Photovoltaic fault prediction and diagnosis method
By constructing the equipment impact curve of photovoltaic power generation units, the problem of distinguishing dust coverage and aging factors in large photovoltaic power plants is solved, intelligent and automated maintenance decisions are realized, and management efficiency is improved. It is suitable for large photovoltaic power plants.
Patent Information
- Application Number
- CN202411414925.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-11
- Publication Date
- 2025-01-07
AI Technical Summary
In large-scale photovoltaic power stations, it is difficult to effectively distinguish between dust coverage and photovoltaic panel damage, resulting in management difficulties, the existing inspection methods are inefficient and difficult to achieve targeted maintenance.
By building the equipment impact curve of the photovoltaic power generation unit, combining dust coverage and aging factors, we can judge the equipment status and realize automated early warning and maintenance decisions, including building the equipment impact curve during the operation period, predicting the equipment impact curve, and judging the intersection to determine maintenance or replacement requirements. .
It realizes intelligent and automated judgment and distinctive maintenance of photovoltaic power generation units, improves management efficiency, reduces human inspection needs, and is suitable for large photovoltaic power plants.
Smart Images
Figure CN119276220A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of new energy technology, and in particular to a photovoltaic fault prediction and diagnosis method. Background Art
[0002] Photovoltaic power station is a facility that uses solar photovoltaic panels to convert sunlight into electricity. Photovoltaic power stations have developed rapidly around the world and have become an important part of clean energy. The following is an introduction to several aspects of photovoltaic power stations: Principle: The working principle of photovoltaic power stations is based on the photoelectric effect. After absorbing sunlight, solar photovoltaic panels generate electron-hole pairs, and the electron-hole pairs are separated and recombined inside the battery. In this process, a weak current is generated, which can be processed by a power conversion device (such as an inverter) to output clean electricity. Structural composition: Photovoltaic power stations are mainly composed of the following parts: Ground or rooftop solar photovoltaic panels: used to receive and convert sunlight energy. Power conversion device: such as an inverter, which converts direct current into alternating current. Control system: monitors and controls the operation of photovoltaic power stations, including functions such as maximum power point tracking (MPPT). Bracket and mounting system: used to fix and install photovoltaic modules and lines connected to the power grid. Monitoring system: real-time monitoring and control of the operating status of photovoltaic power stations, as well as collection and analysis of relevant data. Type: According to the different installation methods of photovoltaic modules, photovoltaic power stations can be divided into ground photovoltaic power stations and rooftop photovoltaic power stations. Scale: The scale of photovoltaic power stations can range from small to large. Common ones include large ground photovoltaic power stations, medium-sized rooftop photovoltaic power stations and small distributed photovoltaic power stations. Advantages: Photovoltaic power stations have obvious advantages in many aspects, such as environmental protection, renewable energy, and saving fossil fuels. With the continuous advancement of technology, the cost of photovoltaic power stations is gradually decreasing, making them more and more valued. In short, photovoltaic power stations, as a clean energy facility, have broad development prospects. The Chinese government also actively supports the development of the photovoltaic industry and has formulated a series of policies and plans to promote the optimization of energy structure and environmental improvement.
[0003] Existing photovoltaic equipment requires relevant personnel to conduct regular inspections of the equipment to detect whether there are any abnormalities in the photovoltaic panels. For large-scale photovoltaic power stations, this will make inspections difficult. It is difficult to distinguish the impact of dust coverage from damage to the photovoltaic panels, and it is impossible to make targeted countermeasures plans, which leads to management difficulties. Summary of the invention
[0004] In order to solve the above technical problems, the present invention provides a photovoltaic fault prediction and diagnosis method for predicting and diagnosing faults of photovoltaic power generation units of photovoltaic power stations, which is particularly suitable for large-scale photovoltaic power stations. The photovoltaic fault prediction and diagnosis method may include:
[0005] S1. Obtain the equipment impact factor d at the time point during the operation period of the photovoltaic power generation unit, and construct the equipment impact curve of the operation period. In the actual working process, in order to better predict and diagnose the photovoltaic power station, we need to divide the prediction area according to the actual prediction ability to obtain regional units. These regional units are photovoltaic power generation units. A photovoltaic power generation unit needs to be in the same area and have the same external environment, and can be regarded as an object for research. The size of the photovoltaic power generation unit can be determined according to the actual situation. The minimum can be a photovoltaic panel, and the maximum can be a small photovoltaic power station within a range.
[0006] S2. Construct the equipment impact curve c for the non-operation period j , and connect them in sequence to obtain the predicted equipment impact curve C j , where j is the chronological number of the operation period that has not occurred, j = 1, 2, .... C0 is the equipment impact curve of the current operation period, and the predicted equipment impact curve C j It is the overall curve between the current operation period and the period j when no operation occurs.
[0007] S3. In predicting the equipment impact curve C j Construct a coordinate system with a vertical coordinate of C 标 In the actual working process, the theoretical prediction of the equipment influence curve C j It is a straight line, but considering the influence of various factors such as equipment aging, performance degradation and dust coverage, this straight line will have a small downward inclination. Since the dust coverage will be cleaned regularly, the downward inclination of this straight line will increase after cleaning, so the predicted equipment impact curve Cj is a wavy downward inclination curve. 标 The value of can be designed according to the performance of the photovoltaic power station, can be obtained based on experience, and of course can also be obtained by calculation.
[0008] S4. Determine and predict the equipment impact curve C j During the monitoring time, the line C 标 Is there any intersection? If yes, an early warning is issued and S5 is executed. If no, no early warning is issued and S6 is executed. In the actual working process, the predicted equipment influence curve C j The predicted length is generally the monitoring time, which depends on the actual situation. The monitoring time can also be less than the predicted device impact curve C j If there is a crossover, it means that during the normal aging process of the equipment, the predicted equipment impact curve C j Below C 标 , the equipment needs to be maintained or replaced.
[0009] S5. Determine and predict the equipment impact curve C j During the monitoring time, the straight line d 标 Is there a crossover, if yes, send a device change signal, if no, send a surface cleaning signal.
[0010] S6. Determine the equipment impact curve and straight line C 标 Is there any intersection? If yes, send a device maintenance signal. If not, do not send a device maintenance signal. This judgment can be used to handle emergency situations.
[0011] Preferably, the operation period can be one day, one week, one month or a period of time within one day, among which one day is more suitable. The number of time points within the operation period can be determined according to actual needs, and generally the number can be set to 1-20. The more the number, the greater the calculation amount and the more accurate the calculated data. The specific number needs to be determined according to actual conditions.
[0012] Preferred: The equipment impact factor Where P is the power generated by the photovoltaic power generation unit, which can be detected in real time by an electric meter, and E is the environmental impact factor, which can be obtained through environmental testing. Of course, the equipment impact factor can also be obtained by directly testing the equipment performance, but this data acquisition method is difficult to detect and has high equipment installation costs.
[0013] Preferably, the factors affecting photovoltaic power generation include maintenance factors and environmental factors.
[0014] Preferred: The factors that need maintenance mainly include photovoltaic panel factors, relay factors, circuit factors and dust coverage factors, which are the main factors for our evaluation of photovoltaic power generation units. If there is a problem with these factors, we believe that there is a problem with the photovoltaic panel, and countermeasures need to be checked, including maintenance, repair or cleaning for recovery. The photovoltaic panel factor is mainly the efficiency of the photovoltaic panel absorbing sunlight and converting it into electrical energy. It is the performance of the photovoltaic panel itself. It will decrease slightly with the use of the photovoltaic panel, but the change in a short period of time is not large. The specific value is generally 19%-21%. If there is a problem with this factor, the photovoltaic panel needs to be maintained or replaced. The relay factor is the relay conversion efficiency, and the circuit factor is the power efficiency transmitted by the meter, circuit, etc. We also believe that its value will not change, and we will not go into details here. The data value of the dust coverage factor will gradually increase over time, and then after cleaning, this value will rapidly decrease, and then repeat the cycle. In summary, under normal circumstances, we can think that the value of the maintenance-influencing factor between management cycles is a cyclical process of decreasing and increasing. The factors that affect the power generation of photovoltaic power stations only need to consider environmental factors to evaluate the performance of the photovoltaic power generation unit itself.
[0015] Preferably: environmental influencing factors may include solar radiation factors, irradiation angle factors and temperature factors. These factors are changing every day or even every moment, but can be obtained based on the detection of the external environment. The solar radiation data value can also be obtained through the weather forecast. Specifically, it can be obtained by looking up the local meteorological bureau to obtain the weekly weather forecast, etc. Of course, it can also be directly detected and obtained, and the details are not repeated here. The irradiation angle data value can be calculated based on the specific local time, the sun's irradiation angle and the installation angle of the solar panel. For the same location, the sun's irradiation angle is fixed every day with one year as a cycle, so we can calculate it, and the details are not repeated here. The data value of the temperature factor can be obtained by obtaining real-time data from the weather forecast and the real-time solar radiation to calculate the internal temperature of the solar panel, or it can be obtained by direct detection.
[0016] Preferred: Environmental Impact Factors Where i is the number of the environmental impact factor parameter, I is the total number of environmental impact factors, i = 1, 2, ..., I; e i is the data value corresponding to the environmental influencing factor numbered i.
[0017] Preferred: The method for constructing the equipment impact curve of the operating period includes: constructing a plane coordinate system, which uses each time point of the operating period as the horizontal coordinate and the proportional value as the vertical coordinate, and implants the calculated equipment parameter factor values in chronological order, connects each point, and then performs smoothing to form the equipment impact curve.
[0018] Preferred: Predicted Equipment Impact Curve C j The obtaining method may include: obtaining the equipment influence curve c of the non-occurring operation period j-1 , when j = 1, c j-1 That is, c0 is the equipment impact curve of the current operation period, which is known data. Obtain the similar curve segment c of the equipment impact curve of the operation period that has not occurred j '. And the equipment impact curve similar curve segment c of the non-operation period j 'Translate on the ordinate so that its starting point is aligned with the predicted equipment impact curve C j-1 The tail points of the predicted equipment impact curve C are obtained. j The equipment impact curve of the non-operation period is similar to the curve segment c j ' and the equipment impact curve c for the non-operation period j-1 , is a line segment, which is a curve formed by connecting time points within the operation period and has two endpoints, which are the starting point and the end point respectively in the order of time occurrence.
[0019] Preferred: Similar curve segment c of the equipment impact curve during the non-operation period j 'The acquisition method may include: extracting similarity evaluation factors of historical operation periods, which may include solar radiation factors, dust cleaning time and temperature factors, and calculating the similarity S between the historical operation period and the operation period j, and then using the equipment influence curve corresponding to the historical operation period with the greatest similarity as the equipment influence curve similarity curve segment c of the operation period j that did not occur j '.
[0020] Preferred: The similarity described Where || is the absolute value symbol, k is the number of the similarity evaluation factor, K is the total number of similarity evaluation factors, k = 1, 2, ..., K; A k is the data value of the historical operation period of the similarity evaluation factor numbered k, A k ' is the data value of the similarity evaluation factor of the number k in the operation period j. The calculation of its specific value is not given as an example here. By calculating with this method, the similarity can be limited to between 0 and 1, which is convenient for calculation and more in line with the needs of curve construction, and will not cause evaluation errors due to deviations.
[0021] Preferably, the predicted equipment impact curve Cj can be optimized, and the specific optimization process includes: obtaining the equipment impact curve of the current operation time period, and converting the predicted equipment impact curve Cj into j Translate to make the predicted equipment impact curve C j The starting point of the equipment influence curve coincides with the end point, and the first-order derivative L1 of the end point of the equipment influence curve is calculated, as well as the predicted equipment influence curve C at this point. j The first-order derivative L2, and then the predicted device influence curve C after translation j The point rotates through an angle of ɑ, where the rotation angle α=arctan[(L1-L2) / (1+L1*L2)]. This method can be used to optimize the value and development trend, making the correction more accurate and practical.
[0022] Preferred: The C 标 =d 标 ×h 标 , where d 标 is the standard equipment impact factor of the normal aging of photovoltaic panel factors, relay factors, and circuit factors in the current working period. Its specific value can be calculated based on the equipment impact factor when the photovoltaic panel was cleaned for the last time. At this time, theoretically h 标 =1, Where δ is the cleanliness factor, which is generally 0.95-1. The specific details are not described here. 标It is the dust cleanliness standard, and its value can be set according to the specific workload of the photovoltaic power station. The general value is 0.9-0.95. Of course, this value is not universal and depends on the local conditions.
[0023] The technical effects and advantages of the present invention are as follows: the maintenance status of the current photovoltaic power generation unit can be determined by this method, and targeted maintenance can be carried out. If it is surface cleaning, if the photovoltaic power station has automatic cleaning equipment, then automatic cleaning can be completed, which provides the possibility of intelligence and realizes the possibility of automatic determination and distinction. The maintenance type can be distinguished by this method, which is convenient for targeted maintenance. Situations such as bad weather and man-made damage can be determined, and there is no need for manual investigation in extreme weather, which is convenient for unified management and improves the automatic detection and management capabilities. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 The present invention is a flow chart of a photovoltaic fault prediction and diagnosis method.
[0025] Figure 2 The present invention provides a schematic flow chart of a method for constructing an equipment impact curve for an operating period in a photovoltaic fault prediction and diagnosis method.
[0026] Figure 3 The predicted equipment influence curve C in the photovoltaic fault prediction and diagnosis method proposed by the present invention j Schematic diagram of the process of obtaining.
[0027] Figure 4 The similar curve segment c of the equipment influence curve during the non-operation period in the photovoltaic fault prediction and diagnosis method proposed by the present invention is j 'Flowchart diagram of the process. DETAILED DESCRIPTION
[0028] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.
[0029] Example 1
[0030] refer to Figure 1In this embodiment, a photovoltaic fault prediction and diagnosis method is proposed, which is used to predict and diagnose the faults of photovoltaic power generation units in photovoltaic power stations, and is particularly suitable for large-scale photovoltaic power stations. The photovoltaic fault prediction and diagnosis method may include:
[0031] S1. Obtain the equipment impact factor d at the time point within the operation period of the photovoltaic power generation unit, and construct the equipment impact curve of the operation period. In the actual working process, in order to better predict and diagnose the photovoltaic power station, we need to divide the prediction area according to the actual prediction ability to obtain regional units. These regional units are photovoltaic power generation units. A photovoltaic power generation unit needs to be in the same area and have the same external environment, and can be regarded as an object for research. The size of the photovoltaic power generation unit can be determined according to the actual situation. The minimum can be a photovoltaic panel, and the maximum can be a small photovoltaic power station within a range. When a large photovoltaic power station has many small photovoltaic power stations in the power generation area, the small photovoltaic power stations in each area can be divided into one or more photovoltaic power generation units. The specific needs are based on the detection capability, the scale of the power station, and the management level. The specific details are not repeated here. The operation period is the time period within the normal operation of the solar photovoltaic panel. The normal working time of the solar photovoltaic panel generally starts when the sun rises and ends when the sun sets. This period can be connected in series or divided as the operation period. The specific needs refer to the local specific conditions and photovoltaic panel equipment parameters. The operating period can be one day, one week, one month or a period of time within a day, among which one day is more appropriate and will not be elaborated here. The number of time points within the operating period can be determined according to actual needs. Generally, the number can be set to 1-20. The more the number, the greater the amount of calculation and the more accurate the calculated data. The specific number needs to be determined according to actual conditions. In the actual working process, the data can be arranged symmetrically at 12 noon and the lighting conditions are sufficient, and the detection time point is fixed every day. For example, the number of time points in the operating period is 3, namely 10 am, 12 noon and 2 pm. At this time point every day, the equipment impact factor d needs to be calculated. The equipment impact factor Where P is the power generation of the photovoltaic power generation unit, which can be detected in real time by an electric meter. In the specific detection process, the PA333 power meter of ZLG Zhiyuan Electronics was used. The two-meter method can be used to test the power. The three-phase output power of photovoltaic power generation can be measured with two-channel power input, which can be directly displayed in real time. Of course, this is just a simple example, and it depends on actual needs. E is the environmental impact factor, which can be obtained through environmental testing. Of course, the equipment impact factor can also be obtained by directly testing the equipment performance, but this data acquisition method is difficult to detect and the equipment installation cost is high. For example, through detection, the power generation power of the photovoltaic power generation unit p = 0.2KW, the environmental impact factor E = 1.5KW, then the equipment parameter factor We divide the factors affecting photovoltaic power generation into maintenance-required factors and environmental factors. The maintenance-required factors mainly include photovoltaic panel factors, relay factors, circuit factors and dust coverage factors, which are the main factors for our evaluation of photovoltaic power generation units. If there is a problem with these factors, we believe that there is a problem with the photovoltaic panel and we need to find countermeasures, including maintenance, repair or cleaning for recovery. The photovoltaic panel factor is mainly the efficiency of the photovoltaic panel absorbing sunlight and converting it into electrical energy. It is the performance of the photovoltaic panel itself. It will decrease slightly with the use of the photovoltaic panel, but the change in a short period of time is not large. The specific value is generally 19%-21%. If there is a problem with this factor, the photovoltaic panel needs to be maintained or replaced. The relay factor is the relay conversion efficiency, and the circuit factor is the power efficiency transmitted by the meter, circuit, etc. We also believe that its value will not change, and we will not go into details here. The dust coverage factor data value will gradually increase over time, and then after cleaning, this value will rapidly decrease, and then the cycle will repeat. In summary, under normal circumstances, we can assume that the value of the factors that need to be maintained between management cycles is a cyclical process of decreasing and increasing. The factors that affect the power generation of photovoltaic power stations only need to consider environmental factors to evaluate the performance of the photovoltaic power generation unit itself. Environmental factors can include solar radiation factors, irradiation angle factors, and temperature factors. These factors are changing every day or even every moment, but they can be obtained based on the detection of the external environment. The solar radiation data value can also be obtained through weather forecasts. Specifically, you can find the local meteorological bureau to obtain the weekly weather forecast, etc., of course, you can also directly detect and obtain it, and the details will not be repeated here. The irradiation angle data value can be calculated based on the specific local time, solar irradiation angle, and the installation angle of the solar panel. For the same location, with one year as a cycle, the solar irradiation angle is fixed every day, so we can calculate it, and the details will not be repeated here. The data value of the temperature factor can be obtained by obtaining real-time data and real-time solar radiation from the weather forecast to calculate the internal temperature of the solar panel, or directly detected, and the details will not be repeated here. Environmental Impact Factors Where i is the number of the environmental impact factor parameter, I is the total number of environmental impact factors, i = 1, 2, ..., I; e i is the data value corresponding to the environmental influencing factor numbered i. For example, in the actual operation of a photovoltaic power station, we conduct statistics at 10 o'clock in the morning. The environmental influencing factors include temperature, solar radiation, and irradiation angle. By querying the weather forecast, we get that the temperature at 10 o'clock is 28 degrees and the solar radiation is 1000W / m 2 , the data value of the irradiation angle factor is 0.5. We calculate the internal temperature of the solar panel to be 30 degrees through the current temperature and solar radiation. This calculation method can find a pre-made solar radiation-temperature information table to obtain the internal temperature of the solar panel. The solar radiation-temperature information table can be obtained through the photovoltaic panel lighting experiment. The details will not be described here. Then, by looking up a pre-set temperature-data value information table, we get the temperature parameter factor of 0.9. Then we calculate the 10-point environmental impact factor E=1000×0.9×0.5=0.45KW. Of course, this example is not universal, and other specific situations will not be described here. The environmental impact factor calculated by this method is accurate, the data acquisition method is simple, and it can be detected in real time, ensuring the accuracy of the calculation. Reference Figure 2 The method for constructing the equipment influence curve of the operation period includes: constructing a plane coordinate system, which uses each time point of the operation period as the horizontal coordinate and the proportional value as the vertical coordinate. The equipment parameter factor values we calculated are implanted in chronological order, and each point is connected and then smoothed to form the equipment influence curve. The specific values are not given as examples here.
[0032] S2. Construct the equipment impact curve c for the non-operation period j , and connect them in sequence to obtain the predicted equipment impact curve C j , where j is the time sequence number of the unoccurred operation period, j = 1, 2, ...; in actual use, the number of j is generally 10, of course, other numerical settings are not excluded. C0 is the equipment impact curve of the current operation period, and the predicted equipment impact curve C j is the overall curve from the current operation period to the non-operation period j, refer to Figure 3 , and the method for obtaining the same may include: obtaining the equipment influence curve c of the non-occurring operation period j-1 , when j = 1, c j-1 That is, c0 is the equipment impact curve of the current operation period, which is known data. Obtain the similar curve segment c of the equipment impact curve of the operation period that has not occurred j '. And the equipment impact curve similar curve segment c of the non-operation period j'Translate on the ordinate so that its starting point is aligned with the predicted equipment impact curve C j-1 The tail points of the predicted equipment impact curve C are obtained. j The equipment impact curve of the non-operation period is similar to the curve segment c j ' and the equipment impact curve c for the non-operation period j-1 , is a line segment, which is a curve formed by connecting the time points within the operation period, and has two endpoints, which are the starting point and the end point in the order of time occurrence. Figure 4 , the equipment impact curve for the non-operation period is similar to the curve segment c j The obtaining method may include: extracting similarity evaluation factors of historical operation periods, the similarity evaluation factors may include solar radiation factor, dust cleaning time and temperature factor, and calculating the similarity S between the historical operation period and the operation period j, the similarity Where || is the absolute value symbol, k is the number of the similarity evaluation factor, K is the total number of similarity evaluation factors, k = 1, 2, ..., K; A k is the data value of the historical operation period of the similarity evaluation factor numbered k, A k ' is the data value of the similarity evaluation factor operation period j with number k. The calculation of its specific value is not given here as an example. By calculating with this method, the similarity can be limited between 0 and 1, which is convenient for calculation and more in line with the needs of curve construction. Evaluation errors will not be caused by deviations. Then the equipment impact curve corresponding to the historical operation period with the greatest similarity is used as the equipment impact curve similarity curve segment c of the operation period j that did not occur. j We can also optimize the predicted equipment impact curve Cj. The specific optimization process includes: obtaining the equipment impact curve of the current operation time period and converting the predicted equipment impact curve Cj into j Translate to make the predicted equipment impact curve C j The starting point of the equipment influence curve coincides with the end point, and the first-order derivative L1 of the end point of the equipment influence curve is calculated, as well as the predicted equipment influence curve C at this point. j The first-order derivative L2, and then the predicted device influence curve C after translation j The point rotates through an angle of ɑ, and the rotation angle α=arctan[(L1-L2)(1+L1*L2)]. By optimizing through this method, the optimization of the numerical value and the development trend can be completed, and the correction is more accurate and practical.
[0033] S3. In predicting the equipment impact curve C j Construct a coordinate system with a vertical coordinate of C 标 In the actual working process, the theoretical prediction of the equipment influence curve Cj It is a straight line, but considering the influence of various factors such as equipment aging, performance degradation and dust coverage, this straight line will have a small downward inclination. Since the dust coverage will be cleaned regularly, the downward inclination of this straight line will increase after cleaning, so the predicted equipment impact curve C j It is a wavy curve that slopes downward. 标 The value of can be designed according to the performance of the photovoltaic power station, can be obtained based on experience, and of course can also be obtained by calculation. 标 =d 标 ×h 标 , where d 标 is the standard equipment impact factor of the normal aging of photovoltaic panel factors, relay factors, and circuit factors in the current working period. Its specific value can be calculated based on the equipment impact factor when the photovoltaic panel was cleaned for the last time. At this time, theoretically h 标 =1, Where δ is the cleanliness factor, which is generally 0.95-1. The specific details are not described here. 标 The dust cleanliness standard can be set according to the specific workload of the photovoltaic power station. The general value is 0.9-0.95. Of course, this value is not universal and depends on the regional conditions. For example, the last time the photovoltaic power station was cleaned was 10 days ago. The cleanliness factor of the cleaning was 0.99, and the equipment impact factor after cleaning was 0.13. Then, after calculation, we can get The dust cleanliness standard value is 0.9, and C can be obtained by calculation. 标 =0.1313×0.9=0.1182. Of course, this is just a simple example and is not universal. Other numerical settings will not be elaborated here.
[0034] S4. Determine and predict the equipment impact curve C j During the monitoring time, the line C 标 Is there any intersection? If yes, an early warning is issued and S5 is executed. If no, no early warning is issued and S6 is executed. In the actual working process, the predicted equipment impact curve C j The prediction length is generally the monitoring time, which is generally set to 10 operation periods. The specific time can be 10 days, a week or a day, etc., depending on the actual situation. The monitoring time can also be less than the predicted equipment impact curve C j If there is a crossover, it means that during the normal aging process of the equipment, the predicted equipment impact curve C j Below C 标 , the equipment needs to be maintained or replaced.
[0035] S5. Determine and predict the equipment impact curve C j During the monitoring time, is there any intersection with the straight line d'? If yes, a device replacement signal is sent, if not, a surface cleaning signal is sent. d' is the equipment impact factor of equipment scrapping, which depends on the scrapping standards of each photovoltaic power station. The general value can be 0.5-0.1, depending on the actual situation. This method can be used to determine the maintenance status of the current photovoltaic power generation unit, and targeted maintenance can be carried out. If it is surface cleaning, if the photovoltaic power station has automatic cleaning equipment, then automatic cleaning can be completed, which provides the possibility of intelligence and realizes the possibility of automatic judgment and distinction. This method can be used to distinguish the types of maintenance, which is convenient for targeted maintenance.
[0036] S6. Determine the equipment impact curve and straight line C 标 Is there any intersection? If so, a device maintenance signal is sent; if not, no device maintenance signal is sent. This judgment can be used to handle emergency situations, such as severe weather such as sandstorms, strong winds, hail, and man-made damage. It determines that no manual inspection is required after extreme weather, which facilitates unified management and improves automatic detection and management capabilities.
[0037] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.
Claims
1. A photovoltaic fault prediction and diagnosis method, characterized in that: The photovoltaic fault prediction and diagnosis method comprises: S1. Obtain the equipment impact factor d at a time point during the operation period of the photovoltaic power generation unit, and construct an equipment impact curve during the operation period; S2. Construct the equipment impact curve c for the non-operation period j , and connect them in sequence to obtain the predicted equipment impact curve C j , where j is the time sequence number of the non-occurring operation period, j = 1, 2, ...; S3. In predicting the equipment impact curve C j Construct a coordinate system with a vertical coordinate of C 标 straight line; S4. Determine and predict the equipment impact curve C j During the monitoring time, the line C 标 Is there any intersection? If yes, a warning is issued and S5 is executed. If no, no warning is issued and S6 is executed. S5. Determine and predict the equipment impact curve C j Whether there is an intersection with the straight line d' during the monitoring time, if yes, a device replacement signal is sent, if not, a surface cleaning signal is sent; S6. Determine the equipment impact curve and straight line C 标 Is there any intersection? If yes, send the equipment maintenance signal; if no, do not send the equipment maintenance signal.
2. A photovoltaic fault prediction and diagnosis method according to claim 1, characterized in that: A work period is one of a day, a week, a month, or a period of time within a day.
3. A photovoltaic fault prediction and diagnosis method according to claim 1, characterized in that: The equipment impact factor Where P is the power generation of the photovoltaic power generation unit, and E is the environmental impact factor.
4. A photovoltaic fault prediction and diagnosis method according to claim 3, characterized in that: Factors that affect photovoltaic power generation include maintenance factors and environmental factors.
5. A photovoltaic fault prediction and diagnosis method according to claim 4, characterized in that: Environmental factors Where i is the number of the environmental impact factor parameter, I is the total number of environmental impact factors, i = 1, 2, ..., I; e i is the data value corresponding to the environmental influencing factor numbered i.
6. A photovoltaic fault prediction and diagnosis method according to claim 1, characterized in that: The method for constructing the equipment impact curve of the operating period includes: constructing a plane coordinate system, which uses each time point of the operating period as the horizontal coordinate and the proportional value as the vertical coordinate, implanting the equipment parameter factor values in chronological order, connecting each point, and then smoothing it to form the equipment impact curve.
7. A photovoltaic fault prediction and diagnosis method according to claim 1, characterized in that: Predicted Equipment Impact Curve C j The acquisition method includes: obtaining the equipment impact curve c of the non-occurring operation period j-1 , obtain the similar curve segment c of the equipment impact curve during the non-operation period j ', and the equipment impact curve similar curve segment c of the non-operation period j 'Translate on the ordinate so that its starting point is aligned with the predicted equipment impact curve C j-1 The tail points of the predicted equipment impact curve C are obtained. j .
8. A photovoltaic fault prediction and diagnosis method according to claim 7, characterized in that: The similar curve segment c of the equipment impact curve during the non-operation period j The acquisition method includes: extracting the similarity evaluation factors of the historical operation period, calculating the similarity S between the historical operation period and the operation period j, and then taking the equipment influence curve corresponding to the historical operation period with the greatest similarity as the similarity curve segment c of the equipment influence curve of the non-occurring operation period j j '.
9. A photovoltaic fault prediction and diagnosis method according to claim 8, characterized in that: Similarity evaluation factors include solar radiation factor, dust cleaning time and temperature factor.
10. A photovoltaic fault prediction and diagnosis method according to claim 1, characterized in that: The C 标 =d 标 ×h 标 , where d 标 is the standard equipment influencing factor of the normal aging of photovoltaic panel factors, relay factors, and circuit factors during the current working period, h 标 It is the dust cleanliness standard.