JP6093465B1 - Power generation diagnosis method and power generation diagnosis apparatus for solar power generation system - Google Patents

Abstract

A power generation diagnosis method for a solar power generation system that enables power generation diagnosis for each component with high accuracy and in a short time. A method for diagnosing power generation of a solar power generation system constituted by a plurality of constituent elements, the step of acquiring information on the power generation amount of each constituent element, and each constituent element for the total power generation amount of the solar power generation system Power generation of each component using the step of calculating and storing the output ratio, the output ratio in the time zone to be diagnosed, and at least part of the output ratio for the past predetermined number of days measured in the same time zone And a step of determining an abnormality. [Selection] Figure 2

Description

  The present invention relates to a power generation diagnosis method and a power generation diagnosis apparatus for a photovoltaic power generation system.

  Efforts for renewable energy are becoming more active in order to improve problems such as depletion of fossil fuels, air pollution, and global warming. Renewable energy is a general term for energy that can be used permanently, and includes sunlight, wind power, hydropower, geothermal heat, biomass, and the like.

  The most common form of utilization of renewable energy is solar power generation. Photovoltaic power generation is a power generation method that directly converts solar energy into electric power using a solar cell that converts light energy into electricity using the photovoltaic effect. Since it is available anywhere and can convert inexhaustible solar energy into electric power cleanly without generating exhaust gas such as carbon dioxide, it is expected to be one of the powerful solutions for energy problems and environmental problems. In recent years, there has been a rapid expansion due to the promotion of government and other public institutions.

  One type of photovoltaic power generation system is a system constituted by a solar cell string (hereinafter also simply referred to as “string”) in which a plurality of solar cell panels are connected in series. Since solar cell strings are usually installed outdoors, solar cell panels and cables connecting them may be deteriorated or damaged due to the influence of ultraviolet rays, wind and rain, etc., which may cause power generation failure. However, due to circumstances such as the remote location of the solar power generation system, it is often impossible to check the solar cell string directly, and the amount of power generation may vary greatly depending on the amount of solar radiation. It has been difficult to quickly and accurately detect power generation failures caused by the power generation device.

  As a technique for solving such a problem, a technique has been proposed in which the output of solar cell strings is compared to determine the degree of divergence, and the degree of divergence is compared with a reference value to detect a string abnormality (patent). Reference 1).

JP 2012-104750 A

  However, according to the above method, since the failure is diagnosed by comparing the components of the photovoltaic power generation system such as the string, it is possible to diagnose which component is broken by only one comparison. However, there is a problem that it is difficult to identify a component considered to be a failure only after comparing all the components.

  Moreover, since each component with a large fluctuation | variation of electric power generation is compared, there existed a problem that a diagnosis for a short time would become difficult if diagnostic accuracy was not enough and it tried to raise diagnostic accuracy.

  The present invention has been made in view of the above, and an object of the present invention is to provide a power generation diagnosis method and a power generation diagnosis device for a solar power generation system capable of performing power generation diagnosis of components such as strings with high accuracy and in a short time. It is in.

A power generation diagnosis method for a photovoltaic power generation system according to the present invention is as follows.
A method for performing power generation diagnosis of a photovoltaic power generation system configured by a plurality of components by a computer,
The computer obtaining information on the power generation amount of each component;
Storing each output ratio for each of the components with respect to the total power generation amount of the photovoltaic power generation system, calculated from the power generation amount of each of the acquired components;
Calculating each normal range for each output fraction of each component using at least a part of the most recent of the respective output ratios of a predetermined past number of days in a zone diagnosed and ing time and same time zone,
Comparing each normal range and each output ratio to determine power generation abnormality.

Moreover, the power generation diagnostic device of the photovoltaic power generation system according to the present invention is:
A power generation diagnostic device for a solar power generation system configured by a plurality of components,
An acquisition unit for acquiring information on the power generation amount of each of the components;
A storage unit that stores each output ratio for each of the components with respect to the total power generation amount of the photovoltaic power generation system, calculated from the power generation amount of each of the acquired components.
The most recent past predetermined number of days in a diagnostic object and name Ru time zone and the time zone using at least a part of each output fraction was calculated each normal range for each output fraction of each component, the respective normal range And a determination unit that compares the output ratios with each other to determine power generation abnormality.

  According to such a configuration, the power generation diagnosis is performed based on the ratio of the power generation amount of each component to the power generation amount of the entire system using the power generation amount of the entire system that is more stable than the power generation amount of each individual component. Compared with the configuration in which the power generation amounts of the constituent elements are compared, it is possible to make a diagnosis with higher accuracy.

  In addition, in the configuration for comparing the power generation amounts of the components, it is difficult to identify which component is abnormal by only one comparison, but each component with respect to the power generation amount of the entire system According to the above configuration using the output ratio, in many cases, it is possible to detect whether each component has a power generation abnormality by a single determination.

  As described above, according to the power generation diagnosis method and power generation diagnosis apparatus for a photovoltaic power generation system according to the present invention, power generation diagnosis for each component can be performed with high accuracy in a short time.

It is explanatory drawing which shows typically the solar power generation system with which power generation diagnosis is performed by the power generation diagnostic apparatus concerning one Embodiment of this invention. It is a lineblock diagram of the power generation diagnostic device concerning one embodiment of the present invention. It is a flowchart which shows the process sequence of the electric power generation diagnostic apparatus concerning one Embodiment of this invention. It is an example of the output ratio in a normal solar power generation system. It is an example of the output ratio of the solar power generation system including the power generation abnormality to which the power generation diagnosis of the present embodiment is effectively applied.

  Hereinafter, the power generation diagnostic device of the solar power generation system according to the present embodiment will be described in detail. However, all of the following embodiments are examples, and do not give a limited interpretation of the present invention. It should be noted that the drawings are drawings for explaining the embodiments according to the present invention, and the size of each component may not always be accurate.

  FIG. 1 is an explanatory diagram schematically showing a solar power generation system 1 to which the power generation diagnosis method according to the present embodiment is applied. The solar power generation system 1 includes a string 12 (12a, 12b, 12c) in which a plurality of solar battery panels 11 including a plurality of solar battery cells 10 are connected in series, a power conditioner 20 to which each string 12 is connected, Is provided. The DC power generated by the solar battery cell 10 is converted into AC by the power conditioner 20 and supplied to the power supply system 30 such as a commercial system.

  The photovoltaic power generation system 1 shown in FIG. 1 is a system called a multi-string method (multiple-input method) using a power conditioner 20 having a boosting function with respect to the input of each string 12.

  In the multi-string method, the input voltage is controlled for each string 12 and can be directly connected to the power conditioner 20 without using an intermediate device such as a junction box or a boosting unit required for the collective input method. There is a merit that it becomes easy.

  In contrast, in the collective input method in which the strings 12 are connected to the power conditioner 20 via a connection box that aggregates the strings 12, each string 12 is connected to the connection box via a booster unit as necessary. In the present embodiment, the multi-string solar power generation system 1 will be described. However, the power generation diagnostic apparatus according to the present embodiment can be any system as long as the output of each component such as a string is measured. It can also be applied to. Note that “output” in this specification refers to generated power, and may be generated current when a predetermined condition described later is satisfied.

  The solar battery cell 10 is generally in a square shape of about 10 to 15 cm, and converts light energy into electric power by utilizing the photovoltaic effect.

  The solar battery panel 11 is configured by collecting a plurality of solar battery cells 10. The general solar cell panel 11 is packaged in a state in which the surface is protected by tempered glass and the back surface is protected by a weather-resistant resin called a back sheet, and the inside is filled with a transparent sealing resin and shielded from the outside air It is stored in. The solar cell panel 11 is also referred to as a solar cell module.

  The solar cell string 12 is configured such that a plurality of solar cell panels 11 are connected in series to obtain a desired voltage.

  The power conditioner 20 includes a DC / DC converter 21 that boosts the voltage of the string 12 to a desired value, and an inverter 22 that converts DC power into AC power. In addition to this, it has a protection function necessary for connecting to the grid 30 and a power generation control function described later.

  Since the power conditioner 20 in this embodiment includes DC / DC converters 21a, 21b, and 21c for the inputs of the strings 12a, 12b, and 12c, the number of connections of the solar cell panels 11 is different for each string 12. The number may be adjusted as appropriate according to the installation environment.

  The DC / DC converter 21 of the power conditioner 20 in the present embodiment has a power generation control function. In the photovoltaic power generation system 1 according to the present embodiment, MPPT (Maximum Power Point Tracking) control is used. The MPPT control is a control technique for causing each solar cell panel to output the maximum amount of power generation, and the power generation is controlled so that the current and voltage of the solar cell panel 11 become optimum values. In the multi-string method, MPPT control is performed for each string 12 and the output can be optimized, so that power generation efficiency can be improved. As the MPPT control, for example, a known technique such as a mountain climbing method can be used.

  Since the power generation amount of each string 12 is substantially proportional to the light energy incident on the solar cell panel 11, there is a strong correlation between the power generation amount and the amount of solar radiation. The amount of power generation according to the amount of solar radiation cannot be obtained. Thus, it is conceivable to determine the failure of the string using a solar radiation meter, but in this case, the solar power generation system 1 must be newly provided with a solar radiation meter.

  Moreover, the situation where sunlight is regularly blocked | interrupted by the influence of the building in the vicinity of the solar cell panel 11, a telephone pole, etc. is also assumed. Here, since the sun moves with time, the influence of the shadow caused by the obstacle also changes depending on the time zone. For this reason, there may be a time zone in which power generation is normally performed, or there may be a partial time zone in which power generation is not possible at all. Depending on the location of the solar radiation meter, it may be difficult to perform accurate power generation diagnosis.

  In order to solve such a problem, the power generation diagnosis apparatus according to the present embodiment performs power generation diagnosis using the output ratio of each string 12 with respect to the total power generation amount of the system 1.

  The power generation diagnostic device 40 according to the present embodiment includes the acquisition unit 41 that acquires information on the power generation amount of the component such as the string 12 and the total power generation amount of the system 1 from the power generation amount of each component acquired by the acquisition unit. The storage unit 42 that calculates and stores the output ratio of each component with respect to the above, the output ratio in the time zone to be diagnosed, and at least part of the output ratio for the past predetermined number of days in the same time zone stored in the storage unit And a determination unit 43 that determines power generation abnormality of each component (see FIG. 2). The power generation diagnostic device 40 can be connected to the photovoltaic power generation system 1 by electrical or electromagnetic communication means.

  Hereinafter, the processing procedure performed by the power generation diagnosis apparatus according to the present embodiment will be described with reference to the flowchart shown in FIG.

  The solar power generation system 1 includes measurement means for measuring the power generation amount of each string 12 as a component. In the power generation diagnosis device according to the present embodiment, first, the acquisition unit acquires information on the power generation amount of each string 12 (step S10). The power generation amount here refers to DC power, but a direct current may be used as the power generation amount when a predetermined condition described later is satisfied.

Next, from the power generation amount of each string 12, the total power generation amount P of the system 1 and the output ratio C _i of each string 12 with respect to the total power generation amount P of the system 1 are calculated and stored in the storage unit (step) S20).

Here, the total power generation amount P of the system 1 is calculated by the following equation 1.
[Formula 1]
P = P ₁ + P ₂ +... + P _n
n: Number of strings
P: Total power generated by the system (DC power)

Further, the output ratio C _i of each string 12 is calculated by the following equation 2. The following formula is a calculation formula when the multi-string solar power generation system 1 is assumed.
[Formula 2]
C _i = P _i / P
P _i : power generation amount of the i-th string (DC power)
C _i : Output ratio of the i-th string to the total power generation amount of the system

Note that in equations 1 and 2 above, but using the generated power as power generation amount P _i of total power generation amount P and the string 12 of the system 1, if the voltage condition is approximately equivalent with the generated current Also good.

The acquisition of the power generation amount P _i of each string 12 and the calculation of the output ratio C _i for each string 12 in steps S10 and S20 are preferably performed at regular intervals, especially every 1 minute to several hours. Is preferred. In addition, when acquisition of the power generation amount P _i of each string 12 is performed in a short cycle, the output ratio C _i for each string 12 is calculated using the integrated value or average value of P _i for each fixed time. However, the data processing load can be reduced by doing in this way.

Whether the output ratio C _{i for} each string 12 calculated in step S20 is classified and stored in the storage unit for each predetermined time period (hereinafter referred to as a diagnosis time period) for power generation diagnosis. Alternatively, it is stored in the storage unit together with time information at the time of measurement, and is later classified for each diagnosis time zone so that data processing can be performed. Further, instead of storing the output ratio C _i for each string 12, the power generation amount P _i for each string 12 is stored, and the output ratio C _i for each string 12 is calculated whenever necessary. May be.

  In the calculation formulas of the above formulas 1 and 2, the string 12 has been described as a component in the system 1. However, for example, in a photovoltaic power generation system having a plurality of power conditioners, the component is a power conditioner. You can also. In this case, it is possible to perform power generation diagnosis in units of power conditioners by replacing the above-mentioned Pi with the output of each power conditioner and replacing P with the output of the entire system constituted by a plurality of power conditioners. . Further, when a plurality of power plants are located in the same area, the entire individual power plants may be regarded as components. For example, in the case of an eco-town or the like in which houses equipped with a solar power generation system are gathered, if the present invention is applied with individual dwelling units as constituent elements, the dwelling unit in which power generation abnormality has occurred can be quickly identified, and Inspection and maintenance of the solar power generation system can be carried out.

  In addition, the component can be set to any of strings, arrays in which multiple strings are connected in parallel, power conditioners, and so on. The power generation diagnostic apparatus according to the present embodiment can be applied by taking the string connected as a component and the system as a system.

  Thus, as long as the power generation amount of each component can be measured, the power generation diagnostic device according to the present embodiment can be applied regardless of how the system and the components configuring the system are set. .

  It should be noted that the components such as the string 12 do not necessarily have to be installed adjacent to each other, and may be installed separately as long as the solar radiation amount conditions are substantially equal.

Next, the standard deviation is calculated from the output ratio C _i for the past predetermined number of days in the predetermined diagnosis time zone (step S30). The power generation diagnosis is performed every time unit (1 minute to several hours) with a predetermined width. Since it is a diagnosis of photovoltaic power generation, it is obvious that the time zone to be diagnosed is daytime, but for example, when diagnosis is performed in units of 10 minutes, 9:10 to 9:20, 9:20 to Power generation diagnosis is performed for each diagnosis time period such as 9:30, 9:30 to 9:40,.

If the time unit for performing power generation diagnosis is too short, the calculation load increases. Further, since the easily occur misjudgment due to noise (variations in the data), it is necessary to take measures such as to loosely determination to expand the width of deviation value T _i to the normal range. On the other hand, if the time unit for performing power generation diagnosis is too long, the information will be diluted if there is a slight decrease in output due to power generation abnormality to be detected. Defects such as inability to detect occur. In particular, the DC / DC converter 21 of the power conditioner 20 according to the present embodiment has an MPPT control function. However, when this function fails, the output suddenly drops and recovers in the initial stage, and then the output gradually decreases. There is a tendency to continue. Therefore, if the time unit for performing power generation diagnosis is too long, the initial failure of the MPPT control function cannot be detected, and abnormality detection is delayed. For these reasons, the time unit for performing power generation diagnosis is preferably 3 minutes to 30 minutes, and more preferably 5 minutes to 20 minutes. This makes it possible to detect power generation abnormality accurately and in a short time.

  As an example, in a multi-string solar power generation system in which three strings 12 are connected to one power conditioner 20 and MPPT control is performed for each string 12, the string output observed when the system is normal FIG. 4 shows the change in the ratio, and FIG. 5 shows the change in the string output ratio observed in the initial stage of the abnormality in the MPPT control of the second system string 12 in the same system.

4, the output ratio of 5 corresponds to C _i of the present invention. When the system is normal, the output ratio of each string is generally stable as shown in FIG. 4, but when an abnormality occurs in the MPPT control, the initial abnormality is as shown in FIG. The output drops and recovers repeatedly for a short time and with a low frequency. The power generation diagnosis according to the present invention is performed by detecting the difference between the normal output ratio as shown in FIG. 4 and the abnormal output ratio as shown in FIG. When the time is set longer than the time, a slight decrease in output as shown in FIG. 5 becomes an error when viewed from the whole, so that it is difficult to detect. In such a setting, this abnormality is detected only after several days to several weeks have passed since the abnormality first occurred and the output drop becomes normal. In order to quickly detect the abnormality in the early stage of occurrence, it is necessary to set the time unit for performing power generation diagnosis to be shorter than several hours.

The past data range setting in step S30 will be described. The power generation diagnostic apparatus according to the present embodiment is configured to determine the normality / abnormality of power generation by comparing the output ratio C _i of each component for each diagnostic time zone with past data, that is, the past in the same diagnostic time zone. A power generation diagnosis is performed by detecting a discontinuous change in the output ratio C _i from. Here, the range to be compared as the past data needs to be set in consideration of the change of the season, and as a guideline, the range is 1 day to 90 days from the previous day.

The seasonal change mentioned above takes into consideration that the height of the sun changes according to the seasonal change. When the height of the sun changes, the time zone where the shadow affects may change depending on the positional relationship between the obstacle and the solar cell panel 11. In addition, since the surface of the solar cell panel 11 may not be a perfect plane but may be bent, the output ratio C _i of each string 12 is affected by a change in the solar radiation angle due to a change in the height of the sun. Can also be considered.

Therefore, from the viewpoint of reducing the influence of seasonal changes, it is preferable that the period used as past data is short. On the other hand, if this period is too short, the number of data used as past data is reduced, and the reliability of standard deviation σ _i described later is lowered. Therefore, the range compared as past data is preferably, for example, about 3 to 60 days up to the previous day, and more preferably 7 to 30 days up to the previous day. The past data used for the diagnosis does not necessarily include the measurement values for the previous day, and only the highly reliable data in this period may be extracted and used.

Next, calculation of the standard deviation σ _i in step S30 will be described. The standard deviation σ _i is calculated from the past data of the output ratio C _i of each string 12 for each diagnosis time period. For example, when the output ratio C _i is calculated every minute and the power generation diagnosis is performed in units of 10 minutes (for example, a time zone of 9:30 to 9:40) using data for the past 30 days, this is required. The number of past data in the diagnosis time zone is C _i value per minute × 10 minutes × 30 days = 300 records, and the standard deviation σ _i may be calculated based on this.

Next, using the standard deviation σ _i obtained in step S30, a deviation value T _i is calculated from the actual measured value of the current output ratio C _i (step S40). In the calculation of the deviation value T _i , the representative value of the output ratio C _i in the diagnosis time zone is used as the actual measurement value of the current output ratio C _i . Representative values, addition can be used an average value of a plurality of output ratios C _i which is measured on the diagnostic time period, it may be used median like.

Next, the determination unit performs power generation diagnosis of each string 12 using the deviation value T obtained in step S40 (step S50). Normal range of deviation T _i the (upper and / or lower limit) is set in advance, if they fit deviation T _i calculated in step S40 is in the range, the power generation of the diagnosis time zone determines "normal" (Step S51), if it is off, it is determined as “abnormal” (Step S52).

The normal range of the deviation value T _i can be set as appropriate based on past statistics, simulations, or the like, and may be changed depending on the diagnosis time zone. For example, since the front sunrise and after sunset easily varied without solar radiation is stable, if spread normal range of their diagnosis time zone, an abnormality by the output ratio C _i of the noise variation of solar radiation results determined The fear of being done can be suppressed. Similarly, the normal range in the diagnostic time zone when the sun is in the middle may be set narrower than the normal range in other diagnostic time zones. Furthermore, you may set so that a normal range may be changed continuously according to the angle of the sun for every diagnostic time slot | zone. In this way, power generation diagnosis can be performed continuously while maintaining optimum diagnosis accuracy for each time zone.

The power generation diagnosis apparatus according to the present embodiment makes a determination using the output ratio C _i of each string 12 with respect to the total power generation amount P of the system 1. The most probable failure mode is when the output of one string decreases. In this case, the output ratio of such a string decreases and the output ratio of the other strings increases relatively slightly. . Therefore, many of the strings determined as “abnormal” are strings whose deviation value T _i is below the lower limit of the normal range.

However, since the deviation value T _i may possibly exceed the upper limit of the normal range, it is desirable to set an upper limit as well. As such a situation, for example, in the case of snowfall or the like, it is conceivable that most of the strings 12 constituting the system 1 are covered with obstacles, and only the remaining part of the strings 12 generates power normally. In this case, since the power generation amount of the entire system 1 is greatly reduced, the output value of the string 12 that normally generates power increases, and thus the deviation value T _i exceeds the upper limit of the normal range, and the abnormality may be found. obtain.

In the above description, the diagnosis method for obtaining the representative value (average value) of the output ratio C _{i for} the diagnosis time zone of today and obtaining the deviation value T _i from the representative value (average value) has been described. However, the value of the output ratio C _i is updated. The power generation diagnosis may be performed by obtaining the deviation value T _i from the latest measured value every time. By doing in this way, although measurement load increases, it becomes possible to perform a power generation diagnosis in real time, and a power generation abnormality can be detected earlier. In this case, in order to reduce the possibility of erroneous determination due to measurement noise, when the deviation value T _i outside the normal range continues for a predetermined number of times, or the deviation value T _i outside the normal range is measured a predetermined number of times within a certain time. An abnormality determination may be made only in cases.

  In addition, when it determines with "abnormal" as a result of the power generation diagnosis in a determination part (step S52), you may make it not use for calculation of future standard deviation by making the measured value into an abnormal value. In this way, the standard deviation can be calculated using only past data when power generation is normal, and the diagnostic accuracy can be improved. When the cause of the abnormality is resolved, the normal return can be determined more accurately.

Next, the restriction | limiting of the power generation diagnosis according to the condition of the individual solar power generation system 1 is demonstrated. As described above, there are cases where the solar power generation system 1 is affected by the surroundings every day at a certain time period, and typical examples include surrounding buildings and telephone poles. The influence of shadows is considered. The power generation diagnosis apparatus according to the present embodiment enables accurate and short-time power generation diagnosis even if these periodic obstacles are affected, but other faults that adversely affect the diagnostic accuracy. In the system 1 in which power generation is assumed, power generation diagnosis is performed only in an arbitrary time zone set in advance (for example, 10:00 to 11:00), or power generation diagnosis is performed in an arbitrary time zone set in advance. May not be performed. Further, in a time zone in which an influence is expected, the normal range of the deviation value T _i can be set loose so that continuous power generation diagnosis can be performed to the extent that no erroneous diagnosis is performed.

Hours of such determination excluded, for example, statistically determined using known numerical analysis techniques the period during which data of the obtained output ratio C _i is discontinuous, the object of the generator diagnose this time period You may make it exclude from a time slot | zone. Although it is obvious that it is not suitable for power generation diagnosis at night, the amount of solar radiation is not stable in the time zone close to sunrise or sunset, so the time zone in which the solar radiation amount is stable without individual circumstances, for example, 10: 00 to 00 The power generation diagnosis can be performed only during 15:00 to improve the diagnosis accuracy.

In addition, when there is a clear lack of solar radiation, the power generation diagnosis may be suspended. This is because, under the condition of insufficient solar radiation, the amount of power generation decreases and is not stable, so performing power generation diagnosis under such circumstances affects the diagnostic accuracy. When the solar power generation system 1 measures the solar radiation amount using a solar radiation meter or the like, the power generation diagnosis may not be performed when the solar radiation amount is a predetermined value or less. Thus, it is possible to avoid the diagnosis in a situation where the diagnosis accuracy cannot be ensured and to reduce the processing load because unnecessary diagnosis is not required. Further, the output ratio C _i measured when the amount of solar radiation is insufficient may not be used as past data. In this way, the reliability of the standard deviation σ _i calculated from the past data is improved. It is possible to improve the diagnostic accuracy.

When the solar power generation system 1 does not measure the amount of solar radiation can be used amount of power generation P _i of the power generation amount P and / or the components of the system 1 as an index in place of solar radiation, more than a predetermined value When the amount of power generation is observed, it may be determined that the amount of solar radiation is greater than or equal to a predetermined value.

Incidentally, the total power generation amount P of the system 1 in the above, the output ratio C _i of each string _12, past output ratio C _i standard deviation sigma _i obtained _from, and the deviation T _i based on the standard deviation sigma _i, the It may be calculated by any configuration of the power generation diagnosis apparatus according to the embodiment, or may be calculated by a calculation means other than the power generation diagnosis apparatus.

In the above-described embodiment, the output ratio C _i for each component has been described as being stored. However, in order to reduce the amount of data to be recorded, the output ratio for each diagnosis time zone that is a time unit for power generation diagnosis. calculating a representative value of C _i, it may be stored only this representative value.

As the representative value, an average value of the output ratio C _i measured in the diagnosis time zone can be used. As an average value calculation method, an arithmetic average (simple average) can be used, and a value extremely different from other values may not be used as an abnormal value (outlier) in calculating the average. . Furthermore, the average value may be obtained by using a weighted average in which the value measured at the center of the diagnosis time zone is increased, and the median value may be used instead of the average value. Then, the representative value of the output ratio C _i for each diagnosis time period obtained in this way is sequentially stored.

In such a configuration, the standard deviation σ _i is calculated using a representative value of the output ratio C _i of each component stored for each diagnosis time period. Therefore, the number of data to be recorded is smaller than the configuration for recording all the output ratios C _i described above, and the number of past data when calculating the standard deviation σ _i can be reduced. The load can be reduced.

In the embodiment described above, a standard deviation sigma _i from the past data, and a deviation value T _i of the current output ratio C _i using such standard deviation sigma _i, but was used in the power generation diagnostic In addition to this, the determination may be performed using a mathematical statistical method such as F test, T test, or chi-square test.

  As described above, according to the power generation diagnostic device for the photovoltaic power generation system according to the present embodiment, diagnosis is performed using the output ratio of each component with respect to the total power generation amount of the system. In addition, since the diagnosis is performed by comparing the actual value of the output ratio of each component with the past data, the abnormality can be detected in a short time after the occurrence of power generation abnormality.

DESCRIPTION OF SYMBOLS 1 Photovoltaic power generation system 10 Solar cell 11 Solar cell panel 12a, 12b, 12c Solar cell string 20 Power conditioner 21a, 21b, 21c DC / DC converter 22 Inverter 30 Power supply system 40 Power generation diagnosis apparatus 41 Acquisition part 42 Storage part 43 Judgment part

Claims (16)

A method for performing power generation diagnosis of a photovoltaic power generation system configured by a plurality of components by a computer,
The computer obtaining information on the power generation amount of each component;
Storing each output ratio for each of the components with respect to the total power generation amount of the photovoltaic power generation system, calculated from the power generation amount of each of the acquired components;
Calculating each normal range for each output fraction of each component using at least a part of the most recent of the respective output ratios of a predetermined past number of days in a zone diagnosed and ing time and same time zone,
Comparing the respective normal ranges with the respective output ratios to determine a power generation abnormality. The power generation diagnosis method for a solar power generation system according to claim 1, wherein the component is any one of a string, an array, and a power conditioner. The determination step uses the standard deviation obtained from at least a part of the output ratios for the predetermined number of days in the past to determine the deviation values of the output ratios in the time zone to be diagnosed.
The power generation diagnosis method for a photovoltaic power generation system according to claim 1 or 2, wherein a power generation abnormality of the component is determined if each deviation value is outside a predetermined normal range. The power generation diagnosis method for a photovoltaic power generation system according to claim 3, wherein each normal range is set to be different for each time period. 5. The power generation diagnosis method for a photovoltaic power generation system according to claim 4, wherein each normal range in a time zone immediately after sunrise and / or just before sunset is set wider than each other normal range. The power generation diagnosis method for a solar power generation system according to claim 4, wherein each of the normal ranges in the time zone in which the sun is midway is set narrower than each of the other normal ranges. The length of the said time slot | zone is 1 minute-2 hours, The power generation diagnostic method of the solar power generation system of any one of Claims 1-6. The length of the said time slot | zone is 3 minutes-30 minutes, The power generation diagnostic method of the solar power generation system of any one of Claims 1-6. The length of the said time slot | zone is 5 minutes-20 minutes, The power generation diagnostic method of the solar power generation system of any one of Claims 1-6. The solar power generation according to any one of claims 1 to 9, wherein in the determination step, a power generation abnormality is determined using the output ratios for at least one day among the output ratios for the past 90 days. System power generation diagnosis method. The solar power generation according to any one of claims 1 to 9, wherein in the determination step, a power generation abnormality is determined using the output ratio of at least three days among the output ratios for the past 60 days. System power generation diagnosis method. The sunlight according to any one of claims 1 to 9, wherein in the determination step, power generation abnormality is determined using each output ratio of at least 7 days among the output ratios for the past 30 days. A power generation diagnosis method for a power generation system. The power generation diagnosis of the solar power generation system according to any one of claims 1 to 12, wherein in the determination step, the power generation abnormality is not determined when a solar radiation amount to the solar power generation system is insufficient. Method. The said determination step WHEREIN: Each said output ratio measured when the amount of solar radiation to the said photovoltaic power generation system is insufficient is not used as an output ratio for the said past predetermined number of days. The power generation diagnosis method of the solar power generation system according to item. The method of diagnosing power generation of a solar power generation system according to claim 13 or 14, wherein the shortage of solar radiation amount is determined from the power generation amount of the component and / or the total power generation amount of the system. A power generation diagnostic device for a solar power generation system configured by a plurality of components,
An acquisition unit for acquiring information on the power generation amount of each of the components;
A storage unit that stores each output ratio for each of the components with respect to the total power generation amount of the photovoltaic power generation system, calculated from the power generation amount of each of the acquired components.
The most recent past predetermined number of days in a diagnostic object and name Ru time zone and the time zone using at least a part of each output fraction was calculated each normal range for each output fraction of each component, the respective normal range And a determination unit that compares the output ratios with each other to determine power generation abnormality.

Images