Photovoltaic power station energy efficiency evaluation method

Abstract

The invention provides a photovoltaic power station energy efficiency evaluation method. A data acquisition layer module, a dynamic energy efficiency model module, an energy efficiency evaluation module and a fault positioning module are included. Wherein the dynamic energy efficiency model module establishes a theoretical power generation model based on an AI prediction technology, and the model dynamically updates model parameters according to real-time environment parameters and equipment states; the energy efficiency evaluation module calculates the actual energy efficiency according to the theoretical generating capacity predicted by the dynamic energy efficiency model and the actually collected inverter output power data, and performs energy efficiency loss grading; the fault positioning module discriminates an abnormal group string through a group string current dispersion rate and determines a hot spot position by means of an infrared image, relates to the technical field of photovoltaic power generation, and achieves precise and real-time power station energy efficiency management through fusion of dynamic environment correction, equipment health degree diagnosis and energy efficiency loss classification, thereby improving the efficiency of power station energy efficiency management. The problems that the evaluation deviation is large and the accuracy is reduced along with time in the existing evaluation technology are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power generation, and particularly to a method for evaluating the energy efficiency of a photovoltaic power station. Background Art

[0002] With the rapid development of the photovoltaic industry, the scale and quantity of photovoltaic power stations are increasing continuously. Accurately evaluating the energy efficiency of photovoltaic power stations is of crucial significance for optimizing the operation of power stations, improving power generation benefits, and reducing operation and maintenance costs. However, there are many deficiencies in the existing energy efficiency evaluation technologies for photovoltaic power stations.

[0003] Most of the existing technologies only use a single generated power or theoretical conversion efficiency for evaluation, completely ignoring environmental factors such as changes in irradiance, fluctuations in temperature, dust shading, etc., as well as the attenuation effects caused by the use of equipment over time. These factors actually have a significant impact on the actual energy efficiency of photovoltaic power stations. Ignoring them will lead to large deviations in the evaluation results and cannot truly reflect the operation status of the power stations.

[0004] In addition, traditional evaluation methods, such as the PR performance ratio, do not dynamically correct the parameters of the theoretical model during the evaluation process. As the photovoltaic power station operates for a long time, situations such as equipment aging and environmental changes occur continuously. If the parameters of the theoretical model always remain unchanged, the accuracy of the evaluation will gradually decrease, and it cannot provide a reliable basis for the effective management and optimization of the power station. Summary of the Invention

[0005] In order to overcome the existing problems, the embodiments of the present application provide a method for evaluating the energy efficiency of a photovoltaic power station, which achieves precise and real-time energy efficiency management of the power station by integrating dynamic environment correction, equipment health diagnosis, and energy efficiency loss classification, and effectively solves problems such as large evaluation deviations and decreasing accuracy over time existing in the existing evaluation technologies.

[0006] The technical solution adopted by the embodiments of the present application to solve its technical problems is as follows:

[0007] A method for evaluating the energy efficiency of a photovoltaic power station includes a data acquisition layer module, a dynamic energy efficiency model module, an energy efficiency evaluation module, and a fault location module;

[0008] Among them, the data acquisition layer module collects irradiance, component temperature, environmental temperature and humidity, dust coverage rate, inverter output power, DC side voltage / current, and string-level data in real time, and the data acquisition layer module uses one or more of wired communication and wireless communication to transmit the collected data to the data processing center in real time;

[0009] Among them, through various sensors and data interfaces, irradiance data is collected in real time to understand the changes in solar radiation energy; the temperature of the collection components is acquired because temperature has a direct impact on the power generation efficiency of photovoltaic modules; the ambient temperature and humidity are collected. Temperature and humidity not only affect the performance of the modules but also may affect the equipment lifespan; the dust coverage rate is measured, as dust occlusion will reduce the light intensity received by the modules; the output power of the inverter and the DC-side voltage / current are obtained, and these parameters reflect the power output and electrical operating status of the power station; and string-level data is obtained to provide a basis for more detailed subsequent analysis;

[0010] The dynamic energy efficiency model module establishes a theoretical power generation model based on AI prediction technology. This model dynamically updates the model parameters according to real-time environmental parameters and equipment status. The dynamic energy efficiency model module uses one or more machine learning algorithms such as neural networks and support vector machines to train and update the model parameters based on historical data and real-time data;

[0011] Among them, a theoretical power generation model is constructed based on AI prediction technology. Using historical data and real-time collected data, through machine learning algorithms such as neural networks and support vector machines, the power generation capacity of a photovoltaic power station under different environmental conditions and equipment status is modeled. The model can dynamically predict the theoretical power generation of the power station according to real-time environmental parameters such as irradiance, temperature, and humidity, as well as factors such as the operating time and aging degree of the equipment. Different from the traditional theoretical model with fixed parameters, this model will automatically adjust the parameters as the data is updated and the environment and equipment status change, maintaining the accuracy of predicting the power generation capacity of the power station;

[0012] The energy efficiency evaluation module calculates the actual energy efficiency and implements energy efficiency loss classification based on the theoretical power generation predicted by the dynamic energy efficiency model and data such as the actually collected inverter output power. The energy efficiency evaluation module calculates the actual energy efficiency according to the theoretical power generation predicted by the dynamic energy efficiency model and data such as the actually collected inverter output power, and introduces an energy efficiency loss classification mechanism. The energy efficiency loss classification is divided based on a preset number of energy efficiency intervals. For different levels of energy efficiency loss, corresponding maintenance strategies and optimization measures are formulated respectively; for example, when there is a slight loss, a maintenance strategy of strengthening the daily inspection frequency and optimizing the equipment monitoring parameters is formulated; when there is a moderate loss, an optimization measure of arranging professional technical personnel to conduct equipment performance detection and adjusting the equipment operation parameters is taken; when there is a severe loss, immediately start a comprehensive equipment overhaul process and replace key faulty components;

[0013] Among them, based on the theoretical power generation predicted by the dynamic energy efficiency model and data such as the actual output power of the inverter collected, the actual energy efficiency of the power station is calculated. By comparing the theoretical and actual power generation situations, the energy efficiency losses caused by environmental factors, equipment performance, etc. are analyzed. An energy efficiency loss grading mechanism is introduced to divide the energy efficiency losses into different levels, such as mild loss, moderate loss, and severe loss. For different levels of energy efficiency losses, corresponding optimization strategies and operation and maintenance suggestions are formulated;

[0014] The fault location module discriminates abnormal strings through the string current dispersion rate and determines the hot spot position with the help of infrared images. The fault location module uses an infrared thermal imager to perform thermal imaging on the suspected abnormal strings, and identifies the high-temperature area in the infrared image through image analysis software to determine the hot spot position;

[0015] Among them, based on the string current dispersion rate to identify abnormal strings. Under normal operating conditions, the currents of each photovoltaic string should be relatively close. When the current of a certain string is significantly different from the currents of other strings, that is, when the string current dispersion rate exceeds a certain threshold, it can be judged that there may be a problem with this string. Further combined with infrared image technology, thermal imaging detection is carried out on the suspected abnormal strings. Since faults such as hot spots will cause local temperature rise, obvious abnormal hot areas will appear in the infrared image, so as to accurately determine the hot spot position and provide accurate information for timely fault repair.

[0016] Preferably, after the fault location module determines the hot spot position in combination with the infrared image, it will estimate the impact degree of the hot spot on the overall power generation efficiency of the photovoltaic power station according to the area size, temperature level of the hot spot, and electrical parameters of the string where the hot spot is located. The calculation formula is:

[0017] I impactr = K1*S + K2*ΔT + K3*I string

[0018] Where I impactr is the impact degree of the hot spot on the power generation efficiency, S is the hot spot area, ΔT is the temperature difference between the hot spot and the normal area, I string is the current of the string where the hot spot is located, and K1, K2, and K3 are weight coefficients calibrated according to the actual operation data of the power station.

[0019] Preferably, the formula for the energy efficiency evaluation module to calculate the actual energy efficiency is:

[0020]

[0021] Where ρ is the actual energy efficiency, E actual is the actual power generation, E theory is the theoretical power generation.

[0022] Preferably, the formula for the fault location module to calculate the string current dispersion rate D is:

[0023]

[0024] where I i is the current of the i-th string, and I n is the average value of the currents of all strings, and μ is the total number of strings; when D is greater than the set threshold D threshold , it is determined that the corresponding string is an abnormal string.

[0025] Preferably, in the dynamic energy efficiency model module, the prediction formula for the theoretical power generation E theory is:

[0026] E theory = f(I rr , T c , H, R d , A t )

[0027] where I rr is the irradiance, T c is the module temperature, H is the ambient humidity, R d is the dust coverage rate, A t is the equipment operation time, and f is a mapping function obtained by training through a machine learning algorithm.

[0028] The advantages of the embodiments of the present application are:

[0029] In the present invention, the influence of environmental factors and equipment attenuation is fully considered, and the theoretical power generation prediction is adjusted in real time through a dynamic energy efficiency model, so that the energy efficiency evaluation result is closer to the actual operation of the power station, effectively reducing the evaluation deviation; each module collects and processes data in real time, can timely detect energy efficiency changes and equipment failures, provides real-time operation status information for power station operation and maintenance personnel, and is convenient for timely taking measures to optimize power station operation; the fault location method based on the string current discreteness rate and infrared image technology can quickly and accurately determine the abnormal string and the hot spot position, improve the fault troubleshooting and repair efficiency, and reduce the power generation loss caused by faults; through the energy efficiency loss grading, targeted strategy suggestions are provided for power station operation and maintenance, the operation and maintenance resources are reasonably arranged, the operation and maintenance cost is reduced, and the overall benefit of the power station is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a schematic flow chart of the photovoltaic power station energy efficiency evaluation method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0031] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention. In addition, for the convenience of description below, the "upper", "lower", "left", "right", etc. cited are consistent with the upper, lower, left, right, etc. of the accompanying drawings themselves. The "first", "second", etc. in the following text are used for distinction in description and have no other special meaning.

[0032] The embodiment of the present application provides a method for evaluating the energy efficiency of a photovoltaic power station to solve the problems in the prior art. It fully considers the influence of environmental factors and equipment attenuation, and adjusts the theoretical power generation prediction in real time through a dynamic energy efficiency model, making the energy efficiency evaluation result closer to the actual operation of the power station and effectively reducing the evaluation deviation; each module collects and processes data in real time, can timely detect energy efficiency changes and equipment failures, provides real-time operation status information for the power station operation and maintenance personnel, and is convenient for taking measures in time to optimize the operation of the power station; the fault location method based on the string current discreteness rate and infrared image technology can quickly and accurately determine the abnormal string and hot spot positions, improve the efficiency of fault troubleshooting and repair, and reduce the power generation loss caused by faults; through energy efficiency loss grading, it provides targeted strategy suggestions for power station operation and maintenance, reasonably arranges operation and maintenance resources, reduces operation and maintenance costs, and improves the overall efficiency of the power station.

[0033] The technical solution in the embodiment of the present application to solve the above problems has the following general idea:

[0034] Embodiment

[0035] This embodiment provides a method for evaluating the energy efficiency of a photovoltaic power station, as Figure 1 shown, including a data acquisition layer module, a dynamic energy efficiency model module, an energy efficiency evaluation module, and a fault location module;

[0036] Among them, the data acquisition layer module collects irradiance, component temperature, ambient temperature and humidity, dust coverage rate, inverter output power, DC side voltage / current, and string-level data in real time. The data acquisition layer module uses one or more of wired communication and wireless communication to transmit the collected data to the data processing center in real time;

[0037] Among them, irradiance data is collected in real time through various sensors and data interfaces to understand the changes in solar radiation energy; the temperature of the collection components is acquired because temperature has a direct impact on the power generation efficiency of photovoltaic modules; environmental temperature and humidity are collected, as temperature and humidity not only affect the performance of the modules but also may affect the service life of the equipment; the dust coverage rate is measured, as dust occlusion will reduce the light intensity received by the modules; the output power of the inverter and the DC-side voltage / current are obtained, and these parameters reflect the power output and electrical operating status of the power station; and string-level data is obtained to provide a basis for more detailed subsequent analysis;

[0038] The dynamic energy efficiency model module establishes a theoretical power generation model based on AI prediction technology. This model dynamically updates the model parameters according to real-time environmental parameters and equipment status. The dynamic energy efficiency model module uses one or more machine learning algorithms such as neural networks and support vector machines to train and update the parameters of the model based on historical data and real-time data;

[0039] Among them, a theoretical power generation model is constructed based on AI prediction technology. Using historical data and real-time collected data, through machine learning algorithms such as neural networks and support vector machines, the power generation capacity of the photovoltaic power station under different environmental conditions and equipment status is modeled. The model can dynamically predict the theoretical power generation of the power station according to real-time environmental parameters such as irradiance, temperature, and humidity, and factors such as the operating time and aging degree of the equipment. Different from the traditional theoretical model with fixed parameters, this model will automatically adjust the parameters with the update of data and the changes in the environment and equipment status, maintaining the accuracy of the prediction of the power generation capacity of the power station;

[0040] The energy efficiency evaluation module calculates the actual energy efficiency and implements energy efficiency loss classification based on the theoretical power generation predicted by the dynamic energy efficiency model and data such as the actual collected output power of the inverter. The energy efficiency evaluation module calculates the actual energy efficiency according to the theoretical power generation predicted by the dynamic energy efficiency model and data such as the actual collected output power of the inverter, and introduces an energy efficiency loss classification mechanism. The energy efficiency loss classification is divided based on a preset number of energy efficiency intervals. For different levels of energy efficiency loss, corresponding maintenance strategies and optimization measures are formulated respectively; for example, when there is a mild loss, a maintenance strategy of strengthening the daily inspection frequency and optimizing the equipment monitoring parameters is formulated; when there is a moderate loss, an optimization measure of arranging professional technical personnel to conduct equipment performance detection and adjusting the equipment operation parameters is taken; when there is a severe loss, a comprehensive equipment overhaul process is immediately started to replace key faulty components;

[0041] Among them, the actual energy efficiency of the power station is calculated based on the theoretical power generation predicted by the dynamic energy efficiency model and data such as the actually collected output power of the inverter. By comparing the theoretical and actual power generation situations, the energy efficiency losses caused by environmental factors, equipment performance, etc. are analyzed. An energy efficiency loss classification mechanism is introduced to divide the energy efficiency losses into different levels, such as mild loss, moderate loss, and severe loss. For different levels of energy efficiency losses, corresponding optimization strategies and operation and maintenance suggestions are formulated;

[0042] The fault location module discriminates abnormal strings through the string current discreteness rate and determines the hot spot position with the help of infrared images. The fault location module uses an infrared thermal imager to perform thermal imaging on suspected abnormal strings, and identifies the high-temperature areas in the infrared images through image analysis software to determine the hot spot position;

[0043] Among them, abnormal strings are identified based on the string current discreteness rate. Under normal operating conditions, the currents of each photovoltaic string should be relatively close. When the current of a certain string is significantly different from the currents of other strings, that is, when the string current discreteness rate exceeds a certain threshold, it can be judged that there may be a problem with that string. Further combined with infrared image technology, thermal imaging detection is performed on the suspected abnormal strings. Since faults such as hot spots will cause local temperature increases, obvious abnormal hot areas will appear in the infrared images, thus accurately determining the hot spot position and providing accurate information for timely fault repair.

[0044] After the fault location module determines the hot spot position in combination with the infrared image, it will estimate the impact degree of the hot spot on the overall power generation efficiency of the photovoltaic power station according to the area size, temperature, and electrical parameters of the string where the hot spot is located. The calculation formula is:

[0045] I impactr = K1*S + K2*ΔT + K3*I string

[0046] Among them, I impactr is the impact degree of the hot spot on the power generation efficiency, S is the hot spot area, ΔT is the temperature difference between the hot spot and the normal area, I string is the current of the string where the hot spot is located, and K1, K2, and K3 are weight coefficients calibrated according to the actual operation data of the power station.

[0047] The formula for the energy efficiency evaluation module to calculate the actual energy efficiency is:

[0048]

[0049] Among them, ρ is the actual energy efficiency, E actual is the actual power generation, and E theory is the theoretical power generation.

[0050] The formula for the fault location module to calculate the string current discreteness rate D is:

[0051]

[0052] where I i is the current of the i-th string, and I n is the average value of the currents of all strings, and μ is the total number of strings; when D is greater than the set threshold D threshold , it is determined that the corresponding string is an abnormal string.

[0053] In the dynamic energy efficiency model module, the prediction formula for the theoretical power generation E theory is as follows:

[0054] E theory = f(I rr , T c , H, R d , A t )

[0055] where I rr is the irradiance, T c is the module temperature, H is the ambient humidity, R d is the dust coverage rate, A t is the equipment operation time, and f is a mapping function obtained by training through machine learning algorithms.

[0056] By adopting the above technical solutions:

[0057] Implementation of the data acquisition layer module: Install high-precision sensors at various key positions in the photovoltaic power station. For example, install irradiance sensors near the surface of the photovoltaic modules to ensure accurate measurement of the light intensity received by the modules; install temperature sensors at appropriate positions inside or on the surface of the modules to monitor the module temperature in real time; use temperature and humidity sensors to monitor the ambient temperature and humidity; measure the dust coverage rate through image recognition technology or dedicated dust detection equipment; obtain the inverter output power and DC side voltage / current data through electrical measurement equipment; set up current acquisition devices at the string connection points to obtain string-level data, and all the collected data is transmitted to the data processing center in real time through wired or wireless communication methods.

[0058] Implementation of the dynamic energy efficiency model module: First, collect the long-term historical data of the photovoltaic power station, including environmental parameter data under different seasons and weather conditions, and operation data during the entire life cycle of the equipment. After cleaning and preprocessing these data, divide them into a training set and a test set. Select a suitable AI algorithm, such as a deep neural network, and use the training set data to train the model, adjust the model parameters, so that the model can accurately fit the relationship between the theoretical power generation and various influencing factors. During actual operation, the model receives the latest data transmitted from the data acquisition layer in real time, and dynamically updates the model parameters according to the current environment and equipment status to predict the theoretical power generation.

[0059] Implementation of the energy efficiency evaluation module: The theoretical power generation predicted by the dynamic energy efficiency model and the actual power data output by the inverter are obtained in real time, and the actual energy efficiency is calculated using the formula: actual energy efficiency = actual power generation / theoretical power generation × 100%. According to the pre-set energy efficiency loss classification standard, for example, when the actual energy efficiency is between 90% - 100%, it is a mild loss; when it is between 70% - 90%, it is a moderate loss; and when it is below 70%, it is a severe loss. The energy efficiency loss of the power station is classified, and corresponding strategies are given for different levels. For example, strengthen daily monitoring in case of mild loss, arrange equipment inspection in case of moderate loss, and immediately conduct a comprehensive overhaul in case of severe loss.

[0060] Implementation of the fault location module: At regular time intervals, for example, every 15 minutes, the string current discreteness rate is calculated. The calculation method is as follows: First, calculate the average value of all string currents, and then calculate the deviation of each string current I i from the average value. When the string current discreteness rate D exceeds the set threshold, mark the string as abnormal. For the abnormal string, use the infrared thermal imager installed in the power station to take a thermal image of it, and identify the high-temperature area in the infrared image through image analysis software to determine the position of the hot spot.

[0061] Finally, it should be noted that: Obviously, the above embodiments are merely examples for clearly explaining the present invention and are not intended to limit the implementation manners. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation manners here. And the obvious changes or modifications derived therefrom are still within the protection scope of the present invention.

Claims

1. A method for evaluating the energy efficiency of a photovoltaic power station, characterized in that, It includes a data acquisition layer module, a dynamic energy efficiency model module, an energy efficiency evaluation module, and a fault location module; Among them, the data acquisition layer module collects irradiance, component temperature, ambient temperature and humidity, dust coverage, inverter output power, DC side voltage / current, and string-level data in real time; The dynamic energy efficiency model module establishes a theoretical power generation model based on AI prediction technology, and this model dynamically updates the model parameters according to real-time environmental parameters and equipment status; The energy efficiency evaluation module calculates the actual energy efficiency and implements energy efficiency loss classification based on the theoretical power generation predicted by the dynamic energy efficiency model and the actually collected inverter output power data; The fault location module discriminates abnormal strings through the string current dispersion rate and determines the hot spot position with the help of infrared images.

2. The method for evaluating the energy efficiency of a photovoltaic power station according to claim 1, characterized in that, The energy efficiency loss classification is divided based on a plurality of preset energy efficiency intervals. For different levels of energy efficiency loss, corresponding maintenance strategies and optimization measures are formulated respectively; for example, when the loss is mild, a maintenance strategy of strengthening the daily inspection frequency and optimizing the equipment monitoring parameters is formulated; when the loss is moderate, professional technicians are arranged to conduct equipment performance detection and optimize the equipment operation parameters; when the loss is severe, immediately start a comprehensive equipment overhaul process and replace key faulty components.

3. The method for evaluating the energy efficiency of a photovoltaic power station according to claim 1, characterized in that, The dynamic energy efficiency model module uses one or more machine learning algorithms such as neural networks and support vector machines to train and update the model parameters based on historical data and real-time data.

4. The method for evaluating the energy efficiency of a photovoltaic power station according to claim 1, wherein The energy efficiency evaluation module calculates the actual energy efficiency according to the theoretical power generation predicted by the dynamic energy efficiency model and the actually collected inverter output power and other data, and introduces an energy efficiency loss classification mechanism.

5. The method for evaluating the energy efficiency of a photovoltaic power station according to claim 1, wherein, The fault location module uses an infrared thermal imager to take thermal images of suspected abnormal strings, and identifies the high-temperature areas in the infrared images through image analysis software to determine the hot spot position.

6. The method for evaluating the energy efficiency of a photovoltaic power station according to claim 1, characterized in that, The data acquisition layer module uses one or more of wired communication and wireless communication to transmit the collected data to the data processing center in real time.

7. The method for evaluating the energy efficiency of a photovoltaic power station according to claim 5, wherein After the fault location module determines the hot spot position in combination with the infrared image, it will estimate the impact degree of the hot spot on the overall power generation efficiency of the photovoltaic power station according to the area size, temperature, and electrical parameters of the string where the hot spot is located. The calculation formula is: I impactr = K1*S + K2*ΔT + K3*I string Among them, I impactr is the influence degree of hot spots on the power generation efficiency, S is the hot spot area, ΔT is the temperature difference between the hot spots and the normal area, and I string is the current of the corresponding string, and K1, K2, and K3 are weight coefficients calibrated according to the actual operation data of the power station.

8. The method for evaluating the energy efficiency of a photovoltaic power station according to claim 4, wherein The formula for the energy efficiency evaluation module to calculate the actual energy efficiency is: where ρ is the actual energy efficiency, E actual is the actual power generation, and E theory is the theoretical power generation.

9. A method for evaluating the energy efficiency of a photovoltaic power station according to claim 5, characterized in that, The formula for the fault location module to calculate the string current dispersion rate D is: Where I i is the current of the i-th string, and I n is the average value of the currents of all strings, and μ is the total number of strings; when D is greater than the set threshold D threshold it is determined that the corresponding string is an abnormal string.

10. A method for evaluating the energy efficiency of a photovoltaic power station according to claim 8, characterized in that, In the dynamic energy efficiency model module, the predicted formula for the theoretical power generation E theory is as follows: E theory = f(I rr , T c , H, R d , A t ) where I rr is the irradiance, T c is the module temperature, H is the ambient humidity, R d is the dust coverage rate, A t is the device operation time, and f is a mapping function obtained by training with a machine learning algorithm.

Images