the generated internal value in any one of the devices constituting the network. The first value is set as the initial value of the above, and the second value is set as the initial value of the generated internal value in other devices. According to this configuration, the internal value generating means generates an internal value by an operation based on the generated generated internal value and the received internal value received by the communication means. When the internal value generating means of each device constituting the network does this, the internal values of all the devices converge to the arithmetic mean value. In addition, the first value is set as the initial value of the generated internal value in any one of the devices constituting the network, and the second value is set as the initial value of the generated internal value in the other devices. Will be done. Therefore, the converged generated internal value is a value calculated from the number of devices constituting the network, the first value, and the second value. The number search means calculates the number of devices constituting the network from the converged generated internal value. As a result, the number of devices constituting the network can be searched even when the management device for managing the entire devices constituting the network is not provided.

In a preferred embodiment of the present invention, the first value is "1", the second value is "0", and the number search means is based on the reciprocal of the converged generated internal value. Further, it is provided with a number calculation means for calculating the number. According to this configuration, it is possible to facilitate the calculation by the number calculation means.

In a preferred embodiment of the present invention, the number search means further includes a convergence detecting means for detecting that the generated internal value has converged based on a change in the generated internal value. According to this configuration, it is possible to appropriately detect that the generated internal values have converged.

In a preferred embodiment of the present invention, the apparatus subtracts the generated internal value from the received internal value, and adds and outputs only those whose subtraction result is a positive value. Then, a random number is set as the initial value of the generated internal value, and after the calculation by the second calculation means, the integration by the integration means, and the transmission / reception of the internal value by the communication means are performed, the generated internal value is set. Based on this, the set value determining means for determining whether to set the first value or the second value is further provided, and the set value determined after the determination by the set value determining means is set. It is set as an initial value of the generated internal value, and processing is performed by the number search means. According to this configuration, a random number is set as the initial value of the generated internal value, and after the process of converging the generated internal value to the maximum value is performed, the first value is set or the second value is set based on the generated internal value. It is decided whether to set the value of. Therefore, it is possible to automatically determine whether to set the first value or the second value without human intervention.

In a preferred embodiment of the present invention, the apparatus subtracts the generated internal value from the received internal value, and adds and outputs only those whose subtraction result is a negative value. Then, a random number is set as the initial value of the generated internal value, and after the calculation by the second calculation means, the integration by the integration means, and the transmission / reception of the internal value by the communication means are performed, the generated internal value is set. Based on this, the set value determining means for determining whether to set the first value or the second value is further provided, and the set value determined after the determination by the set value determining means is set. It is set as an initial value of the generated internal value, and processing is performed by the number search means. According to this configuration, a random number is set as the initial value of the generated internal value, and after the process of converging the generated internal value to the minimum value is performed, the first value is set or the second value is set based on the generated internal value. It is decided whether to set the value of. Therefore, it is possible to automatically determine whether to set the first value or the second value without human intervention.

In a preferred embodiment of the present invention, the set value determining means determines to set the first value when the generated internal value does not change, and when the generated internal value changes. Is determined to set the second value. According to this configuration, it is possible to appropriately determine whether to set the first value or the second value.

The device provided by the second aspect of the present invention transmits an internal value generating means for generating an internal value and an internal value generated by the internal value generating means as a generated internal value to at least one of other devices. A communication means for receiving an internal value transmitted by at least one of the other devices as a reception internal value is provided, and the internal value generating means subtracts the generated internal value from the received internal value and subtracts the result. A calculation means for adding all of the above, an integration means for integrating the calculation result output by the calculation means, and calculating the internal value, and a first value or a second value are set as the initial value of the generated internal value. After that, the number of devices constituting the network by communication is calculated based on the generated internal value converged by the calculation by the calculation means, the integration by the integration means, and the repetition of transmission and reception of the internal value by the communication means. It is characterized by having a means for searching the number of units.
According to this configuration, the number of devices constituting the network can be searched even when the management device for managing the entire devices constituting the network is not provided.

In a preferred embodiment of the present invention, the apparatus subtracts the generated internal value from the received internal value, and adds and outputs only those whose subtraction result is a positive value. Then, a random number is set as the initial value of the generated internal value, and after the calculation by the second calculation means, the integration by the integration means, and the transmission / reception of the internal value by the communication means are performed, the generated internal value is set. Based on this, the set value determining means for determining whether to set the first value or the second value is further provided, and the set value determined after the determination by the set value determining means is set. It is set as an initial value of the generated internal value, and processing is performed by the number search means. According to this configuration, it is possible to automatically determine whether to set the first value or the second value without human intervention.

In a preferred embodiment of the present invention, the apparatus subtracts the generated internal value from the received internal value, and adds and outputs only those whose subtraction result is a negative value. Then, a random number is set as the initial value of the generated internal value, and after the calculation by the second calculation means, the integration by the integration means, and the transmission / reception of the internal value by the communication means are performed, the generated internal value is set. Based on this, the set value determining means for determining whether to set the first value or the second value is further provided, and the set value determined after the determination by the set value determining means is set. It is set as an initial value of the generated internal value, and processing is performed by the number search means. According to this configuration, it is possible to automatically determine whether to set the first value or the second value without human intervention.

In a preferred embodiment of the present invention, the set value determining means determines to set the first value when the generated internal value does not change, and when the generated internal value changes. Is determined to set the second value. According to this configuration, it is possible to appropriately determine whether to set the first value or the second value.

The method provided by the third aspect of the present invention is in a communication network composed of an internal value generating means for generating an internal value and a communication means for communicating with at least one other device. It is a method of searching for the number of devices constituting the network, and uses the internal value generation step in which the internal value generation means generates the internal value and the internal value generated in the internal value generation step as the generation internal value. A transmission step of transmitting to at least one of the other devices and a receiving step of receiving an internal value transmitted by at least one of the other devices as a reception internal value are performed by each device. The value generation step is an integration step of subtracting the generated internal value from the received internal value and adding all the subtraction results, and an integration of integrating the calculation results output in the calculation step to calculate the internal value. A process is provided, a first value is set as an initial value of the generated internal value in any one of the devices constituting the network, and an initial value of the generated internal value is set in the other devices. After the setting step of setting the second value as the value and the setting step, the internal value generation step, the transmission step, and the reception step are repeated, and the network is configured based on the converged generated internal value. It is characterized by including a number calculation process for calculating the number of devices. According to this configuration, the number of devices constituting the network can be searched even when the management device for managing the entire devices constituting the network is not provided.

In a preferred embodiment of the present invention, the setting step is a second operation in which the generated internal value is subtracted from the received internal value, and only those whose subtraction result is a positive value are added and output. After the steps, the random number setting step of setting a random number as the initial value of the generated internal value, and the random number setting step, after the second calculation step, the integration step, the transmission step, and the reception step are performed. The generation includes a determination step of determining whether to set the first value or the second value based on the generated internal value, and the set value determined by the determination step is generated. Set as the initial value of the internal value. According to this configuration, it is possible to automatically determine whether to set the first value or the second value without human intervention.

In a preferred embodiment of the present invention, the setting step is a second operation in which the generated internal value is subtracted from the received internal value, and only those whose subtraction result is a negative value are added and output. After the steps, the random number setting step of setting a random number as the initial value of the generated internal value, and the random number setting step, after the second calculation step, the integration step, the transmission step, and the reception step are performed. The generation includes a determination step of determining whether to set the first value or the second value based on the generated internal value, and the set value determined by the determination step is generated. Set as the initial value of the internal value. According to this configuration, it is possible to automatically determine whether to set the first value or the second value without human intervention.

According to the present invention, the internal value generating means generates an internal value by an operation based on the generated generated internal value and the received internal value received by the communication means. When the internal value generating means of each device constituting the network does this, the internal values of all the devices converge to the arithmetic mean value. Further, a first value is set as the generated internal value in any one of the devices constituting the network, and a second value is set as the generated internal value in the other devices. Therefore, the converged generated internal value is a value calculated from the number of devices constituting the network, the first value, and the second value. For example, when the first value is "1" and the second value is "0", the converged generated internal value is the reciprocal of the number of devices constituting the network. The number search means calculates the number of devices constituting the network from the converged generated internal value. As a result, the number of devices constituting the network can be searched even when the management device for managing the entire devices constituting the network is not provided.

Other features and advantages of the present invention will become more apparent with the detailed description given below with reference to the accompanying drawings.

It is a functional block diagram which shows the internal structure of the apparatus which concerns on 1st Embodiment. It is a figure which shows the connection state of each apparatus in the system which concerns on 1st Embodiment. It is a flowchart for demonstrating the configuration number search process performed by the apparatus which concerns on 1st Embodiment. It is a figure which shows the result of the simulation which confirms that the number of constituents can be searched in each apparatus shown in FIG. It is a figure which shows the other example of the connection state of each apparatus. It is a functional block diagram which shows the internal structure of another Example of a timing generation part. It is a functional block diagram which shows the internal structure of the apparatus which concerns on 2nd Embodiment. It is a flowchart for demonstrating the configuration number search process performed by the apparatus which concerns on 2nd Embodiment. It is a figure which shows the result of the simulation which performed the component number search process which concerns on 2nd Embodiment.

Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings.

FIG. 1 is a functional block diagram showing an internal configuration of the device according to the first embodiment. FIG. 2 is a diagram showing a connection state of each device.

As shown in FIG. 1, the device A sets initial values for the internal value generating unit 1 that generates the internal value X _i , the communication unit 2 that communicates with other devices A, and the internal value X _i. Then, the number search unit 3 is provided to calculate the number of devices constituting the network (hereinafter, referred to as “configured number”) from the converged internal value X _i . Note that FIG. 1 shows only the configurations necessary for searching for the number of constituents. In practice, each device A has other configurations. Even if the other configuration is a configuration for performing consensus cooperative control (that is, a configuration similar to the internal value generation unit 1 for generating a predetermined internal value for control or calculation). It may be a configuration for performing other functions.

The communication unit 2 communicates with another device A. The communication unit 2 receives the internal value X _i generated by the internal value generation unit 1 and transmits it to the communication unit 2 of the other device A. Further, the communication unit 2 outputs the internal value X _j received from the communication unit 2 of the other device A to the internal value generation unit 1. The communication method is not limited, and may be wired communication or wireless communication. The internal value X _i corresponds to the “generated internal value” of the present invention, and the internal value X _j corresponds to the “received internal value” of the present invention.

As shown in FIG. 2A, the device A communicates with another device A, and the entire communicating device A constitutes a network. FIG. 2A shows a state in which eight devices A (A1 to A8) form a network. In this case, the devices A1 to A8 constituting the network correspond to the "system" of the present invention. It should be noted that the actual system is composed of more devices A, but a few cases are shown for the sake of brevity.

The solid arrow shown in FIG. 2A indicates that mutual communication is being performed. The device A1 communicates only with the device A2 and the device A8, and the device A2 communicates only with the device A1 and the device A3. Further, the device A3 communicates only with the device A2 and the device A4, the device A4 communicates only with the device A3 and the device A5, and the device A5 communicates only with the device A4 and the device A6. ing. Further, the device A6 communicates only with the device A5 and the device A7, the device A7 communicates only with the device A6 and the device A8, and the device A8 communicates only with the device A7 and the device A1. ing. As described above, the communication unit 2 of the device A is connected to the communication unit 2 of at least one device A (for example, a device located in the vicinity or a device for which communication has been established) among the devices A constituting the network. The network may be connected as long as it is communicating and a communication path exists for any two devices A constituting the network (hereinafter, this state is referred to as a "connected state"). It is not necessary to communicate with the communication unit 2 of all the constituent devices A.

For example, when the device A is the device A2, the communication unit 2 transmits the internal value X ₂ generated by the internal value generation unit 1 to the communication units 2 of the devices A1 and A3, and the communication unit 2 of the device A1 sends the internal value X ₁ is received, and the internal value X ₃ is received from the communication unit 2 of the device A3.

Returning to FIG. 1, the internal value generation unit 1 is realized by, for example, a microcomputer, and the generated internal value X _i and the internal value X _j of another device A input from the communication unit 2 are _combined . Is used to generate the internal value X _i . Even if the internal value X _i and the internal value X _j are different, the internal value X _i and the internal value X _j converge to a common value by repeating the arithmetic processing in the internal value generation unit 1. As shown in FIG. 1, the internal value generation unit 1 includes a calculation unit 11, a multiplier 12, and an integrator 13.

The calculation unit 11 performs a calculation based on the following equation (1). That is, the calculation unit 11 subtracts the internal value X _i generated by the internal value generation unit 1 from each internal value X _j input from the communication unit 2, and multiplies the calculation result u _i by adding all the subtraction results. Output to the device 12.

For example, when the device A is the device A2 (see FIG. 2A) and the internal value X _j is received from the device A1 and the device A3 at the same time, the calculation unit 11 performs the calculation of the following equation (2) to perform the calculation. The result u ₂ is output.

The multiplier 12 multiplies the calculation result u _i input from the calculation unit 11 by a predetermined coefficient ε and outputs the result to the integrator 13. The coefficient ε is a value that satisfies 0 <ε <1 / d _max and is preset. d _max is the maximum number of internal values X _j that may be input to the communication unit 2 at the same time. The coefficient ε is multiplied by the calculation result u _i in order to prevent the input to the integrator 13 from becoming too large (small) and the fluctuation of the internal value X _i becoming too large.

The integrator 13 generates and outputs an internal value X _i by integrating the values input from the multiplier 12. Integrator 13 generates an internal value X _i by adding the value input from the multiplier 12 to the internal value X _i previously generated. The internal value X _i is output to the communication unit 2 and the calculation unit 11 via the switching unit 31.

In this embodiment, the internal value generating unit 1 uses the internal value X _i of the generated, is input from the communication unit 2, an internal value X _j of the other device A, to generate an internal value X _i. When the internal value X _i is larger than the arithmetic mean value of each internal value X _j , the calculation result u _i output by the calculation unit 11 becomes a negative value. Then, a negative value is added to the integrator 13, so that the internal value X _i becomes smaller. On the other hand, when the internal value X _i is smaller than the arithmetic mean value of each internal value X _j , the calculation result u _i output by the calculation unit 11 becomes a positive value. Then, a positive value is added to the integrator 13, so that the internal value X _i becomes large. That is, the internal value X _i approaches the arithmetic mean value of each internal value X _j . By performing this process in each device A, the internal value X _i of each device A converges to the same value. It has been mathematically proved that the internal value X _i converges to the same convergence value X α (see Non-Patent Documents 1 and 2). It has also been proved that the convergence value Xα is the arithmetic mean value of the initial values of the internal values X _i of each device A, as shown in the following equation (3). n is the number of devices A constituting the network (number of components), and the following equation (3) is the number of devices _n by adding all the initial values of the internal values X _{1 to} X _n of the devices A _{1 to} An. It shows that the arithmetic mean value divided is calculated.

The number search unit 3 is realized by, for example, a microcomputer or the like, and is for searching the number of constituents. As shown in FIG. 1, the number search unit 3 includes a switching unit 31, a timing generation unit 32, a convergence detection unit 33, and a number calculation unit 34.

The switching unit 31 sets an initial value to the internal value X _i . Switching unit 31 is inputted to the internal value X _i from the integrator 13 during normal outputting the internal value X _i as it is to the calculation unit 11 and the communication unit 2. However, when an instruction for inputting an initial value is input from the timing generation unit 32, "0" or "1" is set as the initial value of the internal value X _i and output to the calculation unit 11 and the communication unit 2. The initial value is basically set to "0", but "1" is set in any one of the devices A constituting the network. It should be noted that any device A may be set to "1" as the initial value. For example, "0" may be set for all devices A, and only the device A closest to the user may be changed to "1". In the present embodiment, "1" corresponds to the "first value" of the present invention, and "0" corresponds to the "second value" of the present invention.

The timing generation unit 32 outputs an initial value input instruction to the switching unit 31 at a predetermined timing. The predetermined timing is set in advance as the timing for searching for the number of components. For example, if the number of components is searched for maintenance, it is set according to the timing of maintenance, and if the coefficient corresponding to the number of components is used for control, it is set according to the timing required for the control. .. Since it is necessary to set the initial value at the same timing in each device A, the timing generation unit 32 of each device A shares the time for matching the timing. For example, the timing generation unit 32 may access the Internet to obtain accurate time on a regular basis, or if it is equipped with GPS (Global Positioning System), obtain time information from the received information. You may do so. Further, the timing generation unit 32 may output a timing signal instructing communication to the communication unit 2.

The convergence detection unit 33 detects that the internal value X _i has converged, and outputs the convergent internal value X _i, which is the converged internal value X _i , to the number calculation unit 34. The convergence detection unit 33 determines that the internal value X _i has converged when the internal value X _i output by the switching unit 31 does not change. Specifically, the convergence detection unit 33 calculates the difference between the internal value X _i and the previous internal value X _i, and the internal value X _i changes when the difference is within a predetermined error range. Judge that there is no. Then, when the state continues for a predetermined time, it is determined that the state has converged. In the present embodiment, the number of calculations performed by the calculation unit 11 is used as a predetermined time. The configuration of the convergence detection unit 33 is not limited. The convergence detection unit 33 may output the convergence value Xα. For example, it may be determined that the difference between the internal value X _i and the previous internal value X _i has converged when the difference is within a predetermined error range. However, in this case, the range of the predetermined error needs to be sufficiently narrow. In addition, considering the time required for convergence, when the time has elapsed sufficiently, it is determined that the internal value X _i has converged, and the internal value X _i at this time is output as the convergence value X α. May be good. However, the time required for convergence varies depending on the number of components and the state of communication connection of each device A. When the number of components is small, or when each device A communicates with a large number of devices A, the time required for convergence is short, but in the opposite case, the time required for convergence is long. Therefore, if the number of components and the state of communication connection of each device A can be estimated in advance and the time required for convergence can be estimated, the time may be set accordingly. On the other hand, if the time required for convergence cannot be estimated, it is necessary to set a sufficiently long time.

The number calculation unit 34 calculates the number of constituent units based on the convergence value Xα input from the convergence detection unit 33. As described above, the convergence value Xα is the arithmetic mean value of the initial values of the internal values X _i of each device A (see the above equation (3)). As the initial value of the internal value X _i of each device A, only one "1" is set, and the other values are set to "0". Therefore, since the total value of the initial values is "1", the convergence value Xα is 1 / n, where n is the number of components. Therefore, the number of components n is the reciprocal of the convergence value Xα (n = 1 / Xα). The number calculation unit 34 calculates the reciprocal (1 / Xα) of the convergence value Xα input from the convergence detection unit 33 and outputs it as the number of constituents n. The output number of components n is used for calculation of a coefficient for control, or is displayed on a display unit (not shown) and used for maintenance. The use destination of the constituent number n is not limited.

FIG. 3 is a flowchart for explaining a process for searching for the number of constituents n performed by the apparatus A (hereinafter, referred to as a “number of constituents search process”).

First, it is determined whether or not there is a timing input indicating an instruction for inputting the initial value (S1). Specifically, it is determined whether or not the initial value input instruction has been input from the timing generation unit 32 to the switching unit 31. If there is no timing input (S1: NO), the process returns to step S1 and the determination is repeated. If there is a timing input (S1: YES), the process proceeds to step S2. That is, the specific processing (S2 to S7) is started when the timing input is waited and the timing input is performed.

Next, the initial value is input to the internal value X _i (S2). Specifically, the switching unit 31 sets the initial value "0" or "1" to the internal value X _i . Next, the internal value X _i is transmitted (S3), and the internal value X _j is received (S4). Specifically, the communication unit 2 transmits the internal value X _i input from the internal value generation unit 1 to the other device A, and also receives the internal value X _j from the other device A. Then, the internal value X _i is generated (S5). Specifically, the arithmetic unit 11 performs an operation based on the internal value X _i and the internal value X _j , the multiplier 12 multiplies the operation result u _i by a predetermined coefficient ε, and the integrator 13 calculates the multiplication result. By integrating, the internal value X _i is generated.

Next, it is determined whether or not the internal value X _i has converged (S6). Specifically, the convergence detection unit 33 determines whether or not the internal value X _i has converged. If it has not converged (S6: NO), the process returns to step S3, and steps S3 to S6 are repeated. When it converges (S6: YES), the number of components is calculated (S7), and the search process for the number of components ends. Specifically, when the convergence detection unit 33 determines that the internal value X _i has converged, the convergence value Xα is output to the number calculation unit 34, and the number calculation unit 34 calculates the number of components based on the convergence value Xα. And output. The configuration number search process is not limited to the above.

Next, in each of the devices A1 to A8 shown in FIG. 2, a simulation for confirming that the number of components can be searched will be described.

FIG. 4 is a diagram showing the results of the simulation. FIG. 4A shows the time change of the internal value X _i of each of the devices A1 to A8 shown in FIG. 2A. The vertical axis shows the value of each internal value X _i , and the horizontal axis shows the number of calculations in the calculation unit 11. The simulation was started with the initial value of the device A1 being "1" and the initial value of the devices A2 to A8 being "0".

As shown in FIG. 4A, the internal values X _i of each of the devices A1 to A8 approach the same value as the number of operations increases, and converge to almost the same value when the number of operations is about 80. .. The convergence value Xα is 0.125. Therefore, the number of components n = 1 / Xα = 1 / 0.125 = 8 is calculated.

FIG. 4 (b) shows the time change of the internal value X _i of each of the devices A1 to A8 shown in FIG. 2 (b). FIG. 2B shows a state in which the device A8 is disconnected from the network in FIG. 2A. That is, the device A1 communicates with the device A2 only, and the device A2 communicates with the device A1 and the device A3 only. Further, the device A3 communicates only with the device A2 and the device A4, the device A4 communicates only with the device A3 and the device A5, and the device A5 communicates only with the device A4 and the device A6. ing. Further, the device A6 communicates with the device A5 and the device A7 only, and the device A7 communicates with the device A6 only. The device A8 does not communicate with any of the devices. Even in this case, the devices A1 to A7 are in a connected state. Similar to the simulation in FIG. 4A, the simulation was started with the initial value of the device A1 being “1” and the initial value of the devices A2 to A8 being “0”.

As shown in FIG. 4 (b), the internal value X _i of each device A1~A7 each time the number of calculations is increased approaches the same value, but the internal value X _i of the device A8 remains "0" is there. The internal values X _i of each of the devices A1 to A7 converge to almost the same value when the number of operations is about 200. The convergence value Xα is 0.142857. Therefore, in each of the devices A1 to A7, the number of components n = 1 / Xα = 1 / 0.1422857 = 7 is calculated. On the other hand, the internal value X _i of the device A8 remains "0", the convergence value Xα becomes "0", and the number of components n = 1 / Xα cannot be calculated. Therefore, the device A8 may be disconnected from the network. I understand.

Next, the operation and effect of the device A will be described.

According to this embodiment, the internal value generating unit 1, by the operation of the communication unit 2 and the generated internal value X _i is based on the internal value X _j received, generates the internal value X _i. When the internal value generation unit 1 of each device A constituting the network does this, the internal value X _i in all the devices A converges to the arithmetic mean value. Further, "1" is set as the internal value X _i in any one of the devices A constituting the network, and "0" is set as the internal value X _i in the other devices A. .. Therefore, the convergent value Xα, which is the converged internal value X _i , is the reciprocal of the number of constituents n. The number calculation unit 34 calculates the constituent number n from the convergence value Xα. As a result, the number of devices A constituting the network can be searched even when the management device for managing the entire device A constituting the network is not provided.

Further, according to the present embodiment, "1" is set in one device A and "0" is set in the other device A as the initial value of the internal value X _i . Therefore, the convergence value Xα is the reciprocal of the number of constituents n. Therefore, the number of components n can be easily calculated.

In the present embodiment, the case where each device A performs mutual communication has been described, but the present invention is not limited to this, and one-sided communication may be performed. For example, as shown in FIG. 5, the device A1 only receives from the device A8 and only transmits to the device A2, and the device A2 only receives from the device A1 and only transmits to the device A3. A3 only receives from device A2 and only transmits to device A4, device A4 only receives from device A3 and only transmits to device A5, and device A5 only receives from device A4. , Only transmitting to device A6, device A6 only receiving from device A5, only transmitting to device A7, device A7 only receiving from device A6, only transmitting to device A8, device Even when A8 only receives from the device A7 and only transmits to the device A1, the internal value X _i can be converged to the additive average value. More generally, the internal value X is a state in which an arbitrary device A can be reached (a state of "including a spanning tree" in graph theory) when a destination is traced from a certain device A. It is a condition for converging _i , and it is a state in which an arbitrary device A can be reached by tracing a destination from an arbitrary device A (a "strongly connected" state in graph theory), and , In all devices A, the state where the number of destination devices A and the number of source devices A are equal (“balance graph” state in graph theory) converges the internal value X _i to the arithmetic mean value. It is a condition to make it.

Further, in the present embodiment, the case where "1" is set in one device A and "0" is set in the other device A as the initial value of the internal value X _i has been described, but this is limited to this. I can't. Even if the initial value is other than these, the number of components n can be calculated from the convergence value Xα. For example, assuming that the initial value of one device A is "a" and the initial value of the other device A is "b", the number of components n is n = (ab) / (Xα-b). Can be calculated. By setting b = 0, it can be easily calculated as n = a / Xα, and by setting a = 1, it can be calculated more easily by setting n as the reciprocal of Xα. That is, setting "1" in one device A and setting "0" in the other device A makes the calculation in the number calculation unit 34 the easiest. When the expected number of components n'is set in one device A and "0" is set in the other device A, if the actual number of components n is equal to the expected number of components n', the convergence value. Xα becomes "1". Therefore, in this case, if the convergence value Xα is “1”, it can be determined that the actual number of components is equal to the expected number of components without performing the calculation based on the convergence value Xα.

Further, in the present embodiment, the case where the timing generation unit 32 outputs the initial value input instruction to the switching unit 31 at the timing based on the time information obtained from the Internet or the like has been described, but the present invention is not limited to this. Other examples of the timing generation unit 32 will be described below.

FIG. 6 is a functional block diagram showing an internal configuration of the timing generation unit 32', which is another embodiment. The timing generation unit 32'synchronizes the internal phase θ _i of each device A with a configuration similar to the internal value generation unit 1, and instructs the switching unit 31 to input the initial value at the timing based on the internal phase θ _i. Output. The timing generation unit 32'includes an internal phase generation unit 8 and a timing output unit 9. In addition to the internal value X _i , the communication unit 2 also transmits and receives the internal phase θ _i .

The internal phase generation unit 8 generates the internal phase θ _i for determining the timing at which the initial value input instruction is output to the switching unit 31. The internal phase generation unit 8 outputs the generated internal phase θ _i to the communication unit 2 and the timing output unit 9. Internal phase generating unit 8 includes an internal phase theta _i which generated, is input from the communication unit 2, by using the internal phase theta _j of another apparatus A, and generates an internal phase theta _i. Even if the internal phase θ _i and the internal phase θ _j are different, the internal phase θ _i and the internal phase θ _j converge to a common internal phase by repeating the arithmetic processing in the internal phase generating unit 8. As shown in FIG. 6, the internal phase generation unit 8 includes a calculation unit 81, a multiplier 82, an adder 83, and an integrator 84.

The calculation unit 81 performs a calculation based on the following equation (4). That is, the arithmetic unit 81, from the internal phase theta _j inputted from the communication unit 2, an internal phase theta _i of the internal phase generating unit 8 is generated by subtracting each operation result u _'i obtained by adding all the subtraction results Output to the multiplier 82.

The multiplier 82 outputs to the adder 83 is multiplied by a 'predetermined coefficient ε in _i' computation result u input from the arithmetic unit 81. The coefficient ε'is a value that satisfies 0 <ε'<1 / d' _max and is preset. d' _max is the maximum value of the number of internal phases θ _j that may be input to the communication unit 2 at the same time. The coefficient epsilon 'has a large input to the integrator 84 (small) becomes too, in order to suppress the fluctuation of the internal phase theta _i is too large, the operation result u' intended to be multiplied by _i is there. Therefore, when the processing in the internal phase generation unit 8 is continuous time processing, it is not necessary to provide the multiplier 82.

The adder 83 adds the input from the multiplier 82, the predetermined angular frequency ω _0, and the value Tω ₀ based on the sampling period T of the integrator 84, and outputs the value to the integrator 84. The value Tω ₀ is input at the timing of the sampling frequency f (= 1 / T), and the multiplication result is obtained from the multiplier 82 at the timing when the calculation unit 81 performs the calculation (that is, the timing when the internal phase θ _j is received). Is entered. The integrator 84 generates and outputs the internal phase θ _i by integrating the input from the adder 83. The integrator 84 generates an internal phase theta _i by adding the input from the adder 83 to the internal phase theta _i previously generated. Further, in the integrator 84, in order to prevent the internal phase θ _i from diverging infinitely, the internal phase θ _{i is set} to (0 ≦ θ _i <S) with a predetermined threshold value S (a multiple of 2π) as an upper limit value. Output as a value in the range of. That is, when the internal phase θ _i reaches the threshold value S which is the upper limit value of the range, the integrator 84 clears the internal phase θ _i to “0” which is the lower limit value of the range. The internal phase θ _i is output to the timing output unit 9, the communication unit 2, and the calculation unit 81. In the present embodiment, the sampling frequency f of the integrator 84 is increased so that the internal phase θ _i output by the internal phase generation unit 8 can be used as it is by the timing output unit 9, and the sampling frequency f is adjusted. Tω ₀ is added, but it is not limited to this. For example, the sampling frequency f may be lowered (for example, 1 Hz) or higher. In this case, adjustment may be made when outputting to the timing output unit 9 or by dividing the frequency inside the timing output unit 9. Further, the value to be added may be a value unrelated to the angular frequency ω ₀ (for example, “1”).

Internal phase generating unit 8 includes an internal phase theta _i which generated, is input from the communication unit 2, by using the internal phase theta _j of another apparatus A, and generates an internal phase theta _i. If the internal phase theta _i is greater than the arithmetic mean value of each internal phase theta _j, the operation result u _'i the arithmetic unit 81 is output is a negative value. Then, the internal phase θ _i output from the integrator 84 is reduced and approaches the internal phase θ _j . On the other hand, if the internal phase theta _i is smaller than the internal phase theta _j, the operation result u _'i the arithmetic unit 81 is output is a positive value. Then, the internal phase θ _i output from the integrator 84 is increased and approaches the internal phase θ _j . By performing this process in each device A, the internal phase θ _i of each device A converges to the same value. The internal phase θ _i changes with time, and it can be considered that a component that changes according to the angular frequency ω ₀ and a component that changes so as to compensate for the initial phase shift are combined. When the latter converges to the same value θα, the internal phase θ _i of each device A also converges to the same value. It has been mathematically proved that the latter converges to the same value θα (see Non-Patent Documents 1 and 2). It has also been proved that the convergence value θα is the arithmetic mean value of the initial values of the internal phase θ _i of each device A, as shown in the following equation (5). The following equation (5) shows that the arithmetic mean value obtained by adding all the initial values of the internal phases θ _{1 to} θ _n of the devices A1 to An and dividing by n is calculated.

The timing output unit 9 outputs an initial value input instruction to the switching unit 31 based on the internal phase θ _i input from the internal phase generation unit 8. For example, the timing output unit 9 outputs an initial value input instruction at the timing when the internal phase θ _i becomes “0”. The timing of outputting the initial value input instruction is not limited to "0". Further, the output may be performed at the timing when the number of times the internal phase θ _i becomes “0” becomes the planned number of times. Further, the timing output unit 9 may also output the timing signal divided according to the processing cycle to the communication unit 2.

In the timing generation unit 32'according to the present embodiment, the internal phase θ _i of each device A converges to the same value to synchronize. Therefore, the timing generation unit 32'of each device A can output the initial value input instruction to the switching unit 31 at the same timing.

Incidentally, the convergence value θα of the changing components so as to compensate for deviation of the initial phase of the internal phase theta _i may not be the arithmetic mean value of the initial value of the internal phase theta _i of each device A, the internal phase theta _It suffices if _i converges to the same value. Therefore, the calculation formula set in the calculation unit 81 is different from the above formula (4), and the convergence value θα is, for example, the weighted average value of the initial value, the geometric mean value (geometric mean value), the harmonic mean value, and the Pth order. It may be an average value. Further, the convergence value θα may be set to the maximum value (or the minimum value) of the initial value by using an algorithm described later.

In the first embodiment, it is necessary to set "1" as the initial value of the internal value X _i in any of the devices A. A case where the setting of the initial value is automated will be described below as a second embodiment.

FIG. 7 is a functional block diagram showing an internal configuration of the device A'according to the second embodiment. In the figure, the same or similar elements as those of the device A (see FIG. 1) according to the first embodiment are designated by the same reference numerals. The device A'according to the second embodiment includes a calculation unit 11'and a switching unit 31'instead of the calculation unit 11 and the switching unit 31, and a set value determination unit 35. It is different from the device A according to the first embodiment.

The calculation unit 11'includes an average consensus unit 11a and a maximum consensus unit 11b, and switches between the two according to an instruction from the switching unit 31'. The average consensus unit 11a performs the same function as the calculation unit 11 according to the first embodiment. Therefore, 'If is switched to the mean agreement portion 11a, the internal value generating unit 1, the internal value X _i, the apparatus A' calculation unit 11 converges to the arithmetic mean value of the initial value of the internal value X _i of Let me. In the present embodiment, the average consensus unit 11a corresponds to the "calculation means" of the present invention.

The maximum consensus unit 11b performs an operation based on the following equation (6). alpha _ij is "1" when the X _j> X _i, is a function that becomes "0" when the X _j ≦ X _i. That is, the maximum consensus unit 11b subtracts the internal value X _i generated by the internal value generation unit 1 from each internal value X _j input from the communication unit 2, and only those whose subtraction result is a positive value are subtracted. The calculation result u _i obtained by adding all of them is output to the multiplier 12. In the present embodiment, the maximum consensus unit 11b corresponds to the "second arithmetic means" of the present invention.

For example, in the case of the device A2 shown in FIG. 2A, when the device A'is X ₁ > X ₂ > X ₃ , the maximum agreement unit 11b performs the calculation of the following equation (7), and the calculation result u Output ₂

Internal value generation unit 1 includes an internal value X _i of the generated, is input from the communication unit 2, by using the internal value X _j of the other device A ', and generates an internal value X _i. While the arithmetic unit 11 'is switched to the maximum agreement portion 11b, when the internal value X _i is either smaller than the internal value X _j, the difference is added, the internal value X _i increases. As a result, the internal value X _i approaches the maximum value of the internal value X _j . When the internal value X _i matches the maximum value of the internal value X _j , the operation result u _i becomes "0", and the operation result u _i does not change. However, since this process is performed by each device A', if there is a larger internal value, one of the internal values X _j changes to match the internal value, and the internal value X _i is also inside it. It will match the value X _j . Therefore, the internal value X _i converges to the maximum value as a whole. That is, the internal value generation unit 1 converges the internal value X _i to the maximum value of the initial value of the internal value X _i of each device A'.

When an initial value input instruction is input from the timing generation unit 32, the switching unit 31'generates a random number, sets the random number as the initial value of the internal value X _i , and performs the calculation unit 11'and the communication unit 2'. Output to. At this time, the calculation unit 11'is switched to the maximum consensus unit 11b. Therefore, the internal value generation unit 1 functions to converge the internal value X _i to the maximum value of the initial value of the internal value X _i of each device A'. Then, when the set value is input from the set value determination unit 35, the switching unit 31'switches the calculation unit 11'to the average consensus unit 11a, sets the set value as the initial value of the internal value X _i , and calculates. Output to unit 11'and communication unit 2. Therefore, the internal value generation unit 1 functions to converge the internal value X _i to the arithmetic mean value of the initial value of the internal value X _i of each device A'.

The set value determination unit 35 determines the set value to be input to the switching unit 31'based on the internal value X _i output by the switching unit 31'. Setting value determination unit 35, the initial value of the internal value X _i is the device A 'when the maximum value of the initial value of the internal value X _i of the set value "1" switching unit 31' outputs the In other cases, the set value "0" is output to the switching unit 31'. Since the maximum value is one, the set value "1" is output to the switching unit 31'only in any one device A', and the set value "0" is output to the switching unit 31'in the other devices A'. It will be output. Since the internal value generation unit 1 in which the calculation unit 11'is switched to the maximum consensus unit 11b functions to converge the internal value X _i to the maximum value of the initial value of the internal value X _i of each device A', the internal value generation unit 1 functions. If the initial value of the internal value X _i is not the maximum value of the initial value of the internal value X _i of each apparatus a ', the internal value X _i is changed so as to converge to the maximum value. On the other hand, if the initial value of the internal value X _i is the maximum value of the initial value of the internal value X _i of each apparatus A ', the internal value X _i is not changed. Therefore, it can be determined whether or not the initial value of the internal value X _i is the maximum value depending on whether or not the internal value X _i changes. Setting value determination unit 35, 'and the initial value of the internal value X _i which is outputted after a predetermined time has elapsed (e.g., arithmetic unit 11' switching unit 31 after the operation in is performed a predetermined number of times) in the switching unit 31 ' Compares with the internal value X _i output by, and if there is no change, it is determined that the internal value X _i is the maximum value, and the set value "1" is output to the switching unit 31'. On the other hand, when there is a change, it is determined that the internal value X _i is not the maximum value, and the set value "0" is output to the switching unit 31'. The time required for convergence varies depending on the number of components n and the communication connection of each device A'. When the number of components n is small, or when each device A'is communicating with a large number of devices A', the time required for convergence is short, and even if the predetermined time is shortened, the set value determination unit 35 It is possible to appropriately judge whether or not it is the maximum value. On the contrary, when the time required for convergence is long, it is necessary to lengthen the predetermined time so that the set value determining unit 35 can appropriately determine. The configuration of the set value determination unit 35 is not limited. The set value determining unit 35 may determine whether or not the internal value X _i is the maximum value, and may output the set value according to the determination.

FIG. 8 is a flowchart for explaining the configuration number search process performed by the device A'.

First, it is determined whether or not there is a timing input indicating an instruction for inputting an initial value (S11). Specifically, it is determined whether or not the initial value input instruction has been input from the timing generation unit 32 to the switching unit 31. If there is no timing input (S11: NO), the process returns to step S11 and the determination is repeated. If there is a timing input (S11: YES), the process proceeds to step S12. That is, waiting for the timing input, when there is a timing input, specific processing (S11 to S12 to S24) is started.

Next, a random number is input to the internal value X _i (S12). Specifically, the switching unit 31'generates a random number and sets the random number as the initial value of the internal value X _i . Next, the internal value X _i is transmitted (S13), and the internal value X _j is received (S14). Specifically, the communication unit 2 transmits the internal value X _i input from the internal value generation unit 1 to the other device A', and also receives the internal value X _j from the other device A'. Then, the internal value X _i is generated (S15). Specifically, the calculation unit 11'(maximum agreement unit 11b) performs an operation based on the internal value X _i and the internal value X _j , and the multiplier 12 multiplies the operation result u _i by a predetermined coefficient ε. The integrator 13 integrates the multiplication result to generate the internal value X _i .

Next, it is determined whether or not the predetermined time has elapsed (S16). If the predetermined time has not elapsed (S16: NO), the process returns to step S13, and steps S13 to S16 are repeated. When the predetermined time has elapsed (S16: YES), it is determined whether or not the internal value X _i has changed (S17). Specifically, the setting value determination unit 35 to determine by comparing the initial value of the internal value X _i, and the internal value X _i input from the switching unit 31 '. When the internal value X _i changes (S17: YES), it is determined that the internal value X _i is not the maximum value, and the set value “0” is input to the internal value X _i (S18). On the other hand, when the internal value X _i does not change (S17: NO), it is determined that the internal value X _i is the maximum value, and the set value “1” is input to the internal value X _i (S19). Specifically, the set value determination unit 35 outputs the set value "0" or "1" to the switching unit 31'according to the determination, and the set value input by the switching unit 31'is set to the internal value X _i . Set.

Since steps S20 to S24 are the same as steps S3 to S7 in the flowchart shown in FIG. 3, description thereof will be omitted. The configuration number search process is not limited to the above.

Next, a simulation in which the component number search process according to the second embodiment will be performed will be described.

FIG. 9 is a diagram showing the results of the simulation. 9 (a) shows a simulation for each of the devices A1 to A8 shown in FIG. 2 (a) (the internal configuration is the internal configuration of the device A'according to the second embodiment (see FIG. 7)). It shows the time change of the internal value X _i of each device A1 to A8 when it is performed. The vertical axis shows the value of each internal value X _i , and the horizontal axis shows the number of calculations in the calculation unit 11'. The simulation was started with the random numbers generated in each of the devices A1 to A8 as initial values.

As shown in FIG. 9A, the device A1 in which the initial value of the internal value X _i is the maximum value of 0.92 continues the initial value, and the internal values X _i of the other devices A2 to A8 are the maximum values. It is approaching 0.92. Then, when the number of calculations is 100 times, the set value determination unit 35 of each device A1~A8 determines the set value, respectively, "1" is set in the internal value X _i of the device A1, device A2~A8 "0" is set for the internal value X _i of. Since random numbers generated by the device A1 is in this simulation has become maximum, although "1" was to be set in the internal value X _i of the device A1, generated by any device A1~A8 Depending on whether the random number to be set becomes the maximum value, which one the internal value X _i is set to "1" changes. After that, as the number of operations increases, the internal values X _i of each of the devices A1 to A8 approach the same value, and when the number of operations is about 200, they converge to almost the same value. The convergence value Xα is 0.125. Therefore, the number of components n = 1 / Xα = 1 / 0.125 = 8 is calculated.

FIG. 9B shows a simulation in which eight devices A1 to A8 (internal configuration is the internal configuration of the device A'according to the second embodiment (see FIG. 7)). The time change of the internal value X _i of each device A1 to A8 is shown. The simulation was started with the random numbers generated in each of the devices A1 to A8 as initial values.

As shown in FIG. 9B, the device A6 in which the initial value of the internal value X _i is the maximum value of 0.77 continues the initial value, and the internal values X _i of the devices A7 and A8 have the maximum value of 0. It is approaching 77. Further, the internal value X _i of the device A1 continues to be the initial value 0.28, and the internal value X _i of the devices A2 to A5 is approaching 0.28. Then, when the number of operations is 100, the set value determining unit 35 of each of the devices A1 to A8 determines the set value, and "1" is set for the internal value X _i of the devices A1 and A6, and the device A2. "0" is set for the internal values X _{i of} ~ A5 and A7 to A8. After that, as the number of operations increases, the internal values X _i of each of the devices A1 to A5 approach the same value, and when the number of operations is about 140, they converge to almost the same value. The convergence value Xα is 0.2. Therefore, it is calculated that the number of components n = 1 / Xα = 1 / 0.2 = 5. Further, the internal values X _i of each of the devices A6 to A8 approach the same value, and the number of operations is about 160, and the values converge to almost the same value. The convergence value Xα is 0.33. Therefore, it is calculated that the number of constituents n = 1 / Xα = 1 / 0.33≈3. That is, it can be seen that the devices A1 to A5 form a network of five constituents, and the devices A6 to A8 form a network of three constituents.

Next, the operation and effect of the device A'will be described.

According to the second embodiment, the switching unit 31'sets a random number as the initial value of the internal value X _i when the timing generation unit 32 inputs the initial value input instruction. Then, the internal value generation unit 1 is internally operated by an operation based on the generated internal value X _i and the internal value X _j received by the communication unit 2 in a state where the calculation unit 11'is switched to the maximum consensus unit 11b. Generate the value X _i . By doing this, the internal value generation unit 1 of each device A'that constitutes the network converges the internal value X _i of all the devices A'to the maximum value. The set value determining unit 35 outputs the set value "1" when the initial value of the internal value X _i is the maximum value, and outputs the set value "0" in other cases. Thus, "1" is set as the internal value X _i in 'any one of the apparatus A of the' device A constituting the network, is "0" as an internal value X _i In other devices A ' Set. Therefore, the initial value of the internal value X _i can be automatically set without human intervention. Further, since the subsequent processing is the same as that of the device A according to the first embodiment, the same effect as that of the first embodiment can be obtained in the second embodiment.

In the second embodiment, the case where the maximum consensus unit 11b performs the calculation based on the above equation (6) has been described, but the present invention is not limited to this. If it can be converged to the maximum value, another arithmetic expression may be used. For example, the maximum consensus unit 11b may perform an operation based on the following equation (8). β _ij is a function that becomes "0" in all cases of X _j ≤ X _i (that is, X _i is the maximum value), and becomes "1" in other cases. That is, the calculation unit 11'outputs "0" as the calculation result u _i when the internal value X _i generated by the internal value generation unit 1 is equal to or greater than all the internal values X _j input from the communication unit 2. In other cases, the calculation result u _i is output by subtracting the internal value X _i from each internal value X _j and adding all the subtraction results. Even in this case, the internal value X _i can be converged to the maximum value.

Further, in the second embodiment, whether the set value is set to "1" or "0" is determined depending on whether or not the initial value of the internal value X _i is the maximum value. Not limited. For example, it may be determined whether the set value is "1" or "0" depending on whether or not the initial value of the internal value X _i is the minimum value. In this case, instead of the maximum consensus unit 11b that performs the calculation based on the above equation (6), the minimum consensus unit (not shown) that performs the calculation based on the following equation (9) may be provided. α _ij'is a function such that “1” when X _j <X _i and “0” when X _j ≧ X _i . That is, the minimum consensus unit subtracts the internal value X _i generated by the internal value generation unit 1 from each internal value X _j input from the communication unit 2, and only those whose subtraction result is a negative value are all. The added calculation result u _i is output to the multiplier 12. In the embodiment, the minimum consensus unit corresponds to the "second arithmetic means" of the present invention. In this case, the internal value X _i can be converged to the minimum value.

The minimum consensus unit is not limited to the case where the calculation based on the above equation (9) is performed. Other arithmetic expressions may be used as long as they can be converged to the minimum value. For example, the minimum consensus unit may perform an operation based on the following equation (10). β _ij'is a function that becomes "0" when all X _j ≥ X _i (that is, X _i is the minimum value), and "1" in other cases. That is, the calculation unit 11'outputs "0" as the calculation result u _i when the internal value X _i generated by the internal value generation unit 1 is equal to or less than all the internal values X _j input from the communication unit 2. In other cases, the calculation result u _i is output by subtracting the internal value X _i from each internal value X _j and adding all the subtraction results. Even in this case, the internal value X _i can be converged to the minimum value.

The present invention can be applied to any device having a communication function and communicating with other devices constituting the network. For example, when each inverter device constituting the electric power system synchronizes the internal phase, or when various timings are matched or shifted so as not to match, each inverter device constituting the electric power system matches the internal compensation value. Therefore, it can be applied to the case where the suppression amount of the output active power and the output invalid power is adjusted, and the case where each measuring device matches the internal average value, the maximum value, and the minimum value based on the measured values. It can also be applied to devices other than devices that perform consensus cooperative control.

The device provided with the communication function, the system provided with the device, and the number search method according to the present invention are not limited to the above-described embodiment. The specific configuration of each part of the device having the communication function, the system provided with the device, and the number search method according to the present invention can be freely redesigned.

A, A1 to A8, A'Device 1 Internal value generator (internal value generator)
11 Calculation unit (calculation means)
11'Calculation unit 11a Average consensus unit (calculation means)
11b Maximum agreement (second calculation means)
12 Multiplier 13 Integrator 2 Communication unit 3 Number search unit (internal value generation means, number search means)
31, 31'Switching unit 32, 32' Timing generation unit 8 Internal phase generation unit 81 Calculation unit 82 Multiplier 83 Adder 84 Integrator 9 Timing output unit 33 Convergence detection unit 34 Number calculation unit 35 Setting value determination unit

Claims (10)

A system in which multiple devices make up a communication network.
The device
Internal value generation means to generate internal value and
A communication means that transmits an internal value generated by the internal value generating means as a generated internal value to at least one of other devices, and receives an internal value transmitted by at least one of the other devices as a receiving internal value.
With
The internal value generating means
An arithmetic means for subtracting the generated internal value from the received internal value and adding all the subtraction results.
An integration means that integrates the calculation results output by the calculation means to calculate the internal value, and
After setting the first value or the second value as the initial value of the generated internal value, the generated internal converged by repeating the calculation by the arithmetic means, the integration by the integrating means, and the transmission / reception of the internal value by the communication means. A number search means for calculating the number of devices constituting the network based on the value, and
Is equipped with
The first value is set as the initial value of the generated internal value in any one of the devices constituting the network, and the second value is set as the initial value of the generated internal value in the other devices. Value is set,
A system characterized by that. The first value is "1".
The second value is "0",
The number search means further includes a number calculation means for calculating the number based on the reciprocal of the converged generated internal value.
The system according to claim 1. The number search means further includes a convergence detecting means for detecting that the generated internal value has converged based on a change in the generated internal value.
The system according to claim 1 or 2. The device
A second calculation means that subtracts the generated internal value from the received internal value, adds only those whose subtraction result is a positive value, and outputs the result.
A random number is set as the initial value of the generated internal value, and after the calculation by the second calculation means, the integration by the integration means, and the transmission / reception of the internal value by the communication means are performed, based on the generated internal value. , A set value determining means for determining whether to set the first value or the second value, and
Is further equipped with
After the determination by the set value determining means, the determined set value is set as the initial value of the generated internal value, and the processing by the number search means is performed.
The system according to any one of claims 1 to 3. The set value determining means determines to set the first value when the generated internal value does not change, and sets the second value when the generated internal value changes. Decide that
The system according to claim 4. Internal value generation means to generate internal value and
A communication means that transmits an internal value generated by the internal value generating means as a generated internal value to at least one of other devices, and receives an internal value transmitted by at least one of the other devices as a receiving internal value.
With
The internal value generating means
An arithmetic means for subtracting the generated internal value from the received internal value and adding all the subtraction results.
An integration means that integrates the calculation results output by the calculation means to calculate the internal value, and
After setting the first value or the second value as the initial value of the generated internal value, the generated internal converged by repeating the calculation by the arithmetic means, the integration by the integrating means, and the transmission / reception of the internal value by the communication means. A number search means that calculates the number of devices that make up a network by communication based on the value, and
Is equipped with
A device characterized by that. A second calculation means that subtracts the generated internal value from the received internal value, adds only those whose subtraction result is a positive value, and outputs the result.
A random number is set as the initial value of the generated internal value, and after the calculation by the second calculation means, the integration by the integration means, and the transmission / reception of the internal value by the communication means are performed, based on the generated internal value. , A set value determining means for determining whether to set the first value or the second value, and
Is further equipped with
After the determination by the set value determining means, the determined set value is set as the initial value of the generated internal value, and the processing by the number search means is performed.
The device according to claim 6. The set value determining means determines to set the first value when the generated internal value does not change, and sets the second value when the generated internal value changes. Decide that
The device according to claim 7. A method of searching for the number of devices constituting the network in a communication network composed of devices including an internal value generating means for generating an internal value and a communication means for communicating with at least one other device. And
An internal value generation step in which the internal value generating means generates an internal value, and
A transmission step of transmitting the internal value generated in the internal value generation step as a generated internal value to at least one of the other devices,
A receiving step of receiving an internal value transmitted by at least one of the other devices as a receiving internal value,
Is to be done by each device,
The internal value generation step is
A calculation step of subtracting the generated internal value from the received internal value and adding all the subtraction results.
An integration process that integrates the calculation results output in the calculation process to calculate the internal value, and
Is equipped with
A first value is set as the initial value of the generated internal value in any one of the devices constituting the network, and a second value is set as the initial value of the generated internal value in the other devices. Setting process to set and
After the setting step, the internal value generation step, the transmission step, and the reception step are repeated, and the number of devices constituting the network is calculated based on the converged generated internal value.
A method characterized by having. The setting process is
A second calculation step in which the generated internal values are subtracted from the received internal values, and only those whose subtraction results are positive are added and output.
A random number setting process that sets a random number as the initial value of the generated internal value, and
After the second calculation step, the integration step, the transmission step, and the reception step are performed after the random number setting step, the first value is set or the first value is set based on the generated internal value. The decision process to decide whether to set the value of 2 and
Is equipped with
The set value determined by the determination step is set as the initial value of the generated internal value.
The method according to claim 9.