Information processing device, information processing method, and program

The information processing device and method address the challenge of estimating system inertia in AC interconnected power systems by using a system matrix analysis and high-pass filtering to improve estimation accuracy during normal and abnormal conditions, ensuring stable frequency and synchronization.

JP7593139B2Active Publication Date: 2024-12-03TOKYO ELECTRIC POWER CO HOLDINGS INC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
JP2021010132
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2021-01-26
Publication Date
2024-12-03
Estimated Expiration
2041-01-26

AI Technical Summary

Technical Problem

Existing methods for estimating system inertia in AC interconnected power systems are limited to abnormal conditions and fail to accurately determine inertia during normal operations, especially with the increasing integration of renewable energy sources and power converters that lack inertia, making it difficult to maintain stable frequency and synchronization.

Method used

An information processing device and method that estimates system inertia by analyzing a numerically obtained system matrix of the AC interconnection system, utilizing phase angle and rotational speed measurements to identify the inertia ratio, and applying high-pass filtering techniques to isolate high-frequency components, thereby improving estimation accuracy during both normal and abnormal conditions.

Benefits of technology

Enables precise estimation of system inertia during both normal and abnormal conditions, enhancing the stability and frequency regulation of AC interconnected power systems by accurately determining the inertia ratio of generators and power sources.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007593139000079
    Figure 0007593139000079
  • Figure 0007593139000080
    Figure 0007593139000080
  • Figure 0007593139000081
    Figure 0007593139000081
Patent Text Reader

Abstract

To provide an information processing device capable of estimating not only values related to the inertia of a power system in abnormal conditions but also values related to the inertia of the power system in normal conditions.SOLUTION: An information processing device is provided, comprising a control unit estimating an inertia ratio on the basis of a system matrix numerically obtained from a sway equation of an AC interconnection system. For example, the control unit identifies the system matrix by using records for a constant period of at least one of a phase angle and a rotational speed in normal conditions. The control unit applies an HPF (high-pass filter) or a BPF (band-pass filter) to at least one of measurement amounts used for the identification.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Research and development is being conducted on technology to estimate the system inertia of an AC interconnected system. An AC interconnected system is an AC power system that includes one or more generators and generates electricity synchronously using these one or more generators. The system inertia of an AC interconnected system is the sum of the inertia moments of the rotating bodies (i.e., generator rotors, turbines, etc.) of one or more generators included in the AC interconnected system.

[0003] Here, in an AC interconnection system, the rotors of one or more generators included in the AC interconnection system all rotate synchronously to generate a single system frequency. For this reason, when the system inertia of the AC interconnection system changes due to, for example, the disconnection of a generator in the AC interconnection system, the system frequency fluctuates. Fluctuations in the system frequency can affect the quality of power supply by the AC interconnection system and the stability of the synchronization between the generators. Therefore, from the viewpoint of a stable supply of power by the AC interconnection system, the system inertia of the AC interconnection system is required to always be equal to or greater than a certain value. For these reasons, it is often desired to accurately estimate the system inertia of the AC interconnection system.

[0004] For example, the stability of the frequency of the power supplied by a power system depends on the sum of the inertias of the generators and sources connected to the power system and operating in synchronous with each other, and has been maintained mainly by the inertia of synchronous generators. However, in recent years, there has been an increase in cases where renewable energy sources are connected to the power grid via power converters constructed using semiconductors. Since such power converters do not have inertia, in such cases the system inertia is reduced. In addition, due to an increase in the number of generators and power sources connected to the power grid and the liberalization of the power industry, there are also an increasing number of cases where it is difficult to grasp the specifications or operating status of individual generators and power sources in a unified manner.

[0005] Therefore, a technique for estimating the total inertia of generators and sources connected to a power grid for use in maintaining the frequency of the power supplied by the power grid has attracted attention. One example of such a technique is the inertia estimation method disclosed in Non-Patent Document 1 (see Non-Patent Document 1).

[0006] Incidentally, Non-Patent Document 2 describes the Rate of Change of Frequency (RoCoF) method (see Non-Patent Document 2). Moreover, Non-Patent Document 3 describes estimation of a power oscillation mode by power system identification (see Non-Patent Document 3). [Prior art documents] [Non-patent literature]

[0007] [Non-Patent Document 1] Davide del Giudice and Samuele Grillo, “Analysis of the Sensitivity of the Extended Kalman Filter-Based Inertia Estimation Method to the Assumed Time of Disturbance”, Energies 12, 483 (2019). [Non-Patent Document 2] "Development of a method for estimating system frequency characteristics based on actual measurement results", Report of the Central Research Institute of the Electric Power Industry, Research Report: T94016, May 1995 [Non-Patent Document 3] Hiraiwa et al., "Estimation of Power Oscillation Modes by Power System Identification Using Pseudo Load Models," IEEJ Trans. PE, Vol.127, No.3, 2007 Summary of the Invention [Problem to be solved by the invention]

[0008] However, the inertia estimation method described in Non-Patent Document 1 estimates inertia using large frequency fluctuations that occur during abnormal times, such as when a large-scale generator falls off the power system, and therefore cannot estimate the system inertia of a power system under normal conditions. Furthermore, there may be cases where estimation of the inertia ratio of a power system is useful, not limited to estimation of the system inertia of the power system.

[0009] The present invention has been made in consideration of the above circumstances, and has an objective of providing an information processing device, an information processing method, and a program that can estimate values ​​related to the inertia of a power system not only during abnormal conditions, but also during normal conditions. [Means for solving the problem]

[0010] As an example configuration, a control unit is provided that estimates an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system. the control unit estimates the inertia ratio from the system matrix based on a part of the identified system matrix and a general characteristic of a power flow sensitivity matrix constituting the system matrix, the general characteristic being that the sum of each column of the power flow sensitivity matrix is ​​0; It is an information processing device. As one configuration example, the information processing device includes a control unit that estimates an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, and the control unit estimates the inertia ratio from an identified part of the system matrix based on general characteristics of a power flow sensitivity matrix that constitutes the system matrix, and the general characteristics of the power flow sensitivity matrix are that the power flow sensitivity matrix is ​​a symmetric matrix. As one configuration example, the information processing device includes a control unit that estimates an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, the control unit identifies the system matrix using a record of at least one of a constant phase angle and rotational speed over a certain period of time, the control unit uses, as the phase angle and the rotational speed, at least one of the phase angle and rotational speed of an individual generator in the AC interconnection system, or a representative phase angle and frequency of a partial system including a plurality of generators, and the control unit excludes, in identifying the system matrix, one or more measurement values ​​that are highly dependent on each other based on a numerical index value for the purpose of improving estimation accuracy.

[0011] As an example of the configuration, the inertia ratio is estimated based on the numerically obtained system matrix of the oscillation equation of the AC interconnection system. An information processing method , the inertia ratio is estimated from the system matrix based on a part of the identified system matrix and a general characteristic of a power flow sensitivity matrix constituting the system matrix, the general characteristic being that the sum of each column of the power flow sensitivity matrix is ​​0; It is an information processing method. One configuration example is an information processing method for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnected system, in which the inertia ratio is estimated from an identified part of the system matrix and a general characteristic of a power flow sensitivity matrix that constitutes the system matrix, and the general characteristic of the power flow sensitivity matrix is ​​that the power flow sensitivity matrix is ​​a symmetric matrix. As one configuration example, there is provided an information processing method for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, in which the system matrix is ​​identified using a record of at least one of a constant phase angle and rotational speed over a certain period of time, and at least one of the phase angle and rotational speed of an individual generator in the AC interconnection system or a representative phase angle and frequency of a partial system including a plurality of generators is used as the phase angle and the rotational speed, and in identifying the system matrix, one or more measurement values ​​that are highly dependent on each other are excluded based on a numerical index value in order to improve the estimation accuracy.

[0012] In one embodiment, the method includes causing the computer to execute a step of estimating the inertia ratio based on a numerically obtained system matrix of an oscillation equation of the AC interconnected system. a program for estimating the inertia ratio from the system matrix based on a part of the identified system matrix and a general characteristic of a power flow sensitivity matrix constituting the part, the general characteristic being that the sum of each column of the power flow sensitivity matrix is ​​0; It is a program. As an example configuration, there is provided a program for causing a computer to execute a step of estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, the program estimating the inertia ratio from an identified part of the system matrix based on general characteristics of the power flow sensitivity matrix that constitutes it, the general characteristics of the power flow sensitivity matrix being that the power flow sensitivity matrix is ​​a symmetric matrix. As one configuration example, there is a program for causing a computer to execute a step of estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, the program identifying the system matrix using records of at least one of a constant phase angle and rotational speed over a certain period of time, and using, as the phase angle and the rotational speed, at least one of the phase angle and rotational speed of an individual generator in the AC interconnection system, or a representative phase angle and frequency of a partial system including a plurality of generators, and in identifying the system matrix, excluding one or more measurement values ​​that are highly dependent on each other based on a numerical index value in order to improve the estimation accuracy. Effect of the Invention

[0013] According to the information processing device, the information processing method, and the program of the present invention, it is possible to estimate not only a value related to the inertia of the power system during an abnormality, but also a value related to the inertia of the power system during a normal state. [Brief description of the drawings]

[0014] [Figure 1] 1 is a diagram showing a schematic configuration of an information processing system according to an embodiment of the present invention; [Diagram 2] FIG. 2 illustrates an example of a hardware configuration of an information processing device. [Diagram 3] FIG. 2 is a diagram illustrating an example of a functional configuration of an information processing device. [Figure 4] 2 is a diagram showing an outline of inertia ratio estimation according to the first embodiment of the present invention; FIG. [Diagram 5] FIG. 11 is a diagram showing an outline of inertia estimation according to a second embodiment of the present invention. [Figure 6] FIG. 2 is a diagram showing an example of a verification model according to the first and second embodiments of the present invention. [Figure 7] FIG. 4 is a diagram showing a verification result of inertia ratio estimation (HPF-1) according to the first embodiment of the present invention. [Figure 8] FIG. 11 is a diagram showing a verification result of inertia estimation (HPF-2) according to the second embodiment of the present invention. [Figure 9] FIG. 11 is a diagram illustrating an example of a temporal change in the system frequency at a representative point before and after an event occurrence time. [Figure 10] FIG. 13 is a graph with a fitting function superimposed thereon. [Figure 11] FIG. 13 is a diagram showing an example of a temporal change in power at a representative point before and after an event occurrence time. [Figure 12] FIG. 13 is a graph with a fitting function superimposed thereon. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the following, when the frequency of power measured at a certain point is mentioned, it means the frequency of the voltage measured at that point or the frequency of the current measured at that point.

[0016] In the following, characters in mathematical expressions are modified (e.g., boldface), but when the characters are quoted in a sentence, the modifications are omitted. Except for the modifications, the characters (or character strings) in the mathematical expressions and the characters (or character strings) quoted in the sentence are explained as being the same.

[0017] [Information Processing System] FIG. 1 is a diagram showing a schematic configuration of an information processing system 1 according to an embodiment of the present invention. The information processing system 1 estimates a value related to the system inertia of the AC interconnection system. The value related to the system inertia of the AC interconnection system may be, for example, the value of the system inertia itself or the value of the inertia ratio.

[0018] Here, the AC interconnection system refers to an AC power system that includes one or more generators and generates power synchronously using the one or more generators. For example, the AC power system in the area of ​​Japan consisting of Chubu, Hokuriku, Kansai, Chugoku, Shikoku, and Kyushu (sometimes called the Central and Western system) is an example of an AC interconnection system. The system inertia of an AC interconnection system refers to the sum of the inertia moments of the rotors (i.e., rotors, turbines, etc. of the generators) of one or more generators included in the AC interconnection system. In the following, as an example, a case will be described in which the information processing system 1 estimates a value related to the system inertia of the AC interconnection system R shown in FIG. 1.

[0019] The AC interconnection system R includes z (z is an integer of 2 or more) generators, generators PP1 to PPz, as one or more generators. In the AC interconnection system R, synchronous power generation is performed by these z generators PP1 to PPz. Also, in the AC interconnection system R, the rotors of these z generators PP1 to PPz all rotate synchronously to generate one system frequency. 1, four or more generators PP1 to PPz are illustrated, but z may be any value of 1, 2, or 3. Also, z may be any value of 4 or more.

[0020] In the following description, for convenience of explanation, the system frequency generated by each of the rotating bodies of these z generators PP1 to PPz will be simply referred to as the system frequency. Here, the system frequency is the frequency of rotation of a rotor, but it is known that it is practically equivalent to the frequency of power measured at a point in the AC interconnection system R where the system frequency is measured. For this reason, the system frequency is measured by measuring the frequency of the power. In other words, the system frequency is an example of the frequency of power at each time at a predetermined point in the AC interconnection system. In addition to the z generators PP1 to PPz, the AC interconnection system R includes other facilities such as substations. However, in order to simplify the drawing, the other facilities are not shown in FIG.

[0021] Each of the generators PP1 to PPz is a generator that supplies electric power by, for example, thermal power generation, hydroelectric power generation, nuclear power generation, solar power generation, geothermal power generation, tidal power generation, etc. Some or all of the generators PP1 to PPz may be generators that supply electric power by generating the same type of electric power, or may be generators that supply electric power by generating different types of electric power.

[0022] The information processing system 1 includes a measurement device 10, z measurement devices PQ1 to PQz, and an information processing device 20. The z measuring devices PQ1 to PQz are provided in the z generators PP1 to PPz, respectively. The measurement device 10 and the measurement devices PQ1 to PQz are each connected to the information processing device 20 wirelessly or via wires so as to be able to communicate with each other.

[0023] Here, for example, the measurement results by the measurement device 10 are used when the RoCoF method is used. Therefore, when the RoCoF method is not used in the information processing system 1, the measurement device 10 does not need to be provided. However, in this embodiment, the measurement device 10 will also be described as one configuration example of the information processing system 1. If the information processing system 1 does not include the measuring device 10, the information processing device 20 and the like will not process the measurement results obtained by the measuring device 10.

[0024] The measurement device 10 is provided at a predetermined representative point in the AC interconnection system R. The representative point may be any point among points included in the AC interconnection system R, as long as the system frequency can be measured. Here, the system frequency is measured as the frequency of power measured at a point among points included in the AC interconnection system R where the system frequency is measured. Therefore, the representative point can be said to be a point at which the power supplied by the AC interconnection system R can be measured. For example, the representative point is a substation or the like included in the AC interconnection system R. In the following description, for convenience of explanation, the physical quantity X measured by the measuring device 10 at the representative point will be referred to as the physical quantity X at the representative point. The physical quantity X is, for example, a system frequency, a power, a voltage, a current, a phase angle, etc.

[0025] The measurement device 10 measures the system frequency at each time of the representative point and a value corresponding to the power at each time of the representative point (for example, a voltage, a current, a phase angle, and the power itself, etc.). The measurement device 10 is, for example, a PMU (Phasor Measurement Unit).

[0026] Generally, a PMU is a standardized, inexpensive measurement device that measures (records and transmits) phasor (voltage and phase angle) and frequency at each point, and optionally current and power, with a GPS-based time stamp. This makes it possible to centrally monitor and analyze the time-synchronized records of frequency, phase angle, and interconnection current at each point.

[0027] Each of the measuring devices PQ1 to PQz measures the phase and frequency of each of the generators PP1 to PPz. Each of the measurement devices PQ1 to PQz is, for example, a PMU.

[0028] The information processing device 20 acquires information measured by the measurement device 10 at each time from the measurement device 10. The information processing device 20 stores the acquired information at each time. The information processing device 20 acquires information measured by each of the measuring devices PQ1 to PQz at each time from each of the measuring devices PQ1 to PQz The information processing device 20 stores the acquired information at each time.

[0029] The information processing device 20 is, for example, a workstation, a desktop PC (Personal Computer), a notebook PC, etc. Alternatively, the information processing device 20 may be a tablet PC, a multi-function mobile phone terminal (smartphone), a mobile phone terminal, a PDA (Personal Digital Assistant), etc.

[0030] [Information processing device] FIG. 2 is a diagram illustrating an example of a hardware configuration of the information processing device 20. As shown in FIG. The information processing device 20 includes, for example, a processor 21, a storage unit 22, an input receiving unit 23, a communication unit 24, and a display unit 25. These components are connected to each other via a bus so that they can communicate with each other. Furthermore, the information processing device 20 communicates with the measurement device 10 and the measurement devices PQ1 to PQz via the communication unit 24.

[0031] The processor 21 controls the entire information processing device 20. The processor 21 is, for example, a CPU (Central Processing Unit). The functional units of the processor 21 may be configured using an FPGA (Field-Programmable Gate Array) or the like. The processor 21 executes various programs stored in the storage unit 22.

[0032] The storage unit 22 includes, for example, a hard disk drive (HDD), a solid state drive (SSD), an electrically erasable programmable read-only memory (EEPROM), a read-only memory (ROM), a random access memory (RAM), etc. Note that the storage unit 22 may be an external storage device connected via a digital input / output port such as a universal serial bus (USB) instead of being built into the information processing device 20. The storage unit 22 stores various information, various images, various programs, etc. that are processed by the information processing device 20.

[0033] The input receiving unit 23 is an input device such as a keyboard, a mouse, a touch pad, etc. The input receiving unit 23 may be a touch panel that is integrated with the display unit 25.

[0034] The communication unit 24 includes, for example, an antenna, a digital input / output port such as a USB, an Ethernet (registered trademark) port, and the like.

[0035] The display unit 25 is a display device including, for example, a liquid crystal display panel or an organic EL (Electro Luminescence) display panel.

[0036] FIG. 3 is a diagram illustrating an example of a functional configuration of the information processing device 20. As shown in FIG. The information processing device 20 includes a storage unit 22, an input receiving unit 23, a communication unit 24, a display unit 25, and a control unit 26.

[0037] The control unit 26 controls the entire information processing device 20. The control unit 26 includes, for example, an acquisition unit 261, an estimation unit 262, a display control unit 263, and a storage control unit 264. These functional units included in the control unit 26 are realized, for example, by the processor 21 executing various programs stored in the storage unit 22. In addition, some or all of the functional units may be hardware functional units such as LSI (Large Scale Integration) and ASIC (Application Specific Integrated Circuit).

[0038] The acquiring unit 261 acquires various types of information. For example, the acquiring unit 261 acquires information from the measurement device 10 and the measurement devices PQ1 to PQz.

[0039] The estimation unit 262 performs various estimations. For example, the estimation unit 262 estimates a value related to the inertia of the AC interconnection system R.

[0040] The display control unit 263 generates various images. The display control unit 263 causes the display unit 25 to display the generated images.

[0041] The storage control unit 264 causes the storage unit 22 to store the various pieces of information acquired by the acquisition unit 261 .

[0042] [Definition of system inertia and its estimation] Generally, all the rotating bodies of synchronous machines in an interconnected system rotate synchronously, producing a single frequency. Therefore, by collectively capturing the kinetic energy of these rotating bodies, the inertia of the entire system can be considered. Here, inertia is estimated as kinetic energy, rather than the moment of inertia or unit inertia constant. For example, if there is only one rotating body in a system, its kinetic energy E (units: [J] or [WS]) is expressed by equation (1).

[0043]

number

[0044] Here, J is the moment of inertia of the rotating body [kgm 2 ], and ω is the rotational speed [rad / s]. The frequency of the system is controlled to be roughly constant, and the change in ω is generally less than 1%, even during large-scale disturbances. Therefore, the change in kinetic energy is also limited to that extent. Therefore, in power systems, estimating the kinetic energy of a rotating body and estimating the moment of inertia J are roughly synonymous, although the numerical values ​​involved are different. In the general case where there are multiple rotating bodies, the kinetic energy of the entire system is the simple sum of the kinetic energies of each of them. In the following, the kinetic energy E of the entire system (at the nominal frequency) will be sys For convenience, [Ws] is called system inertia, and its value is the subject of estimation.

[0045]

number

[0046] Put ω back into the variable in equation (2) and sys Differentiate both sides with respect to t, using as a variable. dE sys / dt=ΔP, ΔP=P m -P e By referring to the following relationship, we obtain equation (3).

[0047]

number

[0048] P m is the mechanical input to the rotor, and P e is the electrical output, and ΔP is the difference between them. Equation (3) is the equation of motion of a rotating system (strictly speaking, the equation of motion is obtained by dividing both sides by ω, but it is equivalent). Eliminating J from these equations gives us equation (4).

[0049]

number

[0050] As mentioned above, the change in the rotation speed ω is small, so the factor acting as the formal coefficient in the formula is the nominal rotation speed ω s Next, if we assume that equation (5) holds true, that is, that the rotation speed is equivalent to the voltage frequency, we obtain equation (6) as the system-wide oscillation equation described using frequency.

[0051]

number

[0052]

number

[0053] [Example of HPF method] An example of an HPF method will be shown below, which we will call HPF method 0. HPF method 0 is the machine input P m While the RoCoF method avoids the effects of changes in by limiting the time period under consideration in the time domain, this method attempts to remove the effects of changes in by using an HPF (high pass filter). In order to estimate the inertia, a disturbance is necessary. In principle, HPF method 0 uses constant load fluctuation as a disturbance.

[0054] The basic mathematical model of HPF method 0 will be explained. The basis is equation (6), which is the system-wide oscillation equation. First, perform a Laplace transform on both sides (but omit the initial value term), and then apply the HPF to both sides.

[0055]

number

[0056] This gives us equation (8).

[0057]

number

[0058] First, if the change in mechanical input is sufficiently slow compared to the change in frequency and active power, then by selecting an appropriate HPF, we can obtain equation (9), which then becomes equation (10).

[0059]

number

[0060]

number

[0061] Regarding frequency and electrical output, only the high-speed changing parts are extracted by applying the HPF, so by writing these as equations (11) and (12), equation (13) is obtained.

[0062]

number

[0063]

number

[0064]

number

[0065] Converting this back to the time domain gives us equation (14).

[0066]

number

[0067] The goal is b=f s / 2E sysKnown methods include using the differential value on the left side of equation (14) output from the PMU as is, or numerically differentiating the frequency output from the PMU, to estimate the proportionality constant b, and treating equation (14) as a state equation to configure an extended Kalman filter and estimate the coefficient b.

[0068] (First embodiment) A first embodiment of the present invention will be described. The HPF technique according to this embodiment will be described below, referred to as HPF-1.

[0069] [Estimation of inertia ratio using HPF] FIG. 4 is a diagram showing an outline of inertia ratio estimation according to the first embodiment of the present invention. The information processing device 20 estimates the inertia ratios of the multiple generators PP1-PPz based on the measurement results of the phase angles and frequencies (or, instead of the frequencies, the rotational speed, etc.) of the generators PP1-PPz by the measuring devices PQ1-PQz of the generators PP1-PPz. The information processing device 20 observes (measures) the phase angles and frequencies of the generators PP1-PPz, performs system identification of the equations of motion from their constant fluctuations, and estimates the mutual inertia ratios of the generators PP1-PPz. Here, for example, information (recorded values) for a certain period of time is used as the phase angles and frequencies. At this time, the information processing device 20 removes the effect of frequency feedback control devices (e.g., speed governors, LFCs) that are provided for maintaining the frequency of the power system by HPF for the phase angles and frequencies of each of the generators PP1 to PPz for all of the generators PP1 to PPz. Note that the purpose of these control devices is to maintain the frequency as described below, so the control results are called effects, but from the standpoint of inertia estimation, these control devices should be called influences because they are factors that make estimation difficult.

[0070] The governor is a control device for keeping the rotational speed of a rotating machine constant. Also, LFC (Load Frequency Control) is a load frequency control. Moreover, constant fluctuations are, for example, when there are no special events and the load devices at the consumer are turned on and off. Demand is always fluctuating because it can be changed at the discretion of the consumer. In this embodiment, it is assumed that the demand fluctuation follows a normal distribution (small fluctuations occur frequently, and large fluctuations occur rarely). Δp m Because it is the power of the steam trying to turn the turbine or the power of the water flow trying to turn the water wheel, it cannot be measured directly and changes slowly.

[0071] Here, in Figure 4, the bold lowercase letters are vectors whose number of elements is equal to the number of generators. The inter-vector sensitivity is a matrix. In Figure 4, the bold line path corresponds to a vector whose number of elements is equal to the number of generators. The information processing device 20 performs least-squares estimation of the matrix L in the equation (16) as the sensitivity in the equation (15) from the constant fluctuation.

[0072]

number

[0073]

number

[0074] In equation (16), K is the synchronizing force matrix (or the tidal current sensitivity matrix), and the element ratio (inertia ratio) of J is estimated from the general condition of K. Here, the general condition for K is, for example, at least one of a column sum 0 condition (a condition that the sum of each column is 0) and a symmetry condition (a condition that the matrix is ​​symmetric). Which of these two conditions (properties) of the matrix K to use is selected, for example, by a user. J is the moment of inertia, and J plus ω0 2 Multiply it by / 2 and you get energy.

[0075] Alternatively, the steady-state stability may be estimated by calculating eigenvalues ​​from the estimated series of matrices. A method for estimating eigenvalues ​​from constant fluctuations is known (see Non-Patent Document 3).

[0076] <Condition for matrix K: "column sum = 0"> The power flow equation for the small fluctuation component (high frequency component) is given by equation (17). Δp e is the electrical output of each of the generators PP1 to PPz, and Δδ is the phase angle of each of the generators PP1 to PPz.

[0077]

number

[0078] Δp e The sum of the elements (total fluctuation in generated power) is 0. Here, we consider the case where the load fluctuation is zero and there is only an initial disturbance. In other words, if you multiply equation (17) by a row vector of 1s from the left, both sides become 0.

[0079]

number

[0080] In order for equation (18) to always hold for various changes in Δδ due to the initial disturbance, equation (19) holds.

[0081]

number

[0082] Equation (19) shows that each column sum of K is zero.

[0083] <Condition for matrix K: "Symmetry of K"> When considering two generators A and B, the following two quantities are equal, so K is symmetric. The first quantity is the change in tidal flow from generator B to generator A when only the phase angle of generator A changes slightly, and the second quantity is the change in tidal flow from generator A to generator B when only the phase angle of generator B changes slightly (the same amount as above). Therefore, a method can be used to estimate the inertia ratio by numerically indexing the degree of symmetry of the tidal current sensitivity matrix K and optimizing the index value.

[0084] <Asymmetry of matrix L> In general, L is asymmetric because, under the same conditions, the following two quantities differ by the inertia difference: Note that angular acceleration is the electric torque (∝Δδ) divided by the inertia. The first quantity is the angular acceleration dω / dt of generator B when only the phase angle of generator A changes small, and the second quantity is the angular acceleration dω / dt of generator A when only the phase angle of generator B changes small (the same amount as above).

[0085] It is possible to formulate the problem of finding J such that equation (20) is as symmetric as possible.

number

[0086] In Non-Patent Document 3, the oscillation mode (eigenvalue) of a multi-machine system (a general system with many generators) is estimated from only the observation of the phase angle and rotation speed at all times. Technically, this is equivalent to identifying the system matrix (the A matrix in the state space representation of a linear system dx / dt=Ax). However, it has not been possible to separate the value related to inertia (inertia ratio in this embodiment) from the contents of the system matrix. In this embodiment, this separation is made possible by focusing on the properties of matrix K. As mentioned above, there are two "properties of matrix K (conditions for matrix K)" that we focused on, and either one may be used. It should be noted that, by using the HPF-1 according to this embodiment, it is also possible to obtain system eigenvalues, similarly to Non-Patent Document 3.

[0087] FIG. 4 shows a power system 211, a PMU 212, HPFs 213 and 214, and an estimation calculation unit 215. The power system 211 is an example of an AC interconnection system R. The PMU 212 is an example of each of the measuring devices PQ1 to PQz. Although one PMU 212 is shown in Fig. 4, a PMU is provided for each of the generators PP1 to PPz. The functions of the HPFs 213 and 214 and the function of the estimation calculation unit 215 are an example of the function of the estimation unit 262 .

[0088] The estimation calculation section 215 estimates the inertia ratio using equation (16) or equation (20) based on the result of Δω passing through the HPF 213 and the result of Δδ passing through the HPF 214 .

[0089] [Details of inertia ratio estimation] The estimation of the inertia ratio (ratio of inertia constants) will be explained in detail below. n G Assuming that the phase angle and frequency of each generator can be measured, the inertia ratio between the generators can be estimated from their constant fluctuations.

[0090] <Formulation of the oscillation equation> First, the generator is n G This paper describes a procedure for identifying the oscillation equation (linear domain) of a system with a stand and estimating the inertia ratio based on the equation. The generator is n G Let us assume that there are i (i=1~n G The equation of motion for the rotor of the )th generator is equation (21) (both sides are in [W]).

[0091]

number

[0092] In equation (21), D i is the damping specific to the machine, such as mechanical damping. iSince the change in [rad / s] is small, some approximations are made. On the other hand, if we assume that the voltage behind the generator is constant, the characteristics of the circuit network, including the internal impedance of the generator, for small changes in various quantities are expressed by equations (22) and (23).

[0093]

number

[0094]

number

[0095] n G ×n G The matrix K[w / rad] is a constant matrix based on a linear approximation of the power flow equation. i is δ i is a small change in ΔP e,i P e,i is an infinitesimal change in. In other words, the network is assumed to be contracted so that the number of nodes matches the number of generators. Generally, when a system configuration including an intermediate bus is modeled directly using the power flow equation, K becomes a sparse matrix. However, since the voltage and phase angle of the intermediate bus are quantities that depend on the quantities of the generators, the quantities of the intermediate bus can be eliminated without losing the physical characteristics of the object. Since K defined above is contracted by this procedure, it becomes a dense matrix. Δδ in equation (23) i is the phase angle δ determined by equation (24) i This is a small change of [rad].

[0096]

number

[0097] To summarise the above, we can obtain an equation for the oscillations of a multi-machine system that describes mechanical input as an external factor for small changes.

[0098]

number

[0099]

number

[0100]

number

[0101]

number

[0102] In equations (25) to (28), Δω is i A vector of infinitesimal changes in Δp m P m.i is a vector that lists the infinitesimal changes in , I is the unit matrix, 0 is the zero matrix, and diag() is a diagonal matrix with the elements or vectors in parentheses as diagonal elements.

[0103] <Derivation of the oscillation equation for high frequency components> Here, we derive the oscillation equations related only to the high frequency components of various quantities. By replacing d / dt in equation (25) with the Laplace operator s and multiplying both sides by the transfer function HPF(s) of the high pass filter (HPF), we obtain equation (29).

[0104]

number

[0105] Now, by adding the subscript H to the amount after passing through the HPF, we obtain equation (30).

[0106]

number

[0107] The Laplace operator s is reverted to the differential operator d / dt. In other words, even if we focus only on the high-frequency components, the high-frequency quantities satisfy the characteristics of the original generator and system. Therefore, it is possible to identify the oscillation equation from only the high-frequency components of the quantities. Generally, the change in mechanical input is slower than other quantities, so Δp m,H If we ignore the second term on the right-hand side, which includes

[0108]

number

[0109] Equation (31) will be explained as the perturbation equation for high frequency components. In reality, the load fluctuation is input to equation (31) and acts as an external force, resulting in equation (32).

[0110]

number

[0111] The way load fluctuations work is based on the oscillation equations in which the mechanical input is explicitly written. However, since the load is mainly present at the eliminated intermediate bus, the matrix B that determines the distribution to the generator is formally added. Since only high-frequency components are considered, Δp l,H is essentially random and, in principle, no measurements are performed. It is driven by random inputs and has a state Δω H and Δδ H changes. Δp l,H Even if is not measured, if it is random around 0, it may be possible to identify the system matrix etc. from the measured state quantity. From now on, we will assume such a property for the load fluctuation. The description of the load fluctuation input in the state equation will be omitted.

[0112] Equations (33) to (36) form part of this perturbation equation. Λ is the upper half of the system matrix shown in equation (35).

[0113]

number

[0114]

number

[0115]

number

[0116]

number

[0117] Based on equation (33), we aim to identify the matrix Λ using the high-frequency components of the differential of the generator's rotation speed and the measured values ​​of the rotation speed and phase angle. If Λ can be identified, the entire oscillation equation can be known because the lower half of the system matrix is ​​a combination of unit matrices and zero matrices. Then, for each matrix included in Λ, L is asymmetric, K is symmetric, and J and J are asymmetric. -1 Since we know the characteristics of J, such as that it is a diagonal matrix, it may be possible to extract some characteristics of J from the identified L.

[0118] The estimation framework described above is shown in FIG. 4 (framework for inertia ratio estimation based on system identification).

[0119] As described above, in the information processing system 1 according to this embodiment, the information processing device 20 can estimate not only a value related to the inertia of the power system during an abnormality (in this embodiment, the inertia ratio), but also a value related to the inertia of the power system during normal times (in this embodiment, the inertia ratio).

[0120] Here, in this embodiment, a technique using an HPF has been described, but as another example, a band pass filter (BPF) may be used instead of the HPF. Furthermore, in this embodiment, the case where both the phase angle and the frequency are used has been described, but as another example, either the phase angle or the frequency may be used. In addition, in this embodiment, a case has been described in which a filter (HPF or BPF) is applied to both the phase angle and the frequency. However, as another example, a filter (HPF or BPF) may be applied to either the phase angle or the frequency. Although a method using a filter (for example, an HPF or a BPF) has been described above, as another configuration example, a configuration may be implemented in which similar processing results are obtained without using a filter.

[0121] In addition, instead of the phase angle and rotation speed of individual generators PP1 to PPz in the AC interconnection system R, a representative phase angle and frequency of a partial system including multiple generators may be used as the phase angle and frequency (or rotation speed). In addition, in identifying the system matrix, in order to improve the estimation accuracy, one or more measurement values ​​such as frequencies that are similar to each other and highly dependent on each other may be excluded based on a numerical index value.

[0122] Second embodiment In this embodiment, differences from the first embodiment will be described in detail, and descriptions of similarities to the first embodiment will be omitted or simplified. The HPF technique according to this embodiment will be described below, and will be referred to as HPF-2.

[0123] [Inertia estimation using HPF] FIG. 5 is a diagram showing an outline of inertia estimation according to the second embodiment of the present invention. The information processing device 20 estimates the individual inertia of each of the multiple generators PP1 to PPz based on the measurement results of the phase angle and frequency of each of the generators PP1 to PPz by the measurement devices PQ1 to PQz for each of the generators PP1 to PPz, and the measurement results of the interconnection line power flow by the measurement device 10 of the AC interconnection system R. At this time, the information processing device 20 removes the effect of frequency feedback control (for example, speed governor, LFC) by the HPF for each phase angle of the generators PP1 to PPz and the interconnection line power flow for all the generators PP1 to PPz.

[0124] FIG. 5 shows area 1 and area 2 of the power system, PMU 312, HPFs 313 and 314, and estimation calculation unit 315. The electric power system is an example of an AC interconnected system R. The PMU 312 is an example of the measurement device 10 and each of the measurement devices PQ1 to PQz. Although one PMU 312 is shown in Fig. 5, a PMU (an example of the measurement device 10) is provided in the AC interconnection system, and a PMU (an example of the measurement devices PQ1 to PQz) is provided for each of the generators PP1 to PPz. The functions of the HPFs 313 and 314 and the function of the estimation calculation unit 315 are an example of the function of the estimating unit 262 . In this embodiment, only in FIG. 5, the Σ symbol represents the sum of the elements of the input vector.

[0125] Here, in Fig. 5, the bold lowercase letters are vectors whose number of elements is equal to the number of generators. In Fig. 5, the bold line paths correspond to vectors whose number of elements is equal to the number of generators.

[0126] The information processing device 20 performs system identification based on the phase angle and frequency of each of the generators PP1 to PPz and the constant fluctuation of the interconnection line power flow. The information processing device 20 obtains partial information on the inertia of each of the generators PP1 to PPz (such as the difference between the area inertia total values) and estimates the inertia of each of the generators PP1 to PPz in conjunction with the inertia ratio information obtained by, for example, HPF-1. Here, the information processing device 20 may estimate the inertia of each of the generators PP1 to PPz using inertia ratio information obtained by a method other than HPF-1.

[0127] For example, we will use equation (37), which shows the interconnection line power flow of a two-area system.

[0128]

number

[0129] As the sensitivity of equation (38), vector k 12 T Identify using least squares (partial information for K).

[0130]

number

[0131] The information processing device 20 estimates partial information of J from partial information of L and K (for example, only the sum of two unknown inertias) and acquires predetermined information (for convenience of explanation, referred to as information a1). The information processing device 20 also utilizes information on the ratio of elements of J obtained by a method such as HPF-1 (for convenience of explanation, referred to as information a2). The information processing device 20 performs least-squares estimation of J that maximizes compatibility between the information a1 and the information a2.

[0132] In this embodiment, partial information on the magnitude of inertia is obtained from observations of the interconnection line current and phase angle. In addition, by combining this with inertia ratio information obtained by the HPF-1 or the like, the inertia itself can be estimated. In this embodiment, a method may be used to estimate the inertia of each of the generators PP1 to PPz by summarizing the degree to which all conditions for estimating the inertia of each of the generators PP1 to PPz are satisfied into a single numerical index and optimizing the index value.

[0133] [Inertia Estimation Details] The estimation of inertia will now be described in detail. In inertia estimation using interconnection line power flow information, the inertia of all generators is estimated assuming that at least one interconnection line power flow can be measured. We will explain inertia estimation when interconnection line power flow information is available in addition to generator measurement information.

[0134] <Formulation of the relationship between interconnection line power flow and generator state> Here, we explain a method for estimating inertia using power flow information. The oscillation equation in equation (31) is intended for the entire AC-interconnected system, but we will assume that this consists of two areas and that the interconnection line power flow can be measured. First, the power flow equation for the entire system contained in the matrix L part of equation (31) is equation (39).

[0135]

number

[0136] If we divide this into two areas, we get equation (40).

[0137]

number

[0138] In addition, the number of generators in the two areas is generally different, so 12 and K. 21 is not necessarily a square matrix. Furthermore, for example, the case where there is only one generator in area 1 is also included in this formulation. The HPF-passing component (high-frequency component) ΔP of the current flowing from area 1 to area 2 H12 If we consider this to be the sum of the same components of the power generation output within area 1, we obtain equation (41).

[0139]

number

[0140] Here, 1 T 1, 0 T 2, 1 T is a row vector consisting of an appropriate number of 1s or 0s (the number of generators in the area with each subscript being a number, or the number of generators in the entire system). Area 1 is defined by the row vector in equation (42).

[0141]

number

[0142] Similarly, the reverse power flow ΔP H21 can also be obtained using equation (43).

[0143]

number

[0144] Here, 0 T 1, 1 T 2 is a row vector in which 1 or 0 is arranged for each generator in each area. Also, it is expressed as (44).

[0145]

number

[0146] Here, the current ΔP H12 and the reverse current ΔP H21 Since the relationship in equation (45) is satisfied, we obtain equation (46).

[0147]

number

[0148]

number

[0149] The purpose is to utilize tidal current information for estimation. ΔP H12 and Δδ H From the measured values ​​of the two, the static proportional relationship between them (information related to the tidal current sensitivity) k T 12 It is believed that it is possible to identify

[0150]

number

[0151] k T 12is Δδ H It is a row vector (horizontally long vector) with the same number of elements as in (1). Hereinafter, this vector will be referred to as the power flow sensitivity vector. In this embodiment, one power flow sensitivity vector is identified using information (recorded values) of at least one of the constant phase angle and rotational speed, and at least one of the power value or an equivalent measurement value, for a certain period of time. Comparing the coefficients with the already calculated relational expression (41), we obtain equation (48).

[0152]

number

[0153] Similarly, when comparing the coefficients with the relational expression (46), we obtain equation (49).

[0154]

number

[0155] Here, the information obtained is a single k T 12 However, when we focus on the generator output in area 1 and when we focus on the generator output in area 2, the information obtained can be applied to different purposes. T 10 is a row vector that extracts and sums the upper elements of K, and e T 20 is a row vector that extracts and sums the lower elements, and it turns out that we are getting something like a sum of different parts of K.

[0156] Here, by using the relationship in equation (50), equations (51) and (52) are obtained.

[0157]

number

[0158]

number

[0159]

number

[0160] If the submatrix L of the system matrix can be identified in advance, then J (a diagonal matrix with inertia constants arranged as elements) is the unknown. Considering equations (51) and (52) from the perspective of finding the unknowns, the first equation is a condition equation for the upper elements of J, and the second equation is a condition equation for the lower elements of J. For this reason, from the perspective of ultimately finding all the elements of J, it is considered that there is no overlap of information in equations (51) and (52). Therefore, by adding up equations (51) and (52) side by side, equation (53) is obtained. Equations (55) and (56) are obtained from equation (54).

[0161]

number

[0162]

number

[0163]

number

[0164]

number

[0165] Here, the derivation utilizes the property that diagonal matrices are interchangeable in matrix multiplication. v is a vector obtained by arranging the diagonal elements of the diagonal matrix J. Equation (57) can be obtained by transposing both sides of equation (55).

[0166]

number

[0167] Here, we use the property that a diagonal matrix does not change when transposed. Equation (57) expresses the unknowns as j v The simultaneous equations are as follows: T is unique, so this alone is not enough. v However, unlike the "symmetry condition", equation (57) is at least formally v It has information that allows it to determine the size of the elements.

[0168] The estimation framework described above is shown in Fig. 5 (framework for inertia estimation using interconnection line power flow information).

[0169] For example, matrices L and k are calculated based on the observed phase angle, rotation speed, and power flow of the interconnection line. 12 After identifying the above, one or more equations related to the inertia ratio and one or more equations related to the magnitude of inertia are selected, and the least squares method is applied to these two or more groups of relational equations to find the inertia constant that makes all conditional equations hold as well as possible.The reason we say "group of equations related to the magnitude of inertia" here is because if we can measure a larger number of interconnection line flows (for example, when three areas are connected in a skewer shape, there will be two interconnection line flows), multiple equations will be obtained instead of just one.

[0170] As described above, in the information processing system 1 according to this embodiment, the information processing device 20 can estimate not only a value related to the inertia of the power system during an abnormality (in this embodiment, inertia), but also a value related to the inertia of the power system during normal times (in this embodiment, inertia).

[0171] Here, in this embodiment, a technique using an HPF has been described, but as another example, a band pass filter (BPF) may be used instead of the HPF. Furthermore, in the present embodiment, a case has been described in which both the phase angle and the tie line power flow are used. However, as another example, either the phase angle or the tie line power flow may be used. In addition, in this embodiment, a case has been described in which a filter (HPF or BPF) is applied to both the phase angle and the interconnection line flow. However, as another example, a filter (HPF or BPF) may be applied to either the phase angle or the interconnection line flow. Although a method using a filter (for example, an HPF or a BPF) has been described above, as another configuration example, a configuration may be implemented in which similar processing results are obtained without using a filter.

[0172] In addition, instead of the phase angle and rotation speed of individual generators PP1 to PPz in the AC interconnection system R, a representative phase angle and frequency of a partial system including multiple generators may be used as the phase angle and frequency (or rotation speed). In addition, in identifying the system matrix, one or more measurement values ​​such as frequencies that are highly dependent on each other may be excluded based on a numerical index value in order to improve estimation accuracy.

[0173] (Verification of the first and second embodiments) [Verification of multi-vehicle inertia and inertia ratio estimation method using HPF] An example will be given using three generators. The verification of the first embodiment (HPF-1) and the verification of the second embodiment (HPF-2) will be explained together.

[0174] FIG. 6 is a diagram showing an example of a verification model according to the first and second embodiments of the present invention. In the example of Fig. 6, the stability is evaluated using a stability evaluation model (the Park model is used as the generator model). Load fluctuations independent of each other are applied from the nearest point to each of the generators 431-1 to 431-3. The time interval for the simulation and sampling is set to Δt = 0.005 [s].

[0175] FIG. 6 shows three generators 431-1 to 431-3, PMUs 412-1 to 412-3 provided for each of the generators 431-1 to 431-3, a PMU 421 provided at a representative point, an estimation function 441, and loads L1 and L2.

[0176] Microfluctuations ΔP1 to ΔP3 are applied in the vicinity of the generators 431-1 to 431-3, respectively. The PMUs 412-1 to 412-3 for the generators 431-1 to 431-3 measure δ and ω for each of the generators 431-1 to 431-3. For HPF-1, the estimation function 441 receives the inertia sum value as input. For HPF-2, the PMU421 measures 12 is used for estimation in the estimation function 441.

[0177] Verification will be explained in detail. We verified a method (HPF-1) that estimates the inertia ratio by synchronously measuring the phase angle and rotational speed of each generator and adding numerical differentiation of the rotational speed. We also verified a method (HPF-2) that estimates the inertia of each generator when synchronously measuring the interconnection line power flow.

[0178] The example system consists of three generators 431-1 to 431-3. In terms of area configuration, in Fig. 6, the area near the generator 431-1 on the left side is Area 1, and the rest are Area 2. To vary the load, different random noise (pseudo white noise with an LPF (low pass filter) with a cutoff angular frequency of 0.5 [rad / s] (time constant 2 seconds) applied, with different random number seeds) was applied from the vicinity of each of the generators 431-1 to 431-3. The simulation of the example system was performed using an RMS-based stability evaluation function with a time step of 0.005 seconds. Sampling was also performed at the same intervals according to the calculation step. When identifying part of the oscillation equation using the least squares method, it is considered basic to use the RLS (recursive least squares) method to track changes in the state of the target system, but in this example, the least squares method was performed using all past data at each time point in order to confirm that a solution could be obtained stably.

[0179] FIG. 7 is a diagram showing a verification result of the inertia ratio estimation (HPF-1) according to the first embodiment of the present invention. In the graph shown in FIG. 7, the horizontal axis represents time [s], and the vertical axis represents the inertia [MW·s] of each of generators 431-1 to 431-3. For each of the three generators 431-1 to 431-3, reference characteristics (values ​​of inertia set in the generators on the model, i.e., true values) 1011 to 1013 and estimated characteristics 1021 to 1023 are shown. The reference characteristics and estimated characteristics correspond to each other with the closest characteristics except for the initial period.

[0180] In the example of Figure 7, the least squares method is performed using all past data (up to 1000 in the graph). This is a sequential estimation result from all past information. Since HPF-1 estimates a "ratio," it is an apportioned value with the total inertia value already known. In the example of Figure 7, the constant values ​​in the figure are true values, and the order from lowest to highest is 2020, 3030, and 4040 [MW s]. In the example of FIG. 7, the curves are estimates, which have the same order as the true values, except initially. The final error of the inertial estimation was -17% to +5%. Although it cannot be said that the accuracy was necessarily high, it was possible to estimate with a certain degree of accuracy. The HPF cutoff is 10 [rad / s]. There is no LPF. The inertia ratio condition was that the column sum of K was 0.

[0181] FIG. 8 is a diagram showing a verification result of inertia estimation (HPF-2) according to the second embodiment of the present invention. In the graph shown in FIG. 8, the horizontal axis represents time [s], and the vertical axis represents the inertia [MW·s] of each of generators 431-1 to 431-3. For each of the three generators 431-1 to 431-3, reference characteristics 1111 to 1113 and estimated characteristics 1121 to 1123 are shown. The reference characteristics and estimated characteristics correspond to each other with the closest characteristics except for the initial period.

[0182] In the example of Figure 8, the least squares method is performed using all past data (up to 1000 in the graph). This is a sequential estimation result from all past information. In HPF-2, the "inertia itself" is estimated. In the example of Figure 8, the constant values ​​in the figure are true values, and the order from lowest to highest is 2020, 3030, and 4040 [MW s]. In the example of FIG. 8, the curves are estimates, which have the same order as the true values, except initially. The final error of the inertial estimation was -3% to +25%. Although it cannot be said that the accuracy was necessarily high, it was possible to estimate with a certain degree of accuracy. The HPF cutoff is 10 [rad / s]. There is no LPF. The inertia ratio condition was that the column sum of K was 0.

[0183] In both the verifications of HPF-1 and HPF-2, equation (58) is used as the condition on the inertia ratio.

[0184]

number

[0185] In addition, since the inertia value cannot be estimated in HPF-1, the total inertia value (the true value given in the simulation) is given and plotted. In addition, ω HPF =10[rad / s]. 7 and 8 each show six lines, and as described above, three straight lines are the true values ​​of the inertia of each of the generators 431-1 to 431-3, which are 2020, 3030, and 4040 [MWs] in ascending order. The remaining three curves show the change in the estimation results over time, and are arranged in the same order as the true values ​​after approximately time 100 [s].

[0186] For HPF-1, the error was -17 to +5% when evaluated based on the inertia estimation accuracy of each generator. For HPF-2, the error was -3 to +25%. For HPF-2, the error for the total inertia value was +4%. Although it cannot be said that the accuracy is sufficient, it is possible to estimate the inertia ratio and inertia with a certain degree of accuracy, and it was confirmed that the method is generally appropriate within the range of this numerical example.

[0187] In the results of HPF-1, even if the total inertia is given, the total value of the estimated result may not match the true value. This is because the total inertia is not given as a constraint condition, but as one of the conditional equations considered in the least squares method. In the formulation, L is assumed to be a singular matrix, but in actual estimation, it does not necessarily become a singular matrix due to conditions such as load, various noises, and calculation errors. In such a case, one of the eigenvalues ​​of L, which should be zero, becomes a small non-zero value, and as a result, the conditional equation in the least squares method becomes redundant, and the condition of the total inertia is considered equally to other equations as one of the least squares evaluation targets, making errors unavoidable. Nevertheless, as long as the eigenvalue in question is small, it is considered that the total estimated value of HPF-1 will be roughly close to the true value given.

[0188] (Modifications of the first and second embodiments) <Estimation using substation measurement information> In the above, we have assumed that the generator-related quantities could be measured in the formulation, and therefore we will explain how to estimate using substation data (not generator data) actually measured by a PMU. That is, in the first and second embodiments, the position (or vicinity) of the representative substation may be used instead of the positions (or vicinity) of the generators PP1 to PPz.

[0189] The above explanation is based on the generator oscillation equation. However, it is difficult to directly measure the phase angle of the generator itself, and in reality, there are several hundred generators in a system, so synchronous measurement of all of them using a PMU or similar may not be realistic at present. For this reason, it is possible to use the phase angle and frequency of a limited number of appropriate representative points (which can represent the phase angles of multiple generators) instead of the δ and other values ​​used in the above discussion.

[0190] In this case, it is important that the observation point adequately represents the neighboring generators. It is also important that the information redundancy (similarity of the measured values) is low. The problem of redundancy is considered below. High redundancy can lead to the following problems. As an extreme example, if two pieces of data from the same location are included, the inertia of nearby generators may be counted twice, and there is a possibility that numerical calculations for system identification may be hindered (such as not being able to obtain the inverse matrix). For this reason, it is desirable to remove measurement values ​​that are similar to a certain extent in advance.

[0191] There are several methods for evaluating whether measurement values ​​are similar or independent, but the simplest method is to use the cosine similarity or correlation coefficient within a certain time period as an index. Here, cosine similarity is the cosine value of the angle between two measurement time series, each of which is considered as a vector. The larger the cosine similarity or correlation coefficient, the more similar the data. Both cosine similarity and correlation coefficient are maximum when the two measurement values ​​are identical, and the maximum value is 1. It is believed that it is possible to determine the necessary and sufficient observation points by calculating the cosine similarity between all data, removing the points that are least independent from other data, and examining the effect and influence on the system identification process and estimation results.

[0192] Phase angle Δδ at measurement point 1 H1 and the phase angle Δδ at measurement point 2 H2 Cosine similarity with <Δδ H1 ,Δδ H2> is obtained as in equation (59), where the Σ symbol refers to the time direction.

[0193]

number

[0194] The cosine similarity calculated here is almost equivalent to the correlation coefficient. The correlation coefficient is calculated by subtracting the average value from each quantity, but since the above formula is targeted at high frequencies, the average value is expected to be almost zero. As a result, it can be said that the cosine similarity in this case is equivalent to the correlation coefficient.

[0195] The following procedure can be considered as a procedure for removing measurement information that is similar to other measurement information. First, we calculate the cosine similarity as shown above for all combinations of phase angles, that is, n G (n G Calculate -1) / 2 ways and create a list in order of largest to smallest. Here, i, j, and k are 1 to n G The integer is one of the following: The largest (closest to 1) combination (i,j) is the most similar waveform. Next, decide whether number i or number j should be excluded. Look through the list from the second number onwards to see if i or j appears.

[0196] For example, when the combination (j,k) is first found, the measurement value with number j is most similar to the measurement value with number i, and is also considered to have a certain degree of similarity with number k, so it is removed from the data set. If two or more measurements are to be removed, the above procedure from the list creation onwards is repeated.

[0197] The above describes the issues and countermeasures when there is information redundancy, but conversely, there are also cases where information is insufficient. For example, if the above-mentioned cosine similarity is utilized, by looking at the combinations at the end of the above-mentioned list and their values, if it is found that a specific measurement value is significantly different from other information, it suggests that necessary measurements may be missing. It is possible to deal with such cases of insufficient information.

[0198] Regarding the issues of information redundancy and information insufficiency, we considered considering them based on measured values ​​above. However, another effective measure would be to revise the present method, which was formulated based on measurements at the generator end, to a formulation that assumes measurements at substations. It may be possible to estimate inertia with the highest possible accuracy using currently available measurement equipment, and to achieve the highest possible accuracy with minimal additional measurement when additional measurements are required.

[0199] As mentioned above, in this case, the substation phase angle and frequency may be used instead of the generator phase angle and angular speed based on the generator-based formulation. In this case, the phase angle and frequency must be converted from [deg], [Hz], etc. to [rad], [rad / s] as necessary.

[0200] Third embodiment [Application of HPF-1 to RoCoF] In the following, the three examples of the RoCoF estimation method will be referred to as RoCoF0, RoCoF1, and RoCoF2 for convenience of explanation.

[0201] <rocof0> In RoCoF0, the frequency waveform after an event such as a large-scale power source trip is approximated by a quintic expression or the like (any polynomial will do) to estimate the system inertia. For example, the problem that Non-Patent Document 2 largely solves regarding RoCoF0 is to avoid the influence of the governor response and the like that becomes noticeable several seconds after the event, and to avoid the influence of the oscillation components between generators that are included in the waveform for several seconds after the event, but the following problem remains. In other words, issues remained, such as the dependency of the inertia estimation results on the event time (determined by the analyst) and the insufficient avoidance of the effects of fluctuation components between generators.

[0202] <rocof1> RoCoF1 achieves objective event timing by identifying the event time by considering the pre-event frequency waveform, i.e., the steady-state frequency before the event is also used. In RoCoF1, for example, the pre-event frequency is linearly approximated, and the event time is determined to minimize the squared error over the entire period of consideration, taking into account the continuity condition with a quintic polynomial or the like after the event (any polynomial will do). Although the main purpose of the least squares method is to identify parameters, a square error value is obtained as a by-product. The event time that minimizes this square error value is determined. RoCoF1 may also be applied to power as well as frequency.

[0203] <rocof2> In RoCoF2, it is premised on a combination with RoCoF1, and the event time is determined by utilizing both the frequency waveform and the power change waveform. In RoCoF2, for example, a step change waveform of an appropriate power value (such as the output of a dropped generator, a remaining machine, the total of neighboring remaining machines, etc., or the waveform of power that changes stepwise in neighboring power flows) is evaluated simultaneously with the frequency waveform to determine the event time common to both. In RoCoF2, for each of the frequency f and the power P, least-squares identification is performed to obtain a common event time that minimizes the weighted average of the squared errors. The frequency waveforms used in RoCoF0, RoCoF1, and RoCoF2 described above are representative frequencies in the AC interconnected system R. An example of a representative frequency is the frequency measurement value at a representative point that is determined in advance in the AC interconnected system or determined from some analysis of the event waveform.

[0204] <Application of HPF-1> HPF-1 according to the first embodiment is applicable to RoCoF0, RoCoF1, RoCoF2, or other RoCoF methods. For each of the above RoCoF methods or other RoCoF methods, the inertia ratio information obtained by HPF-1 according to the first embodiment can be applied. Specifically, as will be described later, this inertia ratio information serves as information for determining the representative frequency used in each of the above RoCoF methods.

[0205] Hereinafter, for convenience of explanation, RoCoF to which HPF-1 is applied will be referred to as this RoCoF and described. The method of this RoCoF is a method that, when there is a problem setting of "weights are unknown and events are irregular", when an event occurs, uses data for a certain period in the immediate past to calculate weights by the method of HPF-1 and utilizes them for RoCoF consideration.

[0206] In this RoCoF, when an event occurs, the RoCoF is calculated by taking a weighted average of the frequency waveforms at the time of the event at multiple locations, using the generator inertia ratio (inertia constant ratio) estimated by HPF-1 using recorded values ​​for a certain period of time prior to the event as a weight.

[0207] In this RoCoF, among the measured waveforms in the graphs shown in Figures 9 to 12, a weighted average value of multiple points is used as the frequency waveform shown in Figures 9 and 10. The measured waveform is curve PL1 in the examples of Figures 9 to 10, and curve PL2 in the examples of Figures 11 and 12.

[0208] <Weighted average of multi-point frequencies based on system characteristics just before the event> The weighted average of the frequency waveform will now be described. The ideal weighting for each point frequency is the inertia of the point, which changes from moment to moment. Therefore, the weight w in equation (60) is determined based on the system characteristics obtained from the measured values ​​for a certain period of time (e.g., 30 minutes or 1 hour) before the event. Since it is a weighting coefficient, the total value is already constrained to be 1, so it is sufficient to know the ratio of the inertia constants between points, not the inertia constant itself. w i (i=1~n G ) is the value of the weighting coefficient for the i-th location (e.g., the i-th generator).

[0209]

number

[0210] The ratio of the inertia constants can be obtained, for example, by applying an HPF to the frequency and phase angle at each point and from the sensitivity of various quantities between points. That is, the ratio of the inertia constants can be obtained by the method of the HPF-1 according to the first embodiment. In this case, in the information processing device 20, the estimation unit 262 estimates the ratio of the inertia constants using HPF-1 based on the measurement results obtained from the measuring device 10 and the measuring devices PQ1 to PQz, and uses the estimation result to determine the weighting coefficient and execute the RoCoF method.

[0211] [Details of RoCoF] As described above, the technique of HPF-1 according to the first embodiment can be applied to the conventional RoCoF method, and can also be applied to the novel RoCoF method. The RoCoF method is described in detail below.

[0212] The RoCoF method is known as a method for estimating the system inertia of the AC interconnected system R. The RoCoF method is a method for estimating the RoCoF (i.e., the rate of change of the system frequency) for the system frequency when the system frequency fluctuates, and estimating the system inertia of the AC interconnected system R based on the estimated RoCoF.

[0213] Here, the kinetic energy of the rotor of each of the one or more generators PP1 to PPz included in the AC interconnection system R can be regarded as the moment of inertia of the rotor of each of the one or more generators PP1 to PPz included in the AC interconnection system R. This is because the kinetic energy of the rotor of each of the one or more generators PP1 to PPz included in the AC interconnection system R can be obtained by multiplying the moment of inertia of the rotor by the square of the angular velocity of the rotor (i.e., the square of the frequency of the rotor). Therefore, in the AC interconnection system R, the system inertia of the AC interconnection system R can be calculated as the sum of the kinetic energy of the rotor of each of the one or more generators PP1 to PPz included in the AC interconnection system R. Then, such a sum of the kinetic energy can be related to the RoCoF for the system frequency by an oscillation equation obtained by time-differentiating a formula showing the sum of the kinetic energy. The RoCoF method is a method of using this oscillation equation to obtain the system inertia of the AC interconnection system R based on the RoCoF (more precisely, the sum of the kinetic energy that can be regarded as the system inertia).

[0214] In the AC interconnection system R, the system frequency is usually controlled to always keep a value within a certain range, even though it fluctuates slightly over time due to demand fluctuations. Therefore, in order to obtain the RoCoF used to estimate the system inertia of the AC interconnection system R by the RoCoF method, the system frequency needs to change significantly to deviate from the range. Therefore, in estimating the system inertia of the AC interconnection system R by the RoCoF method, the RoCoF for the system frequency when an event such as a generator being disconnected from the AC interconnection system R or a load being connected to the AC interconnection system is caused is often used. This is because, in such cases, it is known that the output power of the generator that has not been disconnected from the system changes significantly in a step-like manner (step function-like manner), and the frequency changes linearly at least for a short period of time.

[0215] Here, when an event such as the above occurs in the AC interconnection system R, the estimation accuracy of the system inertia of the AC interconnection system R by the RoCoF method depends on the estimation accuracy of the time when the event occurred in the AC interconnection system R. And the estimation accuracy of the time depends on the estimation proficiency of the person who estimates the time. For this reason, in the RoCoF method, it has been desired to establish a method for objectively determining the time, independent of the estimation proficiency of the person who estimates the time.

[0216] In this regard, a method is known in which the system inertia of the AC interconnection system R is estimated by the RoCoF method using the time when the RoCoF falls below 0.4 [Hz / s] as the time when an event occurs in the AC interconnection system R (see Non-Patent Document 2). However, there is no reason to justify using the time when RoCoF falls below 0.4 [Hz / s] as the time when an event occurs in the AC interconnection system R other than the fact that the estimation accuracy of the system inertia of the AC interconnection system R becomes high empirically. In other words, the method described in the above Non-Patent Document 2 cannot eliminate the subjectivity of the empirical rule, and the objectivity of the estimation of the time when an event occurs in the AC interconnection system R is insufficient.

[0217] Therefore, we investigated a new RoCoF method to improve the objectivity of estimating the time of event occurrence. The novel RoCoF method will now be described.

[0218] In this embodiment, the information processing system 1 estimates the system inertia of the AC interconnection system R by the RoCoF method. In the RoCoF method, the RoCoF (i.e., the rate of change of the system frequency) for the system frequency when the system frequency fluctuates is estimated, and the system inertia of the AC interconnection system R is estimated based on the estimated RoCoF.

[0219] Therefore, when estimating the system inertia of the AC interconnection system R using the RoCoF method, the information processing system 1 uses the RoCoF for the system frequency when an event occurs, such as the disconnection of a generator from the AC interconnection system R or the disconnection of a load connected to the AC interconnection system R. In this embodiment, for convenience of explanation, an event such as a disconnection of a generator from the AC interconnection system R or a disconnection of a load connected to the AC interconnection system R will be simply referred to as an event. When an event occurs, a value according to the power at each point of the AC interconnection system R also changes with the system frequency. The value is, for example, the voltage, current, and power itself at each point of the AC interconnection system R, but is not limited to these.

[0220] The measuring device 10 measures the system frequency at each time of the representative point and a value corresponding to the power at each time of the representative point (for example, the voltage, the current, the phase angle, and the power itself, etc.) The measuring device 10 is, for example, a PMU. In the following, for convenience of explanation, the measurement values ​​excluding the system frequency among the values ​​measured by the measurement device 10 will be referred to as non-frequency measurement values. The measurement device 10 may be configured to measure the system frequency at each time of the representative point, but not to measure the non-frequency measurement values ​​at each time of the representative point. In this case, the information processing device 20 acquires non-frequency measurement value information indicating the non-frequency measurement values ​​at each time of the representative point from another device different from the measurement device 10. The measurement device 10 may also be configured to measure the non-frequency measurement values ​​at each time of the representative point, without measuring the system frequency at each time of the representative point. In this case, the information processing device 20 acquires frequency information indicating the system frequency at each time of the representative point from another device different from the measurement device 10.

[0221] The information processing device 20 acquires frequency information indicating the system frequency measured by the measuring device 10 at each time from the measuring device 10. The information processing device 20 stores the acquired frequency information at each time. Furthermore, the information processing device 20 acquires non-frequency measurement value information indicating the non-frequency measurement values ​​measured by the measuring device 10 at each time from the measuring device 10 at each time. The information processing device 20 stores the acquired non-frequency measurement value information at each time. Note that when the measuring device 10 does not measure non-frequency measurement values, the information processing device 20 does not acquire non-frequency measurement value information from the measuring device 10.

[0222] Hereinafter, a case will be described in which the non-frequency measurement value is power as an example. For convenience of explanation, non-frequency measurement value information indicating power as the non-frequency measurement value will be referred to as power information.

[0223] The information processing device 20 has, for example, three estimation modes, a first estimation mode, a second estimation mode, and a third estimation mode, as operation modes for estimating the system inertia of the AC interconnection system R. Note that the information processing device 20 may be configured to have another estimation mode instead of some or all of the three estimation modes, or in addition to some or all of the three estimation modes. Also, the information processing device 20 may be configured to have some of the three estimation modes.

[0224] The first estimation mode is an operation mode that, among operation modes for estimating the system inertia of the AC interconnection system R, involves estimation of an event occurrence time based on frequency information acquired from the measurement device 10. Note that the event occurrence time is the time when an event occurs. In this embodiment, an example of the first estimation mode is a mode of the RoCoF0 or RoCoF1 method.

[0225] The second estimation mode is an operation mode among operation modes for estimating the system inertia of the AC interconnection system R, which involves estimation of an event occurrence time based on power information (an example of non-frequency measurement value information) acquired from the measurement device 10. In this embodiment, an example of the second estimation mode is a mode of the RoCoF0 or RoCoF1 method.

[0226] The third estimation mode is an operating mode among operating modes for estimating the system inertia of the AC interconnection system R, which involves estimating the time when an event occurs based on frequency information acquired from the measurement device 10 and power information acquired from the measurement device 10. In this embodiment, an example of the third estimation mode is a mode of the RoCoF2 method.

[0227] If the information processing device 20 does not have the second estimation mode and the third estimation mode, the information processing device 20 may be configured not to acquire power information.

[0228] In the first estimation mode, the information processing device 20 estimates the event occurrence time based on frequency information of a period corresponding to the received operation among the frequency information stored in advance. More specifically, in the first estimation mode, the information processing device 20 specifies each of the estimation period, the first time, and the second time according to the received operation. Then, the information processing device 20 estimates the event occurrence time based on frequency information of a target period from the specified first time to the specified second time among frequency information indicating the system frequency for each time at the representative point. Thereby, the information processing device 20 can improve the objectivity of the estimation of the event occurrence time. Here, the estimation period is a period during which the event is estimated to have occurred. The target period is a period during which the event occurrence time is estimated. The first time is a time before the estimation period and is the time when the target period starts. The second time is a time after the estimation period and is the time when the target period ends. The target period may be any period as long as it includes the estimation period. For example, the first time is a time several seconds before the start of the estimation period. This is because the grid frequency at the representative point before the event occurs is kept almost constant. Also, for example, the second time is a time several seconds after the start of the estimation period. This is because fluctuations in the grid frequency at the representative point caused by an event often settle within a few seconds.

[0229] Furthermore, in the first estimation mode, when estimating an event occurrence time, the information processing device 20 derives a fitting function that fits the waveform of the grid frequency (i.e., the temporal change in the grid frequency) indicated by the frequency information in the target period by solving an optimization problem. Then, the information processing device 20 estimates the RoCoF for the grid frequency in the estimation period based on the derived fitting function. Then, the information processing device 20 estimates the system inertia of the AC interconnection system R based on the estimated RoCoF and the RoCoF method. In this case, the information processing device 20 can use the result of HPF-1.

[0230] In the second estimation mode, the information processing device 20 estimates the event occurrence time based on power information for a period corresponding to the received operation among the power information stored in advance. More specifically, in the second estimation mode as well, the information processing device 20 specifies the estimated period, the first time, and the second time according to the received operation. Then, the information processing device 20 estimates the event occurrence time based on power information for a target period from the specified first time to the specified second time among the power information indicating the power at each time at the representative point. In this case as well, the information processing device 20 can improve the objectivity of the estimation of the event occurrence time.

[0231] In the second estimation mode, the information processing device 20 derives a fitting function that fits the waveform of the grid frequency indicated by the frequency information in the target period based on the estimated event occurrence time by solving an optimization problem. Then, the information processing device 20 estimates the RoCoF based on the derived fitting function. Then, the information processing device 20 estimates the grid inertia of the AC interconnection grid R based on the estimated RoCoF and the RoCoF method. In this case, the information processing device 20 can use the result of HPF-1.

[0232] In the third estimation mode, the information processing device 20 estimates the event occurrence time based on the frequency information and power information of a period corresponding to the received operation among the frequency information and power information stored in advance. More specifically, in the third estimation mode, the information processing device 20 also specifies the estimation period, the first time, and the second time according to the received operation. Then, the information processing device 20 estimates the event occurrence time based on the frequency information and power information in the target period from the specified first time to the specified second time. In this case, the information processing device 20 can more reliably improve the objectivity of the estimation of the event occurrence time.

[0233] In the third estimation mode, when estimating an event occurrence time, the information processing device 20 derives a fitting function that fits the waveform of the grid frequency (i.e., the change in the grid frequency over time) indicated by the frequency information in the target period by solving an optimization problem. Then, the information processing device 20 estimates the RoCoF for the grid frequency in the aforementioned estimation period based on the derived fitting function. Then, the information processing device 20 estimates the system inertia of the AC interconnection system R based on the estimated RoCoF and the RoCoF method. In this case, the information processing device 20 can use the result of HPF-1.

[0234] When the first estimation mode is selected as the operation mode, the information processing device 20 estimates the system inertia of the AC interconnection system R on the basis of the acquired frequency information. Furthermore, when the second estimation mode is selected as the operation mode, the information processing device 20 estimates the system inertia of the AC interconnection system R based on the acquired power information. Furthermore, when the third estimation mode is selected as the operation mode, the information processing device 20 estimates the system inertia of the AC interconnection system R based on the acquired frequency information and power information.

[0235] <Method for estimating event occurrence time and system inertia in the first estimation mode> Hereinafter, a method for estimating an event occurrence time and a system inertia in the first estimation mode will be described with reference to Fig. 9. Fig. 9 is a diagram showing an example of a temporal change in the system frequency at a representative point before and after an event occurrence time.

[0236] The horizontal axis of the graph shown in Fig. 9 indicates time. The vertical axis of the graph indicates the system frequency at the representative point. A curve PL1 plotted on the graph is an example of a curve obtained by connecting the system frequencies measured at each time by the measurement device 10 with a straight line in chronological order. In other words, the curve PL1 indicates an example of the waveform of the system frequency at the representative point (i.e., the change over time in the system frequency).

[0237] For example, after an event occurs in the AC interconnection system R, the user of the information processing device 20 determines the aforementioned estimated period, the first time, and the second time based on a graph as shown in FIG. 9. For this reason, the information processing device 20 displays a graph as shown in FIG. 9 in response to the accepted operation. At this time, the information processing device 20 accepts information indicating a period including the event occurrence time. Then, the information processing device 20 extracts frequency information indicating the system frequency for each time at the representative point in the accepted period from the frequency information stored in advance. The information processing device 20 generates a graph as shown in FIG. 9 based on the extracted frequency information, and displays the generated graph. Note that the information processing device 20 may be configured not to display such a graph. In this case, the graph as shown in FIG. 9 is generated by another device.

[0238] Next, the user of the information processing device 20 determines the estimated period based on the graph displayed on the information processing device 20. More specifically, the user determines the period estimated to include the event occurrence time based on the graph as the estimated period. For example, the user determines the period including the period in which it is visually possible to determine that the system frequency at the representative point has changed to a first predetermined value or more as the estimated period. This is because the system frequency at the representative point changes to the first predetermined value or more before and after the event occurrence time. The first predetermined value is, for example, about 0.2% of the several-second moving average (for example, a 5-second moving average line) of the system frequency at the representative point, but is not limited to this. Note that the method of determining the estimated period may be other methods as long as the method is capable of determining the period including the event occurrence time as the estimated period instead of this. Note that the information processing device 20 may be configured to specify the period estimated to have occurred as the estimated period based on the extracted frequency information and a machine learning model or the like. The period PD1 shown in FIG. 9 is an example of the estimated period determined by the user based on the graph.

[0239] Next, the user of the information processing device 20 determines a time before the determined estimated period as the first time. For example, the user determines a time a first predetermined time before the time when the estimated period starts as the first time. The first predetermined time is, for example, 2 seconds. The time ts1 shown in FIG. 9 is an example of the first time determined by the user based on the estimated period. Note that the first predetermined time may be shorter than 2 seconds or longer than 2 seconds. Alternatively, the first time may be determined in another manner. The information processing device 20 may be configured to specify the first time, which is the period during which an event is estimated to have occurred, together with the estimated period, based on the extracted frequency information and a machine learning model, etc.

[0240] Next, the user of the information processing device 20 determines a time after the determined estimated period as the second time. For example, the user determines a time that is a second predetermined time after the time when the estimated period ends as the second time. The second predetermined time is, for example, 2 seconds. The time te1 shown in FIG. 9 is an example of the second time determined by the user based on the estimated period. The second predetermined time may be shorter than 2 seconds or longer than 2 seconds. The second predetermined time may be the same as the first predetermined time or may be different from the first predetermined time. The second time may be determined in another way instead of this. The information processing device 20 may be configured to specify the second time together with the estimated period in which the event is estimated to have occurred based on the extracted frequency information and a machine learning model, etc.

[0241] Here, the period from the first time ts1 to the second time te1 is divided into n-1 equal parts, and time ts1 is set as time t1, and time te1 is set as time t n By setting the time, t i Here, n may be any integer equal to or greater than 1. And i represents any integer included in a closed interval between 1 and n. The period PD2 shown in FIG. 9 is a period from time t1 to time t n This is an example of the period from time ts1 to time te1 (the period from time ts1 to time te1), that is, the aforementioned target period.

[0242] The user of the information processing device 20 inputs the estimation period, the first time, and the second time thus determined to the information processing device 20. That is, the information processing device 20 specifies the estimation period, the first time, and the second time in response to the received operation. The information processing device 20 calculates the target period based on the specified first time and second time. The information processing device 20 extracts frequency information for the calculated target period from frequency information stored in advance. The information processing device 20 estimates a fitting function that fits the waveform of the system frequency at the representative point in the target period based on the extracted frequency information by solving an optimization problem that minimizes an objective function. In the following, as an example, a case will be described in which the information processing device 20 estimates the fitting function by solving the optimization problem using the least squares method in the first estimation mode.

[0243] Here, when the information processing device 20 estimates the fitting function by the least squares method, the information processing device 20 treats the event occurrence time as an unknown (fitting parameter) together with one or more real coefficients included in the fitting function. In the following, for convenience of explanation, the candidate of the event occurrence time is referred to as a time candidate t m m is an integer in the closed interval between 1 and n. m is a time included in the estimation period. The information processing device 20 uses, for example, a polynomial as the fitting function. In the following, as an example, a case where the fitting function is a fifth-order polynomial will be described. In this case, the unknowns when the information processing device 20 estimates the fitting function by the least squares method are six real coefficients included in the fitting function, which is a fifth-order polynomial, and a time candidate t m and . Note that the fitting function may be another function such as an exponential function instead of a polynomial. The fitting function may be a polynomial of degree 4 or less, or a polynomial of degree 5 or more.

[0244] For example, at a certain time t x1 Let time candidate t m If you select , the objective function J to be minimized by the least squares method. f are formulated as shown in the following equations (61) to (63).

[0245]

number

[0246]

number

[0247]

number

[0248] The first term in the brackets on the right side of the above formula (61) is the time when the measurement device 10 i The second term in the brackets on the right side of equation (61) is a fitting function, and is expressed as a fifth-order polynomial as in equation (62) or equation (63) depending on the time. m-1 In the period from the first time to the time candidate t m In equation (62), a0, a1, a2, a3, a4, and a5 are real coefficients (i.e., fitting parameters) included in the fitting function. On the right side of equation (62), the time is the time candidate t m That is, the fitting function expressed by the formula (62) is a straight line. This corresponds to linear approximation of the waveform of the grid frequency indicated by the frequency information in the first target period. On the other hand, the right side of the formula (63) is i Equation (62) and equation (63) depend on the time candidate t m Therefore, the above formulas (61) to (63) can be rewritten as the following formulas (64) and (65).

[0249]

number

[0250]

number

[0251] Then, the minimization problem based on the above equation (64) is expressed as the following equation (66).

[0252]

number

[0253] Moreover, the solution of the above equation (66) is expressed as the following equation (67).

[0254]

number

[0255] The method of deriving the above equation (67) is similar to the general least squares method, so a detailed explanation will be omitted.

[0256] Here, the fitting function obtained by the above equation (67) is x1 This is the function that best fits the waveform of the grid frequency indicated by the frequency information during the target period. However, at time t x1 is merely one of the times included in the estimation period and is merely one of the time candidates for the event occurrence time. Therefore, the minimization problem based on the above equation (64) can be defined as the following equation (68) instead of the above equation (66), and the time candidate t m This can be expressed as a minimization problem with unknowns.

[0257]

number

[0258] Here, in the interval [t m1 ,t m2 ] indicates the estimation period, i.e., time t m1 indicates the time when the estimation period begins. Also, time t m2 indicates the time when the estimation period ends. And the interval [t m1 ,t m2 Each time included in t m The time is selected as the

[0259] By solving the minimization problem of the above equation (68), it is possible to estimate a fitting function that best fits the waveform of the grid frequency indicated by the frequency information in the target period, and to obtain a time candidate t m can be estimated as the event occurrence time.

[0260] The RoCoF used in the RoCoF method is a function obtained by differentiating the fitting function obtained in this way once with respect to time, and adding a candidate time t estimated as the event occurrence time to the derivative. m By substituting, we obtain the following equation (69).

[0261]

number

[0262] That is, in the first estimation mode, the information processing device 20 extracts frequency information in the calculated target period, and then estimates a fitting function that best fits the waveform of the grid frequency indicated by the frequency information and estimates the event occurrence time by the least squares method based on the extracted frequency information, each time included in the specified estimation period, and the above formula (68). In other words, in the first estimation mode, the information processing device 20 estimates the event occurrence time by treating each time in the estimation period as a time candidate at which the event occurs and performing linear approximation of the waveform of the grid frequency indicated by the frequency information in the first target period from the first time to the time candidate and polynomial approximation of the waveform of the grid frequency indicated by the frequency information in the second target period from the time candidate to the second time. In such an estimation of the event occurrence time, the information processing device 20 can obtain almost the same estimation result regardless of the selection of the estimation period and the target period as long as the estimation period includes the event occurrence time and the target period includes the estimation period (obviously, this does not apply when the functional form of the fitting function is changed each time estimation is performed). Therefore, in the first estimation mode, the information processing device 20 can improve the objectivity of the estimation of the event occurrence time.

[0263] Here, FIG. 10 is a diagram in which a fitting function is superimposed on the graph shown in FIG. 9. In FIG. 10, a function FF1 superimposed on the graph indicates an example of a fitting function estimated by the information processing device 20 in the first estimation mode. A time tv1 shown in FIG. 10 indicates an example of an event occurrence time estimated by the information processing device 20 together with the fitting function. From FIG. 10, it can be seen that in the first estimation mode, the information processing device 20 performs linear approximation of the waveform of the grid frequency indicated by the frequency information in the first target period, which is a period before the event occurrence time. Also, from FIG. 10, it can be seen that in the first estimation mode, the information processing device 20 performs polynomial approximation of the waveform of the grid frequency indicated by the frequency information in the second target period, which is a period after the event occurrence time.

[0264] Moreover, in the first estimation mode, the information processing device 20 calculates the RoCoF based on the above formula (69), and estimates the system inertia of the AC interconnection system R by the RoCoF method based on the calculated RoCoF. In this case, the information processing device 20 may use the result of HPF-1.

[0265] <Method for estimating event occurrence time and system inertia in the second estimation mode> Hereinafter, a method for estimating an event occurrence time and grid inertia in the second estimation mode will be described with reference to Fig. 11. Fig. 11 is a diagram showing an example of a temporal change in power at a representative point before and after an event occurrence time.

[0266] The horizontal axis of the graph shown in Fig. 11 indicates time. The vertical axis of the graph indicates power at a representative point. Curve PL2 plotted on the graph is an example of a curve obtained by connecting the power measured at each time by the measuring device 10 with a straight line in chronological order. In other words, curve PL2 indicates an example of the waveform of the power at the representative point (i.e., the change in the power over time).

[0267] For example, after an event occurs in the AC interconnection system R, the user of the information processing device 20 determines the estimation period, the first time, and the second time based on a graph such as that shown in Fig. 11. Note that the method of determining the estimation period, the first time, and the second time in the second estimation mode may be similar to the method of determining the estimation period, the first time, and the second time in the first estimation mode. Here, a case will be described in which the method of determining the estimation period, the first time, and the second time in the second estimation mode is different from the method of determining the estimation period, the first time, and the second time in the first estimation mode.

[0268] The information processing device 20 displays a graph as shown in FIG. 11 in response to the received operation. At this time, the information processing device 20 receives information indicating a period including the event occurrence time. Then, the information processing device 20 extracts power information (an example of non-frequency measurement value information) indicating power (an example of non-frequency measurement value) at each time of a representative point during the received period from the power information stored in advance. The information processing device 20 generates a graph as shown in FIG. 11 based on the extracted power information, and displays the generated graph. Note that the information processing device 20 may be configured not to display such a graph. In this case, a graph as shown in FIG. 11 is generated by another device.

[0269] Here, when an event occurs in the AC interconnection system R, the non-frequency measurement value of the representative point also changes with the change in the system frequency of the representative point. That is, in this case, the power of the representative point also changes. For this reason, the user of the information processing device 20 can determine, based on the graph displayed on the information processing device 20, a period that is estimated to include the event occurrence time as the estimated period. For example, the user determines, as the estimated period, a period that includes a period that can be visually determined to have changed the power of the representative point to a second predetermined value or more. This is because the power of the representative point changes to a second predetermined value or more before and after the event occurrence time. The second predetermined value is, for example, about 5% of a few-second moving average (for example, a 5-second moving average line) of the power of the representative point, but is not limited to this. Note that the method of determining the estimated period may be other methods as long as the method can determine the period including the event occurrence time as the estimated period. In addition, the information processing device 20 may be configured to specify, as the estimated period, a period that is estimated to have occurred in the event based on the extracted power information and a machine learning model or the like. A period PD3 shown in FIG. 11 is an example of an estimated period determined by the user based on the graph.

[0270] Next, the user of the information processing device 20 determines the first time and the second time based on the determined estimated period. The method of determining the first time and the second time after the estimated period is determined is the same as the method of determining the first time and the second time in the first estimation mode, so detailed description will be omitted. The time ts2 shown in FIG. 11 is an example of the first time determined by the user based on the estimated period. Also, the time te2 shown in FIG. 11 is an example of the second time determined by the user based on the estimated period. Note that the information processing device 20 may be configured to specify the first time and the second time, together with the estimated period, that is, the period in which an event is estimated to have occurred, based on the extracted power information and a machine learning model, etc.

[0271] Here, as described above, the period from the first time ts2 to the second time te2 is divided into n-1 equal parts, and time ts2 is set as time t1 and time te2 is set as time t n By setting the time, t i The period PD4 shown in FIG. 11 is from time t1 to time t n This is an example of the period from time ts2 to time te2 (the period from time ts2 to time te2).

[0272] The user of the information processing device 20 inputs the estimated period, the first time, and the second time thus determined to the information processing device 20. That is, the information processing device 20 specifies the estimated period, the first time, and the second time in response to the received operation. The information processing device 20 calculates the target period based on the specified first time and second time. The information processing device 20 extracts power information for the calculated target period from the power information stored in advance. The information processing device 20 estimates a fitting function that fits the waveform of power at the representative point in the target period based on the extracted power information by solving an optimization problem that minimizes an objective function. In the following, as an example, a case will be described in which the information processing device 20 estimates the fitting function by solving the optimization problem using the least squares method in the second estimation mode.

[0273] Here, it is known that when an event occurs in the AC interconnection system R, the power at the representative point changes like a step function. More precisely, it is known that in this case, the power changes like a step function not only at the representative point but also at many points in the AC interconnection system R. Therefore, when estimating the fitting function, the information processing device 20 performs linear approximation of the waveforms of the power before and after the event time. That is, the information processing device 20 uses a constant function as the fitting function. Therefore, for the sake of convenience in the following description, the power before the event occurrence time is referred to as p pre The power after the event occurrence time is p post When the information processing device 20 estimates the fitting function by the least squares method, the information processing device 20 estimates one real coefficient (i.e., p pre ), one real coefficient in the fitting function for the second time period (i.e., p post ), and the event occurrence time are treated as unknowns (fitting parameters). Note that the fitting function may be a polynomial, an exponential function, or another function instead of a constant function.

[0274] For example, each time included in the estimation period is a time candidate t m If we choose as , the objective function J to be minimized by the least squares method is p are formulated as shown in the following equations (70) to (72).

[0275]

number

[0276]

number

[0277]

number

[0278] The first term in the brackets on the right side of the above formula (70) is the time when the measurement device 10 i The measured power is shown in Equation (70). The second term in the brackets on the right side of Equation (70) is a fitting function, and is expressed by Equation (71) or Equation (72) depending on the time. m-1 In the first target period up to time t m From time t n In the second target period up to, the fitting function is expressed by formula (72). Therefore, the above formulas (70) to (72) can be rewritten as the following formulas (73) and (74).

[0279]

number

[0280]

number

[0281] Then, the minimization problem based on the above equation (73) is expressed as the following equation (75).

[0282]

number

[0283] Moreover, the solution of the above equation (75) is expressed as the following equation (76).

[0284]

number

[0285] The method of deriving the above equation (76) is similar to the general least squares method, so a detailed explanation will be omitted.

[0286] Here, as mentioned above, in the interval [t m1 ,t m2 ] indicates the estimation period, i.e., time t m1 indicates the time when the estimation period begins. Also, time t m2 indicates the time when the estimation period ends. And the interval [t m1 ,t m2 Each time included in t m The time is selected as the

[0287] By solving the minimization problem of the above equation (75), it is possible to estimate a fitting function that best fits the waveform of the power indicated by the power information in the target period, and to select a time candidate t m can be estimated as the event occurrence time.

[0288] The RoCoF used in the RoCoF method can be obtained by substituting the event occurrence time into the derivative obtained by differentiating once with respect to time the polynomial obtained by solving the minimization problem of equation (75) into which the event occurrence time estimated in this manner is substituted.

[0289] That is, in the second estimation mode, the information processing device 20 extracts power information for the calculated target period, and then estimates a fitting function that best fits the waveform of power indicated by the power information and estimates the event occurrence time by the least squares method based on the extracted power information, each time included in the determined estimation period, and the above formula (75). In other words, in the second estimation mode, the information processing device 20 estimates the event occurrence time by linearly approximating the waveform of power indicated by the power information for the first target period and the waveform of power indicated by the power information for the second target period based on the least squares method. Even in such an estimation of the event occurrence time, the information processing device 20 can obtain almost the same estimation result regardless of the selection of the estimation period and the target period, as long as the estimation period includes the event occurrence time and the target period includes the estimation period (naturally, this does not apply when the functional form of the fitting function is changed each time estimation is performed). Therefore, the information processing device 20 can improve the objectivity of the estimation of the event occurrence time even in the second estimation mode.

[0290] Here, FIG. 12 is a diagram in which a fitting function is superimposed on the graph shown in FIG. 11. In FIG. 12, a function FF2 superimposed on the graph indicates an example of a fitting function derived by the information processing device 20 in the second estimation mode. A time tv2 shown in FIG. 12 indicates an example of an event occurrence time estimated by the information processing device 20 together with the fitting function. From FIG. 12, it can be seen that in the second estimation mode, the information processing device 20 performs linear approximation of the waveform of power indicated by the power information in the first target period, which is a period before the event occurrence time. Also, from FIG. 12, it can be seen that in the second estimation mode, the information processing device 20 performs polynomial approximation of the waveform of power indicated by the power information in the second target period, which is a period after the event occurrence time.

[0291] Also in the second estimation mode, the information processing device 20 estimates the event occurrence time, and then estimates the RoCoF based on the estimated event occurrence time and the above formulas (70) to (76). Then, the information processing device 20 estimates the system inertia of the AC interconnection system R by the RoCoF method based on the estimated RoCoF. In this case, the information processing device 20 may use the result of HPF-1.

[0292] <Method for estimating event occurrence time and system inertia in the third estimation mode> Hereinafter, a method for estimating an event occurrence time and system inertia in the third estimation mode will be described.

[0293] The method for estimating an event occurrence time in the third estimation mode is a method in which a function that is a linear combination of the above equations (61) and (70) is used as an objective function to be minimized in an optimization problem for estimating a fitting function and an event occurrence time.

[0294] That is, in the third estimation mode, the information processing device 20 solves the minimization problem expressed by the following equation (77).

[0295]

number

[0296] That is, in the third estimation mode, the information processing device 20 estimates a fitting function that fits the waveform of the grid frequency in the first estimation mode, estimates a fitting function that fits the waveform of the power in the second estimation mode, and estimates the event occurrence time. f , w p Each of w is a weight in the linear combination of the objective functions in equation (77). f , w p are defined by the following formula (78). For convenience of explanation, the fitting function that fits the waveform of the grid frequency in the first estimation mode will be referred to as the first fitting function. For convenience of explanation, the fitting function that fits the waveform of the power in the second estimation mode will be referred to as the fitting function.

[0297]

number

[0298] σ in the above equation (78) f is the variance of the grid frequency indicated by the frequency information during the target period. p is the variance of power indicated by the power information in the target period. The reason for using each of these two variances as weights is that each of the two terms included in the right side of the objective function expressed by the above formula (78) is the square value of the error. Note that a configuration may be made in which a quantity other than the variance is used as such a weight.

[0299] The user of the information processing device 20 determines the estimated period, the first time, and the second time by the same method as the method of determining the estimated period, the first time, and the second time for the first estimation mode or the second estimation mode. Then, the user of the information processing device 20 inputs the determined estimated period, the first time, and the second time to the information processing device 20. That is, the information processing device 20 specifies the estimated period, the first time, and the second time in response to the received operation. The information processing device 20 calculates the target period based on the specified first time and second time. The information processing device 20 extracts frequency information and power information for the calculated target period from the power information stored in advance. The information processing device 20 solves an optimization problem to minimize the objective function expressed by equation (78) based on the extracted frequency information and power information and the above equations (77) and (78). As a result, the information processing device 20, together with the first fitting function and the second fitting function, calculates a time candidate t when the first fitting function best fits the waveform of the grid frequency indicated by the frequency information in the target period and when the second fitting function best fits the waveform of the power indicated by the power information in the target period. m is estimated as the event occurrence time.

[0300] The method for obtaining the RoCoF used in the RoCoF method may be the method in the first estimation mode or the method in the second estimation mode.

[0301] Also in the third estimation mode, the information processing device 20 calculates the RoCoF based on the above formula (69), and estimates the system inertia of the AC interconnection system R by the RoCoF method based on the calculated RoCoF. In this case, the information processing device 20 may use the result of HPF-1.

[0302] As described above, in the information processing system 1 according to this embodiment, in the information processing device 20, it is possible to apply the results obtained by the HPF-1 technique according to the first embodiment to the RoCoF method.

[0303] Note that a configuration may be implemented in which information obtained by a method other than HPF-1 (for example, inertia ratio) is applied to the RoCoF method.

[0304] (Regarding the above embodiment) [First configuration example] As one configuration example, the information processing device 20 includes a control unit 26 that estimates the inertia ratio based on a numerically obtained system matrix of an oscillation equation of the AC interconnection system R. As one configuration example, in the information processing device 20, the control unit 26 identifies the system matrix by using a record of at least one of the phase angle and the rotation speed at all times for a certain period of time. As one configuration example, in the information processing device 20, the control unit 26 applies an HPF or a BPF to at least one of the measurement quantities used for identification. As one configuration example, in the information processing device 20, the control unit 26 estimates the inertia ratio from the system matrix based on the general characteristics of a part L of the identified system matrix and the power flow sensitivity matrix K constituting it. As one configuration example, in the information processing device 20, a general feature of the power flow sensitivity matrix K is that the sum of each column of the power flow sensitivity matrix K is zero. As one configuration example, in the information processing device 20, a general feature of the power flow sensitivity matrix K is that the power flow sensitivity matrix K is a symmetric matrix. As one configuration example, in the information processing device 20, the control unit 26 numerically indexes the degree of symmetry of the tidal current sensitivity matrix K, and optimizes the index value to estimate the inertia ratio. As one configuration example, in the information processing device 20, the control unit 26 uses, as the phase angle and rotation speed, at least one of the phase angle and rotation speed of individual generators PP1 to PPz in the AC interconnection system R, or a representative phase angle and frequency of a subsystem including a plurality of generators (e.g., a part of the generators PP1 to PPz). For example, in the AC interconnection system, it is possible that the phase angle and rotation speed of individual generators can be measured in a certain subsystem A1, and only the phase angle and rotation speed of a representative generator or a representative substation can be measured in another subsystem B1, and both of these may be used. As one configuration example, in the information processing device 20, the control unit 26 excludes one or more measurement values ​​that are highly dependent on each other based on a numerical index value in order to improve estimation accuracy in identifying the system matrix.

[0305] Note that an information processing method for executing a process similar to that of the information processing device 20 described above, or a program (computer program) for realizing a process similar to that of the information processing device 20 described above may be implemented. In one configuration example, the information processing method estimates the inertia ratio based on a numerically obtained system matrix of the oscillation equation of the AC interconnection system R. As one configuration example, the program is a program for causing a computer (for example, a computer constituting the information processing device 20) to execute a step of estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of the AC interconnection system R.

[0306] [Second configuration example] As one configuration example, the information processing device 20 includes a control unit 26 that estimates the inertia of each of the generators PP1 to PPz that constitute the AC interconnection system R from the inertia ratio of the rotors of the generators PP1 to PPz and partial information related to the magnitude of inertia. As one configuration example, in the information processing device 20, the control unit 26 uses, as the inertia ratio, a value estimated based on a numerically obtained system matrix of the oscillation equation of the AC interconnection system R. As one configuration example, in the information processing device 20, the control unit 26 obtains each piece of partial information related to the magnitude of inertia from a part L of the identified system matrix and at least one identified power flow sensitivity vector k. As one configuration example, in the information processing device 20, the control unit 26 identifies the system matrix by using a record of at least one of the phase angle and the rotation speed at all times for a certain period of time. As one configuration example, in the information processing device 20, the control unit 26 identifies one power flow sensitivity vector k by using records of at least one of the normal phase angle and rotational speed, and at least one of the power value or an equivalent measurement value, for a certain period of time. As one configuration example, in the information processing device 20, the control unit 26 applies an HPF or a BPF to at least one of the measurement quantities used for identification. As one configuration example, in the information processing device 20, the control unit 26 estimates the inertia ratio from the system matrix based on a part L of the identified system matrix and general characteristics of the tidal current sensitivity matrix K that constitutes it, where the general characteristics of the tidal current sensitivity matrix K are at least one of the characteristics that the sum of each column of the tidal current sensitivity matrix K is zero or the characteristic that the tidal current sensitivity matrix K is a symmetric matrix, and the control unit 26 estimates the inertia based on both the information on the estimated inertia ratio and at least one of the partial information on the magnitude of inertia. As one configuration example, in the information processing device 20, the control unit 26 summarizes the degree to which all conditions for estimating the inertia of each of the generators PP1 to PPz are satisfied into a single numerical index, and estimates the inertia of each of the generators PP1 to PPz by optimizing the index value. As one configuration example, in the information processing device 20, the control unit 26 uses, as the phase angle and rotational speed, at least one of the phase angle and rotational speed of individual generators PP1 to PPz in the AC interconnection system R, or a representative phase angle and frequency of a partial system including multiple generators (e.g., some of the generators PP1 to PPz). As one configuration example, in the information processing device 20, the control unit 26 excludes one or more measurement values ​​that are highly dependent on each other based on a numerical index value in order to improve the accuracy of inertia estimation in identifying the system matrix.

[0307] Note that an information processing method for executing a process similar to that of the information processing device 20 described above, or a program (computer program) for realizing a process similar to that of the information processing device 20 described above may be implemented. As one configuration example, the information processing method estimates the inertia of each of the generators PP1 to PPz constituting the AC interconnection system R from the inertia ratio of the rotor of each of the generators PP1 to PPz and partial information related to the magnitude of inertia. As one configuration example, the program is a program for causing a computer (e.g., a computer constituting the information processing device 20) to execute a step of estimating the inertia of each of the generators PP1 to PPz constituting the AC interconnection system R from the inertia ratio of the rotating body of each of the generators PP1 to PPz and partial information regarding the magnitude of inertia.

[0308] Here, a program for realizing any function in any of the devices described above may be recorded (stored) on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed to perform processing. It should be noted that the term "computer system" used here may include an operating system (OS) or hardware such as peripheral devices. In addition, the term "computer-readable recording medium" refers to writable non-volatile memory such as a flexible disk, an optical magnetic disk, a ROM, or a flash memory, a portable medium such as a DVD (Digital Versatile Disc), or a storage device such as a hard disk built into a computer system. Furthermore, the computer-readable recording medium is, for example, a non-transitory recording medium.

[0309] Furthermore, the term "computer-readable recording medium" includes devices that retain a program for a certain period of time, such as volatile memory (e.g., DRAM (Dynamic Random Access Memory)) within a computer system that serves as a server or client when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. The above program may be transmitted from a computer system in which the program is stored in a storage device or the like to another computer system via a transmission medium, or by transmission waves in the transmission medium. Here, the "transmission medium" that transmits the program refers to a medium that has the function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line. The above program may be for implementing some of the above functions. Furthermore, the above program may be a so-called differential file (differential program) that can implement the above functions in combination with a program already recorded in the computer system.

[0310] The functions of any of the components in any of the above-described devices may be realized by a processor. For example, each process in the present embodiment may be realized by a processor that operates based on information such as a program and a computer-readable recording medium that stores information such as a program. Here, the functions of each part of the processor may be realized by, for example, individual hardware, or the functions of each part may be realized by integrated hardware. For example, the processor may include hardware, and the hardware may include at least one of a circuit that processes a digital signal and a circuit that processes an analog signal. For example, the processor may be configured using one or more circuit devices mounted on a circuit board, or one or both of one or more circuit elements. An IC (Integrated Circuit) or the like may be used as the circuit device, and a resistor or a capacitor or the like may be used as the circuit element.

[0311] Here, the processor may be, for example, a CPU. However, the processor is not limited to a CPU, and various processors such as a GPU (Graphics Processing Unit) or a DSP (Digital Signal Processor) may be used. The processor may be, for example, a hardware circuit using an ASIC. The processor may be, for example, a plurality of CPUs, or a hardware circuit using a plurality of ASICs. The processor may be, for example, a combination of a plurality of CPUs and a hardware circuit using a plurality of ASICs. The processor may include, for example, one or more of an amplifier circuit or a filter circuit that processes an analog signal.

[0312] Although an embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to this embodiment, and design modifications and the like are also included within the scope of the present invention that does not deviate from the gist of the present invention. [Explanation of symbols]

[0313] 1...information processing system, 10, PQ1 to PQz...measuring device, 20...information processing device, 21...processor, 22...storage unit, 23...input reception unit, 24...communication unit, 25...display unit, 26...control unit, 211...power system, 212, 312, 412-1 to 412-3, 421...PMU, 213 to 214, 313 to 314...HPF, 215, 315...estimation calculation unit, 261...acquisition unit, 262...estimation unit, 263...display control unit, 264...storage control unit, 431-1 to 431-3, PP1 to PPz...generators, 441...estimation function, 1011 to 1013, 1111 to 1113...reference characteristics, 1021 to 1023, 1121 to 1123...estimated characteristics, FF1 to FF2...function, PL1 to PL2...curve, PD1 to PD4...period, R...AC interconnection system, ts1 to ts2, te1 to te2, tv1 to tv2...time

Claims

1. A control unit is provided for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, the control unit estimates the inertia ratio from the system matrix based on a part of the identified system matrix and general characteristics of a power flow sensitivity matrix constituting the part of the identified system matrix; A general feature of the load flow sensitivity matrix is ​​that each column of the load flow sensitivity matrix sums to zero. Information processing device.

2. The control unit identifies the system matrix using a record of at least one of a phase angle and a rotation speed at all times for a certain period of time. The information processing device according to claim 1 .

3. The control unit applies HPF or BPF to at least one of the measured quantities used for identification.

3. The information processing device according to claim 1 or 2.

4. A control unit is provided for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, the control unit estimates the inertia ratio from the system matrix based on a part of the identified system matrix and general characteristics of a power flow sensitivity matrix constituting the part of the identified system matrix; A general feature of the load flow sensitivity matrix is ​​that it is a symmetric matrix. Information processing device.

5. The control unit numerically indexes a degree of symmetry of the power flow sensitivity matrix and estimates the inertia ratio by optimizing the index value. The information processing device according to claim 4.

6. The control unit identifies the system matrix using a record of at least one of a phase angle and a rotation speed at all times for a certain period of time.

6. The information processing device according to claim 4 or 5.

7. The control unit applies HPF or BPF to at least one of the measured quantities used for identification. The information processing device according to any one of claims 4 to 6.

8. A control unit is provided for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, The control unit identifies the system matrix using a record of at least one of a phase angle and a rotation speed at all times for a certain period of time; The control unit uses, as the phase angle and the rotational speed, at least one of a phase angle and a rotational speed of an individual generator in the AC interconnection system, or a representative phase angle and a representative frequency of a partial system including a plurality of generators; The control unit excludes one or more measurement values ​​that are highly dependent on each other based on a numerical index value in order to improve estimation accuracy in identifying the system matrix. Information processing device.

9. 1. An information processing method for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, comprising: estimating the inertia ratio from the system matrix based on general characteristics of the identified part of the system matrix and the load flow sensitivity matrix constituting the part; A general feature of the load flow sensitivity matrix is ​​that each column of the load flow sensitivity matrix sums to zero. Information processing methods.

10. 1. An information processing method for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, comprising: estimating the inertia ratio from the system matrix based on general characteristics of the identified part of the system matrix and the load flow sensitivity matrix constituting the part; A general feature of the load flow sensitivity matrix is ​​that it is a symmetric matrix. Information processing methods.

11. 1. An information processing method for estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of an AC interconnection system, comprising: Identifying the system matrix using a record of at least one of the phase angle and the rotation speed at all times over a certain period of time; At least one of a phase angle and a rotational speed of an individual generator in the AC interconnected system, or a representative phase angle and a representative frequency of a subsystem including a plurality of generators is used as the phase angle and the rotational speed; In the identification of the system matrix, one or more measurement values ​​that are highly dependent on each other are excluded based on a numerical index value in order to improve estimation accuracy. Information processing methods.

12. On the computer, estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of the AC interconnected grid; A program for executing estimating the inertia ratio from the system matrix based on general characteristics of the identified part of the system matrix and the load flow sensitivity matrix constituting the part; A general feature of the load flow sensitivity matrix is ​​that each column of the load flow sensitivity matrix sums to zero. program.

13. On the computer, estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of the AC interconnected grid; A program for executing estimating the inertia ratio from the system matrix based on general characteristics of the identified part of the system matrix and the load flow sensitivity matrix constituting the part; A general feature of the load flow sensitivity matrix is ​​that it is a symmetric matrix. program.

14. On the computer, estimating an inertia ratio based on a numerically obtained system matrix of an oscillation equation of the AC interconnected grid; A program for executing Identifying the system matrix using a record of at least one of the phase angle and the rotation speed at all times over a certain period of time; At least one of a phase angle and a rotational speed of an individual generator in the AC interconnected system, or a representative phase angle and a representative frequency of a subsystem including a plurality of generators is used as the phase angle and the rotational speed; In the identification of the system matrix, one or more measurement values ​​that are highly dependent on each other are excluded based on a numerical index value in order to improve estimation accuracy. program.