Power factor estimation device, power factor estimation method, and power factor estimation program

Abstract

To provide a power factor estimation device which can improve reliability of a distribution system.SOLUTION: A first power generation output calculation part 13 calculates a solar light power generation output in each time by a maximum tidal current on the basis of the maximum value of a tidal current value in each time in a past predetermined period. A second power generation output calculation part 14 calculates the solar light power generation output in each time by a first vector decomposition by using a value of an operation power factor of the solar light power generation as a numerical constant on the basis of a valid power fluctuation and an invalid power fluctuation in a clear day, and calculates the solar light power generation output in each time by a second vector decomposition by using a value of the power factor of a load specified on the basis of the solar light power generation output in each time by the first vector decomposition on the basis of the valid power fluctuation and the invalid power fluctuation. A power factor determination part 15 specifies a value of the power factor of the load on the basis of the solar light power generation output in each time by the maximum tidal current and the solar light power generation output in each time by the first vector decomposition, and specifies the operation power factor of the solar light power generation on the basis of the solar light power generation output in each time by the maximum tidal current or the solar light power generation output by the second vector decomposition.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a power factor estimation device, a power factor estimation method, and a power factor estimation program. [Background technology]

[0002] In recent years, the amount of renewable energy, especially photovoltaic (PV) power generation, has increased rapidly, and its introduction is expected to continue in the future. The introduction of photovoltaic power generation helps to secure power during the daytime. On the other hand, power flows in both directions, from the system to the consumer, and the amount of power generation can change abruptly depending on the weather, making it more difficult to grasp the state of the system. In recent years, the introduction of distribution line sensor switches and smart meters has diversified the types of data that can be obtained, but it remains difficult to grasp the respective power flows in a distribution system where loads and photovoltaic power generation are generally mixed.

[0003] If the load and the output of photovoltaic power generation are not understood, problems such as the impact on power quality due to voltage rises, incorrect operation in power interchange during distribution line construction or supply interruption accidents, and the formation of vulnerable facilities due to incorrect demand assumptions may occur. Therefore, it is desirable to accurately separate the load and the output of photovoltaic power generation. To address such problems, a method called the "vector decomposition method" has been developed that uses only distribution line sensor information.

[0004] Under the condition that the operating power factor of the photovoltaic power generation and the power factor of the load are constant, the power flow fluctuation can be vector-decomposed into the load and the photovoltaic power generation output on the PQ plane consisting of two axes of active power P and reactive power Q. However, if reactive power is generated by a power factor improvement capacitor, the load curve on the PQ plane does not pass through the origin, and the value of the photovoltaic power generation output cannot be uniquely determined when vector decomposition is performed. Therefore, in order to eliminate the influence of the power factor improvement capacitor, we focus on the characteristic that the ratio of the reactive power fluctuation value to the active power fluctuation value of the load per unit time is roughly constant regardless of the capacitor capacity. In the vector decomposition method, the power flow fluctuation component obtained at each measurement time from a sensor installed on the distribution line is vector-decomposed into the load fluctuation component and the photovoltaic power generation output fluctuation component to estimate the value of the photovoltaic power generation output fluctuation.

[0005] In addition, in recent years, simulations using analysis tools are essential to understand the state of increasingly complex power distribution systems. To verify voltage and current conditions using simulations, it is necessary to set information on the load and photovoltaic power output at any cross section. In particular, the load estimation procedure is more complicated than the photovoltaic power output, which indicates a mechanical value based on the amount of solar radiation, and there is a need to improve the accuracy. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] JP 2016-158371 A [Non-patent literature]

[0007] [Non-Patent Document 1] Junichi Sakaguchi, Satoshi Uemura, Masahiro Matsuda, "Development of a method to estimate photovoltaic power output for each section using distribution line sensor information", Central Research Institute of the Electric Power Industry Research Report R14012, 2015 Summary of the Invention [Problem to be solved by the invention]

[0008] The vector decomposition method performs analysis under the assumption that the power factor of the load and photovoltaic power generation output fluctuations is fixed throughout the day. However, strictly speaking, the actual power factor changes from moment to moment due to the load usage status and the voltage rise suppression function of the PCS (Power Conditioning Subsystem). In conventional power factor estimation methods, a fixed power factor affects the estimation accuracy for data with large variations in power factor at each time slice due to these factors, resulting in poor decomposition accuracy. In this case, it is difficult to grasp the net load amount minus the amount of power generation, which can hinder interchange operations during accidents and make it difficult to maintain the reliability of the power distribution system.

[0009] The disclosed technology has been made in consideration of the above, and aims to provide a power factor estimation device, a power factor estimation method, and a power factor estimation program that improve the reliability of a power distribution system. [Means for solving the problem]

[0010] In one aspect of the power factor estimation device, the power factor estimation method, and the power factor estimation program disclosed herein, the first power output calculation unit calculates the photovoltaic power generation output for each hour due to the maximum power flow based on the maximum power flow value for each hour in a predetermined period in the past. The second power output calculation unit calculates the photovoltaic power generation output for each hour by the first vector decomposition using the value of the operating power factor of the photovoltaic power generation as a constant based on the active power fluctuation and the reactive power fluctuation on a clear day included in the predetermined period in the past, obtains a value of the power factor of the load specified based on the photovoltaic power generation output for each hour by the first vector decomposition, and calculates the photovoltaic power generation output for each hour by the second vector decomposition using the value of the power factor of the load specified based on the active power fluctuation and the reactive power fluctuation. The power factor determination unit specifies the value of the power factor of the load based on the photovoltaic power generation output for each hour due to the maximum power flow and the photovoltaic power generation output for each hour by the first vector decomposition, and specifies the operating power factor of the photovoltaic power generation based on the photovoltaic power generation output for each hour due to the maximum power flow and the photovoltaic power generation output for each hour by the second vector decomposition. Effect of the Invention

[0011] In one aspect, the present invention can improve the reliability of an electrical power distribution system. [Brief description of the drawings]

[0012] [Figure 1] FIG. 1 is a schematic diagram of a power distribution system according to a first embodiment. [Diagram 2] FIG. 2 is a block diagram of the power factor estimating device. [Diagram 3] FIG. 3 is a diagram for explaining vector decomposition of tidal flow fluctuation into load fluctuation and PV output fluctuation. [Figure 4] FIG. 4 is a diagram for explaining vector decomposition of tidal flow fluctuation into load fluctuation and PV output fluctuation. [Diagram 5] FIG. 5 is a diagram showing PV output fluctuations when all PV outputs are under constant power factor control. [Figure 6] FIG. 6 is a diagram for explaining the power factor of the fluctuation component. [Figure 7] FIG. 7 is a flowchart of a power factor estimation process performed by the power factor estimation device according to the first embodiment. [Figure 8] FIG. 8 is a flowchart of the vector decomposition process by the decomposition unit. [Figure 9] FIG. 9 is a diagram showing the results of estimating the power factor when the fluctuation pattern of the active power is used as a reference. [Figure 10] FIG. 10 is a diagram showing the results of estimating the power factor when the fluctuation pattern of the reactive power is used as a reference. [Figure 11] FIG. 11 is a flowchart of a power factor estimation process performed by the power factor estimation device according to the second embodiment. [Figure 12] FIG. 12 is a hardware configuration diagram of the power factor estimating device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0013] Hereinafter, embodiments of the power factor estimation device, the power factor estimation method, and the power factor estimation program disclosed in the present application will be described in detail with reference to the drawings. Note that the power factor estimation device, the power factor estimation method, and the power factor estimation program disclosed in the present application are not limited to the following embodiments. EXAMPLES

[0014] Fig. 1 is a schematic diagram of a power distribution system according to a first embodiment. In the power distribution system 1 according to the present embodiment, a power distribution substation 2, a sensor 3, a load 4, and a photovoltaic power generation (PV) system 5 are arranged. The power distribution substation 2, the load 4, and the photovoltaic power generation system 5 are connected by a power distribution line. A sensor 3 that measures the power status of the power distribution line is arranged between the power distribution substation 2 and the load 4 and the photovoltaic power generation system 5. Furthermore, a power factor estimation device 10 that estimates a power factor using information acquired by the sensor 3 is connected to the sensor 3 by a network.

[0015] The sensor 3 periodically measures the voltage, current, and power factor in the distribution line of the power distribution system 1. For example, a sensor switch in the distribution line to be managed can be used as the sensor 3. Hereinafter, data combining the measurement values ​​of the voltage, current, and power factor measured by the sensor 3 and the measurement time will be referred to as "distribution line sensor information."

[0016] The measurement intervals of the voltage, current, and power factor by the sensor 3 are preferably set according to, for example, the expected fluctuation pitch and the desired estimation accuracy of the value of the load 4 and the PV output, which is the output of the photovoltaic power generation 5. The measurement intervals by the sensor 3 may be set to an appropriate length within the range of, for example, about 30 seconds to 30 minutes, and are set to about 1 minute in this embodiment.

[0017] In this embodiment, the measurement interval by the sensor 3 is used as the predetermined time interval Δt in the calculation of the power factor by the power factor estimation device 10 in the following description. However, the measurement interval by the sensor 3 and the predetermined time interval Δt in the calculation of the power factor may be different. For example, a moving average may be calculated for each predetermined time interval Δt using measurement values ​​acquired by the sensor 3 periodically performing measurements, and the moving average value for each predetermined time interval Δt may be used as the power distribution line sensor information.

[0018] A power factor estimation device 10 according to the present embodiment includes a power calculation unit 11, a power flow maximum value specification unit 12, a first power output calculation unit 13, a second power output calculation unit 14, a power factor determination unit 15, and a storage unit 16, as shown in FIG.

[0019] The storage unit 16 receives power distribution line sensor information for a predetermined period in the past from the sensor 3. Then, the storage unit 16 generates and stores a sensor data file 161 that records combination data that combines the measurement times by date and the measurement values ​​of the voltage, current, and power factor included in the acquired power distribution line sensor information.

[0020] Hereinafter, the predetermined period in the past during which the power distribution line sensor information recorded in the sensor data file 161 was acquired will be referred to as the "sensor data period." This sensor data period is preferably set to an appropriate length of time, taking into consideration, for example, changes in the output of the load 4 and the photovoltaic power generation system 5 connected to the power distribution line to be managed. The sensor data period can be set to an appropriate length of time within a range of, for example, about one week to two months in the past, and is set to about one month in the present embodiment.

[0021] The power calculation unit 11 reads combination data of the measurement time for each day during the sensor data period and the measurement values ​​of voltage, current, and power factor from the sensor data file 161 stored in the memory unit 16. Next, the power calculation unit 11 calculates active power and reactive power for each measurement time for each day using the read combination data. Here, active power corresponds to the power flow value. Then, the power calculation unit 11 outputs power data in which the calculated active power and reactive power are associated with the date and measurement time to the power flow maximum value identification unit 12, the first power generation output calculation unit 13, and the second power generation output calculation unit 14.

[0022] Power flow maximum value identifyr 12 receives power data in which active power and reactive power are associated with dates and measurement times from power calculation unit 11. Next, power flow maximum value identifyr 12 identifies the maximum power flow value for each time in the sensor data period using the power data. That is, power flow maximum value identifyr 12 extracts, for each time of day, the maximum power flow value among the power flow values ​​at the same time on each date, for each time of day in the sensor data period.

[0023] Here, the amount of power generated by the photovoltaic power generation 5 varies depending on the weather. It is considered that there are times when the PV output is almost zero on cloudy days or rainy days. Therefore, it is assumed that there are times when the active power of the power flow and the power consumption of the load 4 are almost equal on cloudy days or rainy days. Therefore, it can be said that the distribution line sensor information for a predetermined period contains actual measurement data in which the PV output is close to 0 at each time of the day, and contains data in which the power flow and the load 4 are almost equal at each time of the day. It is assumed that when the PV output is almost 0, the value of the active power of the power flow at that time is maximum. Therefore, if the load 4 at each time is not affected by the weather, the maximum value of the power flow extracted at each time by the power flow maximum value identification unit 12 can be regarded as the net value of the load 4 at each time. The power flow maximum value identification unit 12 outputs maximum power flow data that combines each time of the day and the maximum power flow value to the first power generation output calculation unit 13. The maximum value of the power flow identified for each time corresponds to the net value of the load 4 for each time, and each power flow maximum value data represents the net value of the load 4 at each time.

[0024] The first power generation output calculation unit 13 calculates the output of the photovoltaic power generation unit 5 on a specific clear day using the maximum value of the tide current at each time of the day and the tide flow rate on a specific clear day. As described above, the maximum value of the tide current extracted for each time by the tide current maximum value identification unit 12 can be regarded as the net load 4 at each time. Therefore, the PV output for each time on a specific day can be calculated by subtracting the tide current value actually measured for each time on a specific day from the maximum tide current value for each time during the sensor data period. A method for calculating the PV output by the first power generation output calculation unit 13 will be described in detail below.

[0025] The first power generation output calculation unit 13 receives an input of power data in which active power and reactive power are associated with dates and measurement times from the power calculation unit 11. It also receives an input of power flow maximum value data from the power flow maximum identification unit 12, which combines each time of day with the maximum power flow value.

[0026] Next, the first power generation output calculation unit 13 selects a specific sunny day in the sensor data period. Here, the first power generation output calculation unit 13 may set the specified day as a sunny day, or may obtain weather information for each day in the sensor data period from a weather information database or the like and specify a sunny day from the percentage of clouds or the like. Also, if there are multiple sunny days, the first power generation output calculation unit 13 selects one specific sunny day from the multiple sunny days.

[0027] Then, the first power output calculation unit 13 acquires power data for a specific sunny day from the power data in which active power and reactive power are associated with dates and measurement times. Next, the first power output calculation unit 13 uses the power data for the specific sunny day to determine a power flow value for each hour on the specific sunny day. Then, the first power output calculation unit 13 subtracts the power flow value for each hour on the specific sunny day from the maximum power flow value for each hour represented by the power flow maximum data to calculate a power flow difference value. Next, the first power output calculation unit 13 outputs the calculated difference value to the power factor determination unit 15 as the PV output for the specific sunny day. Hereinafter, the output of the photovoltaic power generation unit 5 on the specific sunny day calculated by the first power output calculation unit 13 is referred to as the "PV output for each hour due to maximum power flow". In addition, the first power output calculation unit 13 outputs information on the specific sunny day to the second power output calculation unit 14.

[0028] The second power generation output calculation unit 14 performs vector decomposition using distribution line sensor information related to a specific sunny day, and calculates the output of the photovoltaic power generation 5 for each hour by a method different from that of the first power generation output calculation unit 13. A method of calculating the output of the photovoltaic power generation 5 using vector decomposition by the second power generation output calculation unit 14 will be described below.

[0029] Under conditions where the operating power factor of the photovoltaic power generation 5 and the power factor of the load 4 are constant, the tidal flow fluctuation can be vector-decomposed into the load 4 and the output of the photovoltaic power generation 5 on a PQ plane consisting of two axes of active power P and reactive power Q. However, if reactive power is generated by a power factor improvement capacitor or the like, the load curve on the PQ plane will no longer pass through the origin, and the output value of the photovoltaic power generation 5 cannot be uniquely determined when vector decomposition is performed. In the following, the load 4 will be simply referred to as the load, and the output of the photovoltaic power generation 5 will be simply referred to as the PV output.

[0030] Therefore, in order to eliminate the influence of the power factor improvement capacitor, attention is focused on the characteristic that the ratio of the reactive power fluctuation value to the active power fluctuation value of the load 4 per unit time is roughly constant regardless of the capacitor capacity. In this case, the PV output fluctuation value can be calculated by vector decomposing the power flow fluctuation component obtained at each measurement time from the sensor 3 installed on the distribution line into the load fluctuation component and the PV output fluctuation component.

[0031] Fig. 3 is a diagram for explaining vector decomposition of tidal fluctuation into load fluctuation and PV output fluctuation. More specifically, Fig. 3 shows an example in which a vector representing tidal fluctuation falls in the first quadrant. Fig. 4 is a diagram for explaining vector decomposition of tidal fluctuation into load fluctuation and photovoltaic power generation output fluctuation. More specifically, Fig. 4 shows an example in which a vector representing tidal fluctuation falls in the second quadrant.

[0032] When the power flow fluctuation at a specific time is depicted on the PQ plane, the power flow fluctuation can be decomposed into load fluctuation and PV output fluctuation, as shown in Figures 3 and 4. For example, when the load 4 increases, both the active power and reactive power increase, and at the same time, when the PV output increases, the active power tends to decrease. Then, when the load fluctuation is large, the combined power flow fluctuation falls into the first quadrant as shown in Figure 3, and when the PV output fluctuation is large, it falls into the second quadrant as shown in Figure 4.

[0033] Here, the magnitude of the vectors for the power flow fluctuation, load fluctuation, and PV output fluctuation are expressed as the apparent power fluctuation ΔS and ΔS L , ΔS G Then, the cosine values ​​cosχ, cosθ, and cosφ are the respective power factors. Using these values, the PV output fluctuation ΔP can be calculated from the following formula (1) or (2) using the sine theorem in the example shown in Figure 3 or Figure 4. G Formula (1) is used when the tidal current fluctuation vector shown in Figure 3 falls in the first quadrant, and formula (2) is used when the tidal current fluctuation vector shown in Figure 4 falls in the second quadrant.

[0034]

number

[0035]

number

[0036] In addition, the load fluctuation ΔP L Similarly, it can be found from the following formula (3) or (4) using the sine theorem. Formula (3) is used when the tidal current fluctuation vector shown in Figure 3 falls in the first quadrant, and formula (4) is used when the tidal current fluctuation vector shown in Figure 4 falls in the second quadrant.

[0037]

number

[0038]

number

[0039] The formula for calculating PV output fluctuations in all cases, including cases where the tidal current fluctuation vector enters the third or fourth quadrant, is summarized in Table 1.

[0040] [Table 1]

[0041] The power factor used below is not the power factor of the output of the photovoltaic power generation 5 and the load 4 themselves, but the power factor of each of the fluctuations. FIG. 5 is a diagram showing the PV output fluctuation when all PV outputs are controlled to constant power factor. FIG. 6 is a diagram for explaining the power factor of the fluctuations. As shown in FIG. 5, if it is assumed that all PV outputs are controlled to constant power factor, the power factor of the output and the power factor of the fluctuations can be considered to be the same value. On the other hand, the load 4, which mainly includes reactive power of the power factor improvement capacitor, does not pass through the origin on the PQ plane as shown in FIG. 6, and is therefore different from the power factor of the output of the photovoltaic power generation 5 and the load 4 themselves. In the following, in order to distinguish them from the power factors of the output of the photovoltaic power generation 5 and the load 4 themselves, these are called the "PV fluctuation power factor" and the "load fluctuation power factor".

[0042] Using the above, a method for calculating the PV output value and the value of the load 4 using vector decomposition by the second power generation output calculation unit 14 will now be specifically described. The second power generation output calculation unit 14 acquires information on a specific sunny day from the first power generation output calculation unit 13. Next, the second power generation output calculation unit 14 acquires power data from the power calculation unit 11. Then, the second power generation output calculation unit 14 extracts power data representing active power and reactive power for each time of the specific sunny day from the acquired power data.

[0043] Next, the second power generation output calculation unit 14 calculates the apparent power fluctuation ΔS of the tidal current from the active power fluctuation ΔP and the reactive power fluctuation ΔQ in a predetermined time interval Δt on a specific sunny day, using the following equation (5).

[0044]

number

[0045] Next, the second power generation output calculation unit 14 determines the quadrant from the signs of the active power fluctuation ΔP and the reactive power fluctuation ΔQ, and calculates the PV output fluctuation ΔP to be applied from the calculation formulas summarized in Table 1. G and load fluctuation ΔP LThen, the second power generation output calculation unit 14 determines a PV output fluctuation ΔP G (t) and the load fluctuation ΔP, which is the fluctuation value of load 4 between time [t-Δt] and time [t] L Calculate (t).

[0046] Here, the PV output fluctuation ΔP G When calculating (t), sinθ and sinφ corresponding to the value of the load fluctuation power factor cosθ and the value of the PV fluctuation power factor cosφ are used. In this embodiment, the operating power factor of the photovoltaic power generation 5 is assumed to be a case in which reactive power is injected with a constant power factor operation. This is because, although separation is possible even when the operating power factor of the photovoltaic power generation 5 is 1.0, when the power factor of the load fluctuation of the load 4 also becomes 1.0, the vectors of both become parallel, making separation difficult. In this case, the second power generation output calculation unit 14 can obtain the power factor angle θ of the load fluctuation and the power factor angle φ of the PV output fluctuation using the following formulas (6) and (7). These are constant values ​​in the calculation formulas in Table 1.

[0047]

number

[0048]

number

[0049] Then, the second power generation output calculation unit 14 calculates the obtained PV output fluctuation ΔP G (t) and load fluctuation ΔP L (t) is integrated for each time, and the PV output P at time T is expressed by the following formulas (8) and (9). G (T) and load P L (T) can be obtained.

[0050]

number

[0051]

number

[0052] Here, let us consider the correlation coefficient between the PV output for each hour due to maximum power flow and the PV output calculated by vector decomposition using the calculation formula in Table 1. In this case, the correlation coefficient changes with the change in the load fluctuation power factor cosθ, but the PV fluctuation power factor cosφ does not affect the correlation coefficient regardless of the value. This is for the following reason. That is, the vector decomposition method uses a different calculation formula depending on the quadrant of the power flow fluctuation, as shown in Table 1. In addition, both the power factor angle θ of the load fluctuation and the power factor angle φ of the PV output fluctuation are fixed regardless of the quadrant. In addition, the power factor angle θ of the load fluctuation is applied to a different calculation formula for each quadrant. On the other hand, the part of cosφ / sin(θ+φ) that is affected by the power factor angle φ of the PV output fluctuation does not change regardless of the quadrant, so a fixed value for 24 hours is multiplied in the integration to calculate the PV output. From this, it can be seen that the correlation coefficient changes with the load fluctuation power factor cosθ without being affected by the change in the power factor angle φ of the PV output fluctuation.

[0053] In the calculation formula of Table 1, the value of the load fluctuation power factor cosθ and the value of the PV fluctuation power factor cosφ are used. Therefore, in the first stage of power factor estimation, the second power generation output calculation unit 14 fixes the PV fluctuation power factor cosφ as a constant, performs vector decomposition with the load fluctuation power factor cosθ, and calculates the PV output fluctuation ΔP at time t for each value of the load fluctuation power factor cosθ while changing the phase of the power factor angle θ of the load fluctuation. G Calculate (t).

[0054] The range in which the value of the load fluctuation power factor cosθ is changed is not limited to a specific range of values, and is preferably set appropriately in an appropriate range, taking into consideration the value of the load fluctuation power factor cosθ expected in the distribution line to be managed, for example. The value of the load fluctuation power factor cosθ may be changed within the range of -1 to +1, for example.

[0055] Moreover, the interval of change, which is the interval of change when the value of the load fluctuation power factor cosθ is changed, is not limited to a specific value, and is preferably set to an appropriate value taking into consideration, for example, the estimation accuracy and the expected calculation time, etc. Specifically, the value of the load fluctuation power factor cosθ can be changed in intervals of about 0.01, for example.

[0056] Next, the second power generation output calculation unit 14 calculates the PV output fluctuation ΔP G By accumulating (t), the PV output P at time T corresponding to each value of the load fluctuation power factor on a specific sunny day is calculated. G Calculate (T).

[0057] The second power generation output calculation unit 14 calculates the PV output P at time T on a specific sunny day by vector decomposition with the PV variable power factor cosφ as a constant. G (T) is output to the power factor determination unit 15 as the PV output on a specific sunny day. Hereinafter, the PV output on a specific sunny day calculated by the second power generation output calculation unit 14 with the PV fluctuation power factor cosφ as a constant will be referred to as the "PV output for each hour by vector decomposition with the PV output fluctuation power factor as a constant." The PV output for each hour by vector decomposition with the PV output fluctuation power factor as a constant is calculated by the value of the load fluctuation power factor cosθ, a constant representing the PV fluctuation power factor, the time t and the PV output P G The data is a combination of the (T) values.

[0058] Thereafter, the second power generation output calculation unit 14 receives an input of the determined value of the load fluctuation power factor cosθ from the power factor determination unit 15. Next, as a second stage of power factor determination, the second power generation output calculation unit 14 performs vector decomposition with the PV fluctuation power factor using the load fluctuation power factor cosθ determined by the power factor determination unit 15, and calculates the PV output fluctuation ΔP G Calculate (t).

[0059] The range within which the value of the PV variable power factor cosφ is changed is not limited to a specific range of values, and is preferably set appropriately within an appropriate range in consideration of, for example, values ​​stipulated in the grid interconnection regulations for the photovoltaic power generation 5 as a distributed power source. The value of the PV variable power factor cosφ may be changed within a range of, for example, 0.85 to 1.0.

[0060] Moreover, the interval of change, which is the interval of change when the value of the PV variable power factor cosφ is changed, is not limited to a specific value, and is preferably set to an appropriate value taking into consideration, for example, the estimation accuracy, the expected calculation time, etc. The value of the PV variable power factor cosφ may be changed in intervals of about 0.01, for example, like the load variable power factor.

[0061] Next, the second power generation output calculation unit 14 calculates the PV output fluctuation ΔP G By accumulating (t), the PV output P at time T corresponding to each value of the PV variable power factor on a specific sunny day is calculated. G Calculate (T).

[0062] Thereafter, the second power generation output calculation unit 14 calculates the PV output P at time T on a specific sunny day by vector decomposition using the determined load fluctuation power factor. G (T) is output to the power factor determination unit 15 as the output of the photovoltaic power generation 5 on a specific sunny day. Hereinafter, the PV output on a specific sunny day calculated using the load fluctuation power factor determined by the second power generation output calculation unit 14 will be referred to as the "PV output for each hour by vector decomposition using the determined load fluctuation power factor." The PV output for each hour by vector decomposition using the determined load fluctuation power factor is calculated by the determined value of the load fluctuation power factor cosθ, the value of the PV fluctuation power factor cosφ, the time t, and the PV output P G The data is a combination of the (T) values.

[0063] The power factor determination unit 15 receives information on the PV output at each time due to the maximum power flow from the first power generation output calculation unit 13. In addition, as a first stage of power factor determination, the power factor determination unit 15 receives information on the PV output at each time due to vector decomposition with the PV output fluctuation power factor set as a constant from the second power generation output calculation unit 14.

[0064] Next, the power factor determination unit 15 compares the PV output at each time due to the maximum power flow with the PV output at each time corresponding to each value of the load fluctuation rate calculated by vector decomposition with the PV output fluctuation power factor as a constant. Then, the power factor determination unit 15 identifies the load fluctuation power factor that has the highest correlation between the PV outputs at the same time.

[0065] Here, the power factor determination unit 15 calculates a correlation coefficient between PV output values ​​at the same time for each value of the load fluctuation power factor. When calculating the correlation coefficient, the power factor determination unit 15 may perform moving averaging of the PV output values ​​calculated for each time of the predetermined time interval Δt for each time longer than the predetermined time interval Δt, and then calculate the correlation coefficient between the moving average values. For example, when the predetermined time interval Δt is one minute and the PV output value for each minute is calculated, the power factor determination unit 15 may calculate the moving average value for each hour of the PV output value for each minute, and calculate the correlation coefficient between the moving average values ​​for each hour.

[0066] The power factor determination unit 15 evaluates a correlation coefficient that changes depending on the load fluctuation power factor cosθ with the PV fluctuation power factor cosφ set as a constant, and selects the load fluctuation power factor cosθ. For example, the power factor determination unit 15 calculates a correlation coefficient based on the active power between the PV output at each time by vector decomposition with the PV output fluctuation power factor set as a constant and the PV output at each time by the maximum power flow. Then, the power factor determination unit 15 determines the load fluctuation power factor cosθ with the highest correlation as the estimated result of the load fluctuation power factor cosθ. When the load fluctuation power factor cosθ changes in the range of -1.0 to 1.0 and the search is performed in increments of 0.01, the power factor determination unit 15 can determine the load fluctuation power factor cosθ by searching 2 / 0.01=200 ways. After that, the power factor determination unit 15 notifies the second power generation output calculation unit 14 of the determined load fluctuation power factor cosθ.

[0067] Next, as a second stage of power factor determination, the power factor determination unit 15 receives, from the second power generation output calculation unit 14, information on the PV output for each time period obtained by vector decomposition using the determined load fluctuation power factor.

[0068] Next, the power factor determination unit 15 compares the PV output at each time due to the maximum power flow with the PV output at each time calculated by vector decomposition using the determined load fluctuation power factor. Then, the power factor determination unit 15 identifies a PV fluctuation power factor that minimizes the sum of the differences between the PV outputs at the same time. The process of identifying a combination of power factor values ​​by the power factor determination unit 15 will be specifically described below. Here, too, the power factor determination unit 15 may perform moving averaging of the PV output values ​​calculated at each time of the predetermined time interval Δt for a time longer than the predetermined time interval Δt, and then calculate the difference between the moving average values.

[0069] The power factor determination unit 15 evaluates the difference that changes depending on the PV variable power factor after the load variable power factor is set to the determined value, and determines the PV variable power factor. For example, the power factor determination unit 15 calculates the difference at each time based on the active power between the PV output at each time by vector decomposition using the determined load variable power factor and the PV output at each time by the maximum power flow. Then, the power factor determination unit 15 determines the PV variable power factor cosφ that minimizes the sum of the differences as the estimation result of the PV variable power factor cosφ. When the PV variable power factor cosφ changes in the range of 0.85 to 0.1 and the search is performed in increments of 0.01, the power factor determination unit 15 can estimate the PV variable power factor cosφ by searching 0.16 / 0.01=16 ways.

[0070] Through the above process, the power factor determination unit 15 determines the value of the load fluctuation power factor cosθ when the correlation coefficient is maximized and the value of the PV fluctuation power factor cosφ when the difference is minimized, in other words, the combination of the phase of the power factor angle θ of the load fluctuation and the phase of the power factor angle φ of the PV output fluctuation. Here, the power factor determination unit 15 according to this embodiment can estimate the load fluctuation power factor cosθ and the PV fluctuation power factor cosφ by searching 200+16=216 ways. In contrast, when the PV fluctuation power factor cosφ is not fixed and the load fluctuation power factor cosφ and the PV output fluctuation cosφ are each changed to evaluate the correlation function, 200×16=3200 ways are searched. In this way, the power factor determination unit 15 according to this embodiment can significantly reduce the amount of calculation in the power factor determination process. Thereafter, the power factor determination unit 15 outputs the estimated load fluctuation power factor cosθ and PV fluctuation power factor cosφ to the separation unit 17.

[0071] The separation unit 17 receives the determined load fluctuation power factor cosθ and PV fluctuation power factor cosφ from the power factor determination unit 15 and stores them. Next, the separation unit 17 receives information on the voltage, current, and power factor of the estimation target at each time measured by the sensor 3 from the sensor 3. Then, the separation unit 17 performs vector decomposition on the received information on the voltage, current, and power factor to separately determine the output of the load 4 and the photovoltaic power generation 5 in the power distribution system 1. A method for calculating the output of the photovoltaic power generation 5 by the separation unit 17 will be described below in detail.

[0072] The separation unit 17 calculates active power and reactive power for each time using the acquired information on the measurement values ​​of the voltage, current, and power factor for each time of the estimation target. Next, the separation unit 17 calculates active power fluctuations and reactive power fluctuations for each time using the active power and reactive power for each time. Next, the separation unit 17 calculates the apparent power fluctuations and power flow fluctuation vector angle of the power flow for each time using the active power fluctuations and reactive power fluctuations for each time.

[0073] Next, the separation unit 17 identifies the quadrant in which the power flow fluctuation exists according to the magnitude of the power flow fluctuation vector angle and the combination of the active power fluctuation and the reactive power fluctuation. The separation unit 17 then calculates the PV output fluctuation for each time using an equation selected according to the identified quadrant and the values ​​of the load fluctuation power factor cosθ and the PV fluctuation power factor cosφ acquired from the power factor determination unit 15. Next, the separation unit 17 integrates the calculated PV output fluctuation for each time to obtain the output of the photovoltaic power generation unit 5.

[0074] 7 is a flowchart of the power factor estimation process by the power factor estimation device according to Example 1. Next, the flow of the power factor estimation process by the power factor estimation device 10 according to the present embodiment will be described with reference to FIG.

[0075] The power calculation unit 11 acquires combination data of the measurement time for each day during the sensor data period and the measurement values ​​of the voltage, current, and power factor from the sensor data file 161 stored in the memory unit 16. Next, the power calculation unit 11 calculates the active power and reactive power for each measurement time for each day using the read combination data (step S1). Then, the power calculation unit 11 outputs power data in which the calculated active power and reactive power are associated with the date and measurement time to the power flow maximum value identification unit 12, the first power generation output calculation unit 13, and the second power generation output calculation unit 14.

[0076] Next, power flow maximum value identifyr 12 receives an input of power data in which active power and reactive power are associated with dates and measurement times from power calculator 11. Next, power flow maximum value identifyr 12 uses the power data to identify the maximum power flow value for each time in the sensor data period (step S2).

[0077] Next, the first power generation output calculation unit 13 receives an input of power data in which active power and reactive power are associated with a date and a measurement time from the power calculation unit 11. Also, it receives an input of maximum power flow data combining each time of the day with the maximum power flow value from the power flow maximum specification unit 12. Next, the first power generation output calculation unit 13 selects a specific sunny day in the sensor data period. Next, the first power generation output calculation unit 13 acquires power data for the specific sunny day from the power data in which active power and reactive power are associated with a date and a measurement time. Next, the first power generation output calculation unit 13 uses the power data for the specific sunny day to determine the power flow value for each hour on the specific sunny day. Then, the first power generation output calculation unit 13 calculates a power flow difference value by subtracting the power flow value for each hour on the specific sunny day from the maximum power flow value for each hour represented by the power flow maximum data, and sets the PV output for each hour due to the maximum power flow (step S3).

[0078] The second power output calculation unit 14 obtains information on the specific sunny day from the first power output calculation unit 13. Next, the second power output calculation unit 14 obtains power data from the power calculation unit 11. The second power output calculation unit 14 then extracts power data representing active power and reactive power for each hour on the specific sunny day. Next, the second power output calculation unit 14 calculates the apparent power fluctuation ΔS of the tidal current from the active power fluctuation ΔP and the reactive power fluctuation ΔQ during a predetermined time interval Δt on the specific sunny day using equation (5). Next, the second power output calculation unit 14 determines the quadrant from the signs of the active power fluctuation ΔP and the reactive power fluctuation ΔQ, and selects the ΔP to be applied from the calculation formulas summarized in Table 1. G and ΔP L Next, the second power generation output calculation unit 14 performs vector decomposition with the load fluctuation rate using the PV fluctuation power factor cosφ as a constant, and calculates the PV output fluctuation ΔP G Then, the second power generation output calculation unit 14 calculates the PV output fluctuation ΔP GThe PV output for each time by vector decomposition with the PV output fluctuation power factor as a constant is calculated by accumulating the above values ​​(step S4). The PV output for each time by vector decomposition with the PV output fluctuation power factor as a constant is calculated by PV output P corresponding to each value of the load fluctuation power factor cosθ when the PV fluctuation power factor cosφ is a constant. G (T).

[0079] The power factor determination unit 15 receives information on the PV output at each time due to the maximum power flow from the first power generation output calculation unit 13. As a first stage of power factor determination, the power factor determination unit 15 also receives information on the PV output at each time by vector decomposition with the PV output fluctuation as a constant from the second power generation output calculation unit 14. Then, the power factor determination unit 15 compares the PV output at each time due to the maximum power flow with the PV output at each time calculated by vector decomposition with the PV output fluctuation power factor as a constant, and selects the load fluctuation power factor when the correlation between the PV outputs is highest (step S5).

[0080] The second power generation output calculation unit 14 receives the determined value of the load fluctuation power factor cosθ from the power factor determination unit 15. Next, the second power generation output calculation unit 14 performs vector decomposition with the PV fluctuation power factor cosφ using the load fluctuation power factor cosθ determined by the power factor determination unit 15, and calculates the PV output fluctuation ΔP G Then, the second power generation output calculation unit 14 calculates the PV output fluctuation ΔP G The PV output for each time by vector decomposition using the determined load fluctuation power factor is calculated by accumulating the PV output P corresponding to each value of the PV fluctuation power factor cosφ when the determined load fluctuation power factor cosθ value is used. G (T).

[0081] The power factor determination unit 15 receives information on the PV output for each time by vector decomposition using the determined load fluctuation power factor from the second power generation output calculation unit 14. Next, the power factor determination unit 15 calculates the difference between the PV output for each time by vector decomposition using the determined load fluctuation power factor and the PV output for each time by maximum power flow, based on the active power for each time slice. Then, the power factor determination unit 15 selects the PV fluctuation power factor cosφ that minimizes the sum of the differences (step S7).

[0082] The separation unit 17 acquires the load fluctuation power factor cosθ and the PV fluctuation power factor cosφ selected by the power factor determination unit 15. After that, the separation unit 17 receives information on the measured voltage, current, and power factor of the estimation target at each time from the sensor 3. Then, the separation unit 17 performs vector decomposition of the power flow in the power distribution system 1 using the acquired load fluctuation power factor cosθ and PV fluctuation power factor cosφ to obtain the output of the photovoltaic power generation 5 (step S8).

[0083] Fig. 8 is a flowchart of the vector decomposition process by the decomposition unit. Next, the flow of the vector decomposition process by the separation unit 17 will be described with reference to Fig. 8. The process described in the flow of Fig. 8 corresponds to an example of the process executed in step S8 of Fig. 7.

[0084] Then, the separation unit 17 sets a combination of the load fluctuation power factor cos θ and the PV fluctuation power factor cos φ for each pattern (step S11).

[0085] Next, the separator 17 calculates active power and reactive power from the voltage, current, and power factor of the estimation target measured by the sensor 3 at each time (step S12).

[0086] Next, the separator 17 calculates the active power fluctuation and the reactive power fluctuation (step S13).

[0087] Next, the separation unit 17 performs vector decomposition into a load fluctuation component and a PV output fluctuation component using the load fluctuation power factor cosθ and the PV fluctuation power factor cosφ selected by the power factor determination unit 15 (step S14). Specifically, the separation unit 17 identifies a quadrant in which the power flow fluctuation exists, and calculates the load fluctuation and the PV output fluctuation for each time using an equation selected according to the identified quadrant and the values ​​of the load fluctuation power factor cosθ and the PV fluctuation power factor cosφ acquired from the power factor determination unit 15.

[0088] Thereafter, the separation unit 17 accumulates the PV output fluctuations for each time and calculates the output of the photovoltaic power generation 5 (step S15). Similarly, the separation unit 17 accumulates the load fluctuations for each time and calculates the value of the load 4.

[0089] As described above, when calculating the PV fluctuation power factor and the load fluctuation power factor, the power factor estimating device according to the present embodiment performs vector decomposition with the load fluctuation power factor with the PV fluctuation power factor as a constant to determine the load fluctuation power factor. Next, the power factor estimating device performs vector decomposition with the PV fluctuation power factor using the determined load fluctuation power factor to determine the PV fluctuation power factor.

[0090] As a result, when performing vector decomposition, it is possible to determine the load fluctuation power factor and the PV fluctuation power factor by vector decomposition of the number of patterns of the load fluctuation power factor and the number of patterns of the PV fluctuation power factor without performing vector decomposition on combination patterns when both the load fluctuation power factor and the PV fluctuation power factor are changed. As a result, the power factor estimation device according to the present embodiment can significantly reduce the amount of calculation in the power factor determination process, and can reduce the processing load of the power factor estimation process. Therefore, by easily performing appropriate power factor estimation, it is possible to improve the accuracy of the decomposition and accurately grasp the net load amount minus the power generation amount, thereby improving the reliability of the power distribution system. EXAMPLES

[0091] Next, a second embodiment will be described. The power factor estimation device according to this embodiment is different from the first embodiment in that it estimates the power factor using reactive power. A power factor estimation device 10 according to this embodiment is also represented by a block diagram in FIG. 2. In the following description, the description of the functions of each unit similar to those of the first embodiment will be omitted.

[0092] In the conventional method of power factor estimation using vector decomposition, the correlation coefficient between the result of applying the vector decomposition method and the PV output at each time due to maximum power flow is calculated for each combination of the PV fluctuation power factor cosθ and the load fluctuation power factor cosφ. After that, combinations of power factors with large correlation coefficients are extracted, and the combination of the PV fluctuation power factor cosθ and the load fluctuation power factor cosφ that minimizes the total difference for each time section is selected as the optimal combination. In this method, the fluctuation pattern of active power is used as the basis for both the correlation coefficient and the difference between the estimated values.

[0093] In this case, if the accuracy of the hourly PV output due to maximum power flow is not good, the difference may be small in the case of a power factor that does not actually exist depending on the time section. The PV variable power factor cosθ is assumed to be in the range of 0.85 to 1.0, and in that case, the power factor that minimizes the difference is 0.85 or 1.0, and the integration of the difference may ultimately result in an estimated power factor that is not optimal.

[0094] This is thought to be due to the fact that the load conditions are not the same during the measurement period when the maximum power flow method is applied, and also because there are sections in which there is no or small fluctuation in the PV output.In such cases, the vectors of both the active power fluctuation and the reactive power fluctuation approach zero, so even a slight deviation will cause the estimated power factor to change significantly.

[0095] Therefore, the power factor estimating device 10 according to the present embodiment employs the fluctuation pattern of reactive power instead of active power as the basis for comparison.

[0096] In the vector decomposition method, both the PV fluctuation power factor cosθ and the load fluctuation power factor cosφ are assumed to be fixed throughout the day, so it is thought that there will be no large difference in the accuracy of the power factor estimation for active power and reactive power in terms of the correlation coefficient. On the other hand, as the estimated value itself is handled, there will be a difference in the accuracy of the power factor estimation for active power and reactive power in terms of the difference in the estimated value.

[0097] PV output fluctuation ΔP based on active power when the power flow fluctuation in the vector decomposition method is in the first quadrant G is shown in Table 1. In this case, the PV output fluctuation ΔQ based on the reactive power G is calculated by the following formula (10). Hereinafter, the PV output fluctuation based on the active power is referred to as the PV output fluctuation of the active power fluctuation, and the PV output fluctuation based on the reactive power is referred to as the PV output fluctuation of the reactive power fluctuation.

[0098]

number

[0099] Here, the reactive power fluctuations corresponding to the active power fluctuations in each quadrant are summarized in Table 2 below.

[0100] [Table 2]

[0101] ΔP G and ΔQ G The difference between these is cosφ and sinφ, and the rest are the same. When the PV variable power factor cosφ is in the range of 0.85 to 0.1, sinφ is in the range of 0 to 0.527. When the PV variable power factor cosφ is changed in increments of 0.01, sinφ changes in increments of approximately 0.017 to 0.141, and the interval between the estimated values ​​is ΔQ G will be larger.

[0102] In the second stage of power factor determination, the power factor determination unit 15 calculates the PV output P GThe information (T) is input from the second power generation output calculation unit 14.

[0103] Next, the power factor determination unit 15 calculates the difference between the PV output at each time by vector decomposition using the determined load fluctuation power factor and the PV output at each time by the maximum power flow, based on the reactive power at each time slice. Then, the power factor determination unit 15 determines the PV fluctuation power factor that minimizes the sum of the differences as the estimated result of the PV fluctuation power factor.

[0104] Fig. 9 is a diagram showing the results of estimating the power factor when the fluctuation pattern of the active power is used as a reference. Fig. 10 is a diagram showing the results of estimating the power factor when the fluctuation pattern of the active power is used as a reference. In both Figs. 9 and 10, the vertical axis represents power and the horizontal axis represents time. In both cases, the optimal PV variable power factor to be sought is 0.95.

[0105] In FIG. 9, graph 200 represents actual measured values, graph 201 represents the case where the power factor is 0.97, graph 202 represents the case where the power factor is 0.96, graph 203 represents the case where the power factor is 0.95, and graph 204 represents the case where the power factor is 0.94.

[0106] For example, as shown in Fig. 9, if the difference is taken based on the fluctuation pattern of the active power, it is affected by the power factor fluctuation of the PV output at each time due to the maximum power flow, and the power factor determination unit 15 may erroneously estimate a power factor of 0.96. This is because the intervals between the graphs 201 to 204, as represented by the width W1, are short, and even a slight deviation from the actual measurement value makes it highly likely that the power factor will be determined to be erroneous.

[0107] In FIG. 10, graph 210 represents actual measured values, graph 211 represents the case where the power factor is 0.97, graph 212 represents the case where the power factor is 0.96, graph 213 represents the case where the power factor is 0.95, and graph 214 represents the case where the power factor is 0.94.

[0108] On the other hand, when the difference is calculated based on the reactive power fluctuation pattern as shown in Fig. 10, the intervals between the graphs 211 to 214, as represented by the width W2, are wider than when the active power fluctuation pattern is used as the reference. Therefore, the influence of the error of the actual measurement value on the judgment is reduced, and the possibility of estimating the optimal PV variable power factor is increased. For example, in the example of Fig. 10, the power factor determination unit 15 can estimate the PV variable power factor to be 0.95.

[0109] 11 is a flowchart of the power factor estimation process performed by the power factor estimation device according to the embodiment 2. Next, the flow of the power factor estimation process performed by the power factor estimation device 10 according to the embodiment will be described with reference to FIG.

[0110] The power calculation unit 11 acquires combination data of the measurement time for each day during the sensor data period and the measurement values ​​of the voltage, current, and power factor from the sensor data file 161 stored in the memory unit 16. Next, the power calculation unit 11 calculates the active power and reactive power for each measurement time for each day using the read combination data (step S101). Then, the power calculation unit 11 outputs power data in which the calculated active power and reactive power are associated with the date and measurement time to the power flow maximum value identification unit 12, the first power generation output calculation unit 13, and the second power generation output calculation unit 14.

[0111] Next, power flow maximum value identifyr 12 receives an input of power data in which active power and reactive power are associated with dates and measurement times from power calculator 11. Next, power flow maximum value identifyr 12 uses the power data to identify the maximum power flow value for each time in the sensor data period (step S102).

[0112] Next, the first power generation output calculation unit 13 receives an input of power data in which active power and reactive power are associated with a date and a measurement time from the power calculation unit 11. Also, the first power generation output calculation unit 13 receives an input of maximum power flow data, which is a combination of each time of day and the maximum power flow value, from the power flow maximum specification unit 12. Next, the first power generation output calculation unit 13 selects a specific sunny day in the sensor data period. Next, the first power generation output calculation unit 13 acquires power data for the specific sunny day from the power data in which active power and reactive power are associated with a date and a measurement time. Next, the first power generation output calculation unit 13 uses the power data for the specific sunny day to determine the power flow value for each hour on the specific sunny day. Then, the first power generation output calculation unit 13 calculates a power flow difference value by subtracting the power flow value for each hour on the specific sunny day from the maximum power flow value for each hour represented by the power flow maximum data, and sets the calculated power flow difference value as the PV output for each hour due to the maximum power flow (step S103).

[0113] The second power output calculation unit 14 obtains information on the specific sunny day from the first power output calculation unit 13. Next, the second power output calculation unit 14 obtains power data from the power calculation unit 11. The second power output calculation unit 14 then extracts power data representing active power and reactive power for each hour on the specific sunny day. Next, the second power output calculation unit 14 calculates the apparent power fluctuation ΔS of the tidal current from the active power fluctuation ΔP and the reactive power fluctuation ΔQ during a predetermined time interval Δt on the specific sunny day using equation (5). Next, the second power output calculation unit 14 determines the quadrant from the signs of the active power fluctuation ΔP and the reactive power fluctuation ΔQ, and selects the ΔP to be applied from the calculation formulas summarized in Table 1. G and ΔP L Next, the second power generation output calculation unit 14 performs vector decomposition with the load fluctuation rate using the PV fluctuation power factor cosφ as a constant, and calculates the PV output fluctuation ΔP G Then, the second power generation output calculation unit 14 calculates the PV output fluctuation ΔP G are accumulated to calculate the PV output for each time by vector decomposition with the PV output fluctuation power factor as a constant (step S104).

[0114] The power factor determination unit 15 receives information on the PV output at each time due to the maximum power flow from the first power generation output calculation unit 13. As a first stage of power factor determination, the power factor determination unit 15 also receives information on the PV output at each time by vector decomposition with the PV output fluctuation as a constant from the second power generation output calculation unit 14. Then, the power factor determination unit 15 compares the PV output at each time due to the maximum power flow with the PV output at each time calculated by vector decomposition with the PV output fluctuation power factor as a constant, and selects the load fluctuation power factor when the correlation between the PV outputs is highest (step S105).

[0115] The second power generation output calculation unit 14 receives the determined value of the load fluctuation power factor cosθ from the power factor determination unit 15. Next, the second power generation output calculation unit 14 performs vector decomposition with the PV fluctuation power factor cosφ using the load fluctuation power factor cosθ determined by the power factor determination unit 15, and calculates the PV output fluctuation ΔP G Then, the second power generation output calculation unit 14 calculates the PV output fluctuation ΔP G are accumulated to calculate the PV output for each time by vector decomposition using the determined load fluctuation power factor (step S106).

[0116] The power factor determination unit 15 receives information on the output of the photovoltaic power generation unit 5 at each time by vector decomposition using the determined load fluctuation power factor from the second power output calculation unit 14. Next, the power factor determination unit 15 calculates the difference between the PV output at each time by vector decomposition using the determined load fluctuation power factor and the PV output at each time by the maximum power flow, based on the reactive power at each time slice. Then, the power factor determination unit 15 selects the PV fluctuation power factor cosφ that minimizes the sum of the differences (step S107).

[0117] The separation unit 17 acquires the load fluctuation power factor cosθ and the PV fluctuation power factor cosφ selected by the power factor determination unit 15. After that, the separation unit 17 receives information on the measured voltage, current, and power factor of the estimation target at each time from the sensor 3. Then, the separation unit 17 performs vector decomposition of the power flow in the power distribution system 1 using the acquired load fluctuation power factor cosθ and PV fluctuation power factor cosφ to obtain the output of the photovoltaic power generation 5 (step S108).

[0118] As described above, when estimating the PV variable power factor, the power factor estimating device according to the present embodiment obtains the difference between the PV output at each time by vector decomposition and the PV output at each time on a clear day by maximum power flow, with the reactive power at each time slice being the reference. Then, the power factor estimating device selects the PV variable power factor with the smallest sum of the differences with the reactive power as the estimation result.

[0119] As a result, the granularity of the PV output corresponding to the PV variable power factor becomes coarser and larger variations are allowed, compared to when estimating the PV power factor fluctuation based on the active power. Therefore, the effect of errors in the actual measurement value is suppressed, and it becomes possible to estimate the PV variable power factor with greater accuracy. In other words, by improving the accuracy of the decomposition, it is possible to accurately grasp the net load amount minus the amount of power generation, and improve the reliability of the power distribution system.

[0120] (Hardware configuration) 12 is a hardware configuration diagram of the power factor estimation device 10. The power factor estimation device 10 includes, for example, a CPU (Central Processing Unit) 91, a memory 92, a hard disk 93, and a network interface 94. The CPU 91 is connected to the memory 92, the hard disk 93, and the network interface 94 via a bus.

[0121] The network interface 94 is a communication interface for communication between the power factor estimation device 10 and an external device. The network interface 94 relays communication between the CPU 91 and the sensor 3, for example.

[0122] The hard disk 93 is an auxiliary storage device. The hard disk 93 realizes, for example, the function of the storage unit 16 illustrated in Fig. 2. The hard disk 93 also holds various programs including programs that realize the functions of the power calculation unit 11, the power flow maximum value identification unit 12, the first power generation output calculation unit 13, the second power generation output calculation unit 14, the power factor determination unit 15, and the separation unit 17 illustrated in Fig. 1.

[0123] The CPU 91 reads out various programs from the hard disk 93, and deploys and executes them on the memory 92, which is a main storage device. In this way, the CPU 91 realizes the functions of the power calculation unit 11, the power flow maximum value identification unit 12, the first power generation output calculation unit 13, the second power generation output calculation unit 14, the power factor determination unit 15, and the separation unit 17, which are illustrated in FIG. [Explanation of symbols]

[0124] 1 Power distribution system 2. Distribution substations 3 Sensors 4. Load 5. Solar power generation 10 Power factor estimator 11 Power Calculation Section 12 Maximum current determination section 13 First power generation output calculation unit 14 Second power generation output calculation section 15 Power factor determination section 16 Memory section 17 Separation part 161 Sensor Data File

Claims

1. a first power generation output calculation unit that calculates a photovoltaic power generation output at each time due to a maximum power flow based on a maximum power flow value at each time during a predetermined period in the past; a second power generation output calculation unit that calculates a photovoltaic power generation output for each hour by first vector decomposition based on active power fluctuations and reactive power fluctuations on clear days included in the specified past period, with a value of an operating power factor of photovoltaic power generation set as a constant, obtains a value of a load power factor specified based on the photovoltaic power generation output for each hour by the first vector decomposition, and calculates a photovoltaic power generation output for each hour by second vector decomposition based on the active power fluctuations and the reactive power fluctuations, with the value of the load power factor set as the specified value; a power factor determination unit that specifies a value of a power factor of the load based on the photovoltaic power generation output at each hour due to the maximum power flow and the photovoltaic power generation output at each hour due to the first vector decomposition, and specifies an operating power factor of the photovoltaic power generation based on the photovoltaic power generation output at each hour due to the maximum power flow and the photovoltaic power generation output at each hour due to the second vector decomposition; A power factor estimation device comprising:

2. the first power generation output calculation unit calculates a photovoltaic power generation output for each hour due to the maximum power flow by subtracting the power flow value for each hour on the clear day from a maximum power flow value for each hour in the past predetermined period; The second power generation output calculation unit calculates the photovoltaic power generation output for each time by the first vector decomposition using the multiple values ​​of the load power factor, for each value of the load power factor, based on the active power fluctuations and the reactive power fluctuations between times that are separated by a predetermined time interval on the clear day, with the value of the operating power factor of the photovoltaic power generation set as a constant, obtains a value of the load power factor specified based on the photovoltaic power generation output for each time by the first vector decomposition, and calculates the photovoltaic power generation output for each time by the second vector decomposition using the multiple values ​​of the photovoltaic power generation operating power factor, for each value of the photovoltaic power generation operating power factor, based on the active power fluctuations and the reactive power fluctuations, with the value of the load power factor set as the specified value. The power factor determination unit compares the photovoltaic power generation output at each time due to the maximum power flow with the photovoltaic power generation output at each time by the first vector decomposition for each value of the power factor of the load, and identifies a value of the power factor of the load with which the correlation is highest, and compares the photovoltaic power generation output at each time due to the maximum power flow with the photovoltaic power generation output at each time by the second vector decomposition for each value of the operating power factor of the photovoltaic power generation, and identifies the operating power factor of the photovoltaic power generation with which the difference is smallest.

2. The power factor estimating device according to claim 1, further comprising:

3. The power factor estimation device according to claim 2, characterized in that the power factor determination unit calculates a difference at each time between the photovoltaic power generation output at each time due to the maximum power flow and the photovoltaic power generation output at each time due to the second vector decomposition for each value of the operating power factor of the photovoltaic power generation, and identifies a value of the operating power factor of the photovoltaic power generation that minimizes a sum of the calculated differences.

4. 4. The power factor estimation device according to claim 3, wherein the power factor determination unit calculates the difference for each time period using reactive power as a reference.

5. The second power generation output calculation unit, using a value of the operating power factor of the photovoltaic power generation as the constant, calculating a photovoltaic power generation output fluctuation during the predetermined time interval based on a power flow apparent power fluctuation and a power flow fluctuation vector angle based on an active power fluctuation and a reactive power fluctuation between times preceding and succeeding by a predetermined time interval on the clear day, and a plurality of values ​​of a load power factor, and calculating a photovoltaic power generation output for each time by the first vector decomposition using the calculated photovoltaic power generation output fluctuation, for each value of the load power factor; The value of the load power factor is set as a specified value, and based on the apparent power fluctuation and the power flow fluctuation vector angle of the power flow based on the active power fluctuation and the reactive power fluctuation between times preceding and succeeding by a specified time interval on the sunny day and multiple values ​​of the operating power factor of the solar power generation, a solar power generation output fluctuation for the specified time interval is calculated, and the calculated solar power generation output fluctuation is used to calculate the solar power generation output for each time by a second vector decomposition for each value of the operating power factor of the solar power generation.

5. The power factor estimation device according to claim 2, wherein the power factor estimation device is a power factor estimation device.

6. Calculating the solar power generation output for each hour due to the maximum tidal current based on the maximum tidal current value for each hour in a specified period in the past; calculating a photovoltaic power generation output for each hour by a first vector decomposition based on the active power fluctuation and the reactive power fluctuation on a clear day included in the predetermined period in the past, with the value of an operating power factor of the photovoltaic power generation being a constant; determining a value of a power factor of a load based on the hourly photovoltaic power generation output due to the maximum power flow and the hourly photovoltaic power generation output due to the first vector decomposition; Calculating a time-based photovoltaic power generation output by a second vector decomposition, assuming a power factor value of the load as a specified value based on the active power fluctuation and the reactive power fluctuation; Identifying an operating power factor of the photovoltaic power generation based on the photovoltaic power generation output for each hour due to the maximum power flow and the photovoltaic power generation output for each hour due to the second vector decomposition. A power factor estimation method comprising:

7. Calculating the solar power generation output for each hour due to the maximum tidal current based on the maximum tidal current value for each hour in a specified period in the past; calculating a photovoltaic power generation output for each hour by a first vector decomposition based on the active power fluctuation and the reactive power fluctuation on a clear day included in the predetermined period in the past, with the value of an operating power factor of the photovoltaic power generation being a constant; determining a value of a power factor of a load based on the hourly photovoltaic power generation output due to the maximum power flow and the hourly photovoltaic power generation output due to the first vector decomposition; Calculating a time-based photovoltaic power generation output by a second vector decomposition, assuming a power factor value of the load as a specified value based on the active power fluctuation and the reactive power fluctuation; Identifying an operating power factor of the photovoltaic power generation based on the photovoltaic power generation output for each hour due to the maximum power flow and the photovoltaic power generation output for each hour due to the second vector decomposition. A power factor estimation program for causing a computer to execute a process.

Images