Distributed photovoltaic state monitoring and evaluating method

Abstract

The invention discloses a distributed photovoltaic state monitoring and evaluation method, and the method specifically comprises the following steps: A1, collecting the data of a photovoltaic state through collection equipment, and relates to the technical field of distributed photovoltaic. According to the distributed photovoltaic state monitoring and evaluating method, data of a photovoltaic state is acquired through an acquisition device, and after the data is preprocessed, common feature identification, extraction and verification are carried out on typical data of various power generation anomalies of a photovoltaic panel through a common marking combined fault data feature set and a criterion set; rapid fault identification of different manufacturer difference photovoltaic devices is realized, more than 85% of various photovoltaic module faults can be accurately diagnosed, identification data are identified through a photovoltaic detection index system and a photovoltaic health model, and fault data obtained through extraction and verification are analyzed and evaluated to give optimization suggestions and abnormal state early warning. And summarizing the data to an intelligent management and control algorithm model to realize distributed photovoltaic intelligent management and control.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of distributed photovoltaic technology, in particular to a distributed photovoltaic state monitoring and evaluation method. BACKGROUND

[0002] According to the patent document, the name is: a distributed photovoltaic operation state evaluation method (the patent publication number is: CN117634742A, the patent publication date is: 2024-03-01), including: setting data monitoring period interval, automatically collecting temperature, humidity and light intensity data around the distributed photovoltaic, obtaining environmental comprehensive score after data processing; automatically collect the direct current and direct voltage data of each solar cell panel, and take an image of the surface of each solar cell panel, obtain abnormal comprehensive score after data processing; automatically collect the solar radiation intensity in the current air, the solar cell panel surface temperature and the electric energy output data, and obtain the performance comprehensive score after data processing; the comprehensive calculation of the environmental comprehensive score, the abnormal comprehensive score and the performance comprehensive score, generate the comprehensive score of the distributed photovoltaic operation state of each monitoring period, compare with the pre-set score level table, give the corresponding evaluation level according to the comparison result, get more comprehensive and accurate evaluation result.

[0003] Based on the above document, the existing photovoltaic equipment manufacturers are numerous, the models are different, the system component quantity is huge, the hot spot, the shadow shielding, the panel ground insulation impedance is too low, the power generation is low and other fault phenomena are common, the fault fast identification and positioning are lack of signification, the combined fault data feature set and criterion set, the fault troubleshooting workload is large, the hysteresis is strong, and the photovoltaic distribution is "point many and wide", the service environments such as mountainous area, coastal area, aquaculture farm, tropical island are different, there is no unified, universal photovoltaic system monitoring index system and evaluation and early warning method, therefore, the present application provides a distributed photovoltaic state monitoring and evaluation method. SUMMARY

[0004] In view of the shortcomings of the prior art, the present application provides a distributed photovoltaic state monitoring and evaluation method, which solves the problems that the existing photovoltaic equipment manufacturers are numerous, the models are different, the system component quantity is huge, the hot spot, the shadow shielding, the panel ground insulation impedance is too low, the power generation is low and other fault phenomena are common, the fault fast identification and positioning are lack of signification, the combined fault data feature set and criterion set, the fault troubleshooting workload is large, the hysteresis is strong, and the photovoltaic distribution is "point many and wide", the service environments such as mountainous area, coastal area, aquaculture farm, tropical island are different, there is no unified, universal photovoltaic system monitoring index system and evaluation and early warning method.

[0005] To achieve the above purpose, the present application is realized by the following technical scheme: a distributed photovoltaic state monitoring and evaluation method, specifically comprising the following steps:

[0006] A1, first using a collection device to collect data on the state of photovoltaics;

[0007] A2, and based on the collected data transmitted to the data control terminal, and complete the processing operation of the collected data, through the collection of various power generation abnormal typical data of photovoltaic panels and the identification, extraction and verification of common characteristics, construct common sign combination fault data feature set and criterion set;

[0008] A3, based on the processed data to establish a photovoltaic monitoring index system and a photovoltaic health model, realize the abnormal early warning of photovoltaic power generation system state;

[0009] A4, through the intelligent management and control algorithm model of the data summary, realize the intelligent management and control of distributed photovoltaic system.

[0010] Preferably, the collection device comprises a photovoltaic sensor and a transmission module, and the real-time collection of voltage, current, temperature, light intensity and other key operating data of the photovoltaic system is transmitted to the data control terminal for data processing operation.

[0011] Preferably, the data processing operation of the control terminal for the collected data in A2 is:

[0012] B1, first, the collected raw data is preprocessed, and the preprocessing is cleaning, denoising and standardizing the collected raw data;

[0013] B2, the preprocessed data is converted into a numerical value, and according to the integration and screening of the numerical value, the required data is stored and used for subsequent model construction.

[0014] Preferably, the establishment method of the feature set model in A2 is:

[0015] A photovoltaic mathematical model is constructed, and mechanism analysis is performed to determine the feature parameters under the typical working conditions of photovoltaic panel failure, such as no power generation, dust accumulation, damage, and abnormal power generation, such as over-capacity power generation and night power generation. Real-time simulation calculation of photovoltaic model is carried out based on embedded photovoltaic array, core parameters such as output and volt-ampere simulation data under different working conditions are obtained, and typical data features are extracted. Further, the real-time data of photovoltaic system under different working conditions are measured in the experiment or actual scene, and compared with the theoretical structure and simulation data, the photovoltaic system fault data feature set is constructed.

[0016] Preferably, the establishment method of the criterion set model in A2 is:

[0017] The system extracts and accurately identifies the features of typical fault data collected from photovoltaic panel power generation anomalies. It clarifies the common characteristics of typical fault data under the conditions of numerous photovoltaic equipment manufacturers and different models. It obtains the indicative and combined criteria for typical operating conditions such as photovoltaic panel not generating power, photovoltaic panel dust accumulation, photovoltaic panel damage, and power generation anomalies such as overcapacity power generation and nighttime power generation. Based on the mapping relationship between typical operating conditions and data features, it constructs a photovoltaic system fault data criterion set.

[0018] Preferably, the photovoltaic testing index system in A3 is established in the following way:

[0019] Based on data mining technology, the regional factors affecting the safety of photovoltaic systems are deeply extracted, and the data is summarized and analyzed to compile a list of risk factors. The Delphi method is used to judge the reliability, controllability, monitorability, and stability of photovoltaic systems, and a unified photovoltaic monitoring indicator system is established.

[0020] Preferably, the photovoltaic health model in A3 is established in the following way:

[0021] A photovoltaic health evaluation model is established based on improved combined weighting and dynamic fuzzy theory.

[0022] Preferably, the intelligent control algorithm model in A4 is established in the following way:

[0023] Based on the working principle of photovoltaic arrays, the original circuit calculation method is optimized to reduce the complexity of modeling. Digital twin technology and probabilistic graph theory are used to realize remote visualization and intelligent control of photovoltaic systems.

[0024] Beneficial effects

[0025] This invention provides a method for monitoring and assessing the condition of distributed photovoltaic systems. Compared with existing technologies, it has the following advantages:

[0026] (1) The method for monitoring and evaluating the condition of distributed photovoltaics collects data on the condition of photovoltaics through a data acquisition device. After data preprocessing, the method identifies, extracts and verifies the common features of various power generation anomalies of photovoltaic panels through a common characteristic combined fault data feature set and a criterion set. This enables rapid fault identification of photovoltaic equipment from different manufacturers and can accurately diagnose more than 85% of various photovoltaic module faults.

[0027] (2) This method for monitoring and evaluating the condition of distributed photovoltaics analyzes and evaluates fault data identified, extracted and verified through a photovoltaic detection index system and a photovoltaic health model, and provides optimization suggestions and early warnings of abnormal conditions. The abnormal conditions and optimization suggestions analyzed and evaluated are summarized into an intelligent control algorithm model. The intelligent control algorithm model is designed based on an analysis model combining digital twins and probabilistic graphs. The model analysis accuracy is greater than 80%. The intelligent control algorithm solution model based on a combination of data inference and pattern recognition is developed to realize intelligent control of distributed photovoltaics. Attached Figure Description

[0028] Figure 1 This is a flowchart of the evaluation method of the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Please see Figure 1 This invention provides two technical solutions:

[0031] Example 1: A method for monitoring and assessing the condition of distributed photovoltaic systems, specifically including the following steps:

[0032] A1. First, use data acquisition equipment to collect data on the photovoltaic status;

[0033] A2. Based on the data collected, the data is transmitted to the control terminal, and the data processing operation is completed. Through the collection of typical data of various power generation anomalies of photovoltaic panels and the identification, extraction and verification of common features, a common characteristic combined fault data feature set and criterion set are constructed.

[0034] A3. Based on the processed data, establish a photovoltaic monitoring indicator system and a photovoltaic health model to achieve early warning of abnormal conditions in the photovoltaic power generation system.

[0035] A4. Through the intelligent management and control algorithm model based on the evaluated data, intelligent management and control of distributed photovoltaic systems can be achieved.

[0036] In this embodiment of the invention, the acquisition device includes a photovoltaic sensor and a transmission module, and transmits key operating data such as voltage, current, temperature, and light intensity of the photovoltaic system in real time to the data control terminal for data processing. The transmission module is responsible for transmitting the processed data via wired or wireless means. Common transmission methods include RS485, RS232, Ethernet, WIFI, GPRS, etc., and the specific selection depends on the site environment and system requirements.

[0037] In this embodiment of the invention, the data processing operation of the control terminal in A2 for the collected data is as follows:

[0038] B1. First, the collected raw data is preprocessed, which involves cleaning, denoising, and standardizing the collected raw data.

[0039] B2. Convert the preprocessed data into numerical values, and based on the integration and filtering of the numerical values, store the required data for subsequent model building.

[0040] Example 2 differs from Example 1 in that the feature set model in A2 is established in the following way in this embodiment:

[0041] A photovoltaic mathematical model is constructed and its mechanism is analyzed to identify the characteristic parameters under typical operating conditions such as photovoltaic panel non-power generation, photovoltaic panel dust accumulation, photovoltaic panel damage, and power generation anomalies such as overcapacity power generation and nighttime power generation. Real-time simulation calculations of the photovoltaic model are performed based on an embedded photovoltaic array to obtain simulation data of core parameters such as output and volt-ampere under different operating conditions, and typical data features are extracted. Furthermore, real-time data of the photovoltaic system under different operating conditions are measured in experiments or actual scenarios and compared with theoretical structures and simulation data to construct a photovoltaic system fault data feature set. The photovoltaic mathematical model is based on the physical characteristics and working principle of photovoltaic cells and describes the output performance of photovoltaic cells or photovoltaic systems through mathematical equations.

[0042] In this embodiment of the invention, the criterion set model in A2 is established as follows:

[0043] The system extracts and accurately identifies the features of typical fault data collected from photovoltaic panel power generation anomalies. It clarifies the common characteristics of typical fault data under the conditions of numerous photovoltaic equipment manufacturers and different models. It obtains the indicative and combined criteria for typical operating conditions such as photovoltaic panel not generating power, photovoltaic panel dust accumulation, photovoltaic panel damage, and power generation anomalies such as overcapacity power generation and nighttime power generation. Based on the mapping relationship between typical operating conditions and data features, it constructs a photovoltaic system fault data criterion set.

[0044] In this embodiment of the invention, the photovoltaic detection index system in A3 is established as follows:

[0045] Based on data mining technology, the regional factors affecting the safety of photovoltaic systems are deeply extracted, and the data is summarized and analyzed to compile a list of risk factors. The Delphi method is used to judge the reliability, controllability, monitorability, and stability of photovoltaic systems, and a unified photovoltaic monitoring indicator system is established.

[0046] In this embodiment of the invention, the photovoltaic health model in A3 is established as follows:

[0047] A photovoltaic health evaluation model is established based on improved combined weighting and dynamic fuzzy theory.

[0048] In this embodiment of the invention, the intelligent control algorithm model in A4 is established as follows:

[0049] Based on the working principle of photovoltaic arrays, the original circuit calculation method is optimized to reduce the complexity of modeling. Digital twin technology and probabilistic graph theory are used to realize remote visualization and intelligent control of photovoltaic systems.

[0050] In summary, by collecting photovoltaic (PV) status data through acquisition equipment and preprocessing the data, common characteristics of various power generation anomalies in PV panels are identified, extracted, and verified using a common-marker combined fault data feature set and a criterion set. This enables rapid fault identification of PV equipment from different manufacturers, accurately diagnosing over 85% of various PV module faults. Furthermore, the identified, extracted, and verified fault data are analyzed and evaluated using a PV testing index system and a PV health model, providing optimization suggestions and abnormal state warnings. The analyzed and evaluated abnormal states and optimization suggestions are aggregated into an intelligent control algorithm model. This model is designed based on a combination of digital twins and probabilistic graphs, achieving an analysis accuracy greater than 80%. Finally, an intelligent control algorithm solution model based on data inference and pattern recognition is developed, enabling intelligent control of distributed PV systems.

[0051] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art.

[0052] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0053] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring and assessing the condition of distributed photovoltaic systems, characterized in that: Specifically, the following steps are included: A1. First, use data acquisition equipment to collect data on the photovoltaic status; A2. Based on the data collected, the data is transmitted to the control terminal, and the data processing operation is completed. Through the collection of typical data of various power generation anomalies of photovoltaic panels and the identification, extraction and verification of common features, a common characteristic combined fault data feature set and a criterion set are constructed. A3. Based on the processed data, establish a photovoltaic detection index system and a photovoltaic health model to achieve early warning of abnormal conditions in photovoltaic power generation systems; A4. Through the intelligent management and control algorithm model based on the evaluated data, intelligent management and control of distributed photovoltaic systems can be achieved.

2. The distributed photovoltaic state monitoring and assessment method according to claim 1, characterized in that: The data acquisition device includes a photovoltaic sensor and a transmission module, and transmits key operating data of the photovoltaic system, such as voltage, current, temperature, and light intensity, to the data control terminal in real time for data processing.

3. The distributed photovoltaic state monitoring and assessment method according to claim 1, characterized in that: The data processing operations performed by the control terminal in A2 on the collected data are as follows: B1. First, the collected raw data is preprocessed, which involves cleaning, denoising, and standardizing the collected raw data. B2. Convert the preprocessed data into numerical values, and based on the integration and filtering of the numerical values, store the required data for subsequent model building.

4. The distributed photovoltaic state monitoring and assessment method according to claim 1, characterized in that: The method for establishing the feature set model in A2 is as follows: A photovoltaic mathematical model is constructed and its mechanism is analyzed to identify the characteristic parameters under typical operating conditions such as photovoltaic panel non-power generation, photovoltaic panel dust accumulation, photovoltaic panel damage, and power generation anomalies such as overcapacity power generation and nighttime power generation. Real-time simulation calculations of the photovoltaic model are performed based on the embedded photovoltaic array to obtain simulation data of core parameters such as output and volt-ampere under different operating conditions. Typical data features are extracted, and real-time data of the photovoltaic system under different operating conditions are further measured in experiments or actual scenarios. The data are compared with theoretical structures and simulation data to construct a photovoltaic system fault data feature set.

5. The distributed photovoltaic state monitoring and assessment method according to claim 1, characterized in that: The method for establishing the criterion set model in A2 is as follows: The system extracts and accurately identifies the features of typical fault data collected from photovoltaic panel power generation anomalies. It clarifies the common characteristics of typical fault data under the conditions of numerous photovoltaic equipment manufacturers and different models. It obtains the indicative and combined criteria for typical operating conditions such as photovoltaic panel not generating power, photovoltaic panel dust accumulation, photovoltaic panel damage, and power generation anomalies such as overcapacity power generation and nighttime power generation. Based on the mapping relationship between typical operating conditions and data features, it constructs a photovoltaic system fault data criterion set.

6. The distributed photovoltaic state monitoring and assessment method according to claim 1, characterized in that: The establishment method of the photovoltaic monitoring indicator system in A3 is as follows: Based on data mining technology, the regional factors affecting the safety of photovoltaic systems are deeply extracted, and the data is summarized and analyzed to compile a list of risk factors. The Delphi method is used to judge the reliability, controllability, monitorability, and stability of photovoltaic systems, and a unified photovoltaic monitoring indicator system is established.

7. The distributed photovoltaic state monitoring and assessment method according to claim 1, characterized in that: The photovoltaic health model in A3 is established as follows: A photovoltaic health assessment model is established based on improved combined weighting and dynamic fuzzy theory.

8. The distributed photovoltaic state monitoring and assessment method according to claim 1, characterized in that: The method for establishing the intelligent control algorithm model in A4 is as follows: Based on the working principle of photovoltaic arrays, the original circuit calculation method is optimized to reduce the complexity of modeling. Digital twin technology and probabilistic graph theory are used to realize remote visualization and intelligent control of photovoltaic systems.

Images