Search

CN-121984444-A - Photovoltaic power station equipment intelligent early warning system and method based on multi-source data

CN121984444ACN 121984444 ACN121984444 ACN 121984444ACN-121984444-A

Abstract

The invention discloses an intelligent early warning system and method for photovoltaic power station equipment based on multi-source data, and relates to the technical field of intelligent operation and maintenance of photovoltaic, wherein the intelligent early warning system and method firstly synchronously acquire meteorological and electric data of a photovoltaic module and carry out cleaning, alignment and normalization treatment; the method comprises the steps of analyzing real-time slopes of power and irradiance of a component, dividing the component into three periods of low irradiation standby, linear work and saturated work, identifying severe performance degradation equipment, respectively calculating insulation degradation coefficients, performance attenuation indexes, voltage mismatch factors and current consistency indexes at different periods, detecting early insulation faults, evaluating performance attenuation and detecting electrical mismatch problems, further combining accumulated running time and historical fault times of the photovoltaic component, calculating comprehensive degradation degree scores through weighted average, and executing three-level dynamic early warning according to the comprehensive degradation degree scores, so that intelligent evaluation and accurate early warning of the health state of the photovoltaic component are achieved, and the intelligent level of operation and maintenance of a power station and the timeliness of fault response are improved.

Inventors

  • YAN CHAODAN
  • LU ZHENWEI
  • CHEN FANG

Assignees

  • 优得新能源科技（宁波）有限公司

Dates

Publication Date
20260505
Application Date
20260119

Claims (10)

  1. 1. A photovoltaic power station equipment intelligent early warning method based on multi-source data is characterized by comprising the following steps: Step 100, collecting meteorological and electric multisource data of a photovoltaic module in a photovoltaic power station; Step 200, calculating the real-time slope of the direct current power and irradiance of the photovoltaic module in a continuous time window, dividing the time window into a low irradiation standby period, a linear working period and a saturated working period in sequence by comparing with the theoretical slope, and triggering the highest priority early warning for the photovoltaic module with serious performance degradation in the linear period; step 300, calculating insulation degradation coefficient of the photovoltaic module in a low irradiation standby period, calculating performance attenuation index of the photovoltaic module in a linear working period, and calculating voltage mismatch factor and current consistency index of the photovoltaic module in a saturated working period; step 400, calculating an operation life factor by counting the accumulated effective power generation time of each photovoltaic module; And S500, after the daily operation is finished, integrating the period scores of the low irradiation standby period, the linear operation period and the saturation operation period, introducing an operation age factor and a fault density factor, and calculating the comprehensive degradation score to perform photovoltaic module fault early warning.
  2. 2. The intelligent early warning method for the photovoltaic power station equipment based on the multi-source data, as set forth in claim 1, wherein the step S200 includes: Step S201, for each photovoltaic module, normalizing and calculating a real-time slope k (t) between the direct current power P (t) and irradiance I (t) of the photovoltaic module by using a linear regression method in a continuous time window t, wherein the real-time slope k (t) represents the instantaneous conversion efficiency of the photovoltaic module under the current window; Step S202, calculating a theoretical slope k STC of the model of the photovoltaic module under standard conditions, and performing auxiliary judgment on an operation period by comparing a real-time slope k (t) with a theoretical slope k STC , wherein the auxiliary judgment is specifically that a low irradiation standby period is judged when k (t) is less than a1×k STC , a linear operation period is judged when a1×k STC ≤k(t)≤a2×k STC , a saturated operation period is judged when k (t) is more than a2×k STC , and a1 and a2 are coefficient thresholds, and a1 is less than a2 and set by professionals; Step S203, in the current time window, when the real-time slope k (t) under the linear working period is continuously lower than k 0 ×k STC and the duration time duty ratio exceeds theta% of the linear working period, marking as 'serious performance degradation', generating the highest priority early warning and immediately maintaining, wherein theta and k 0 are emergency thresholds and are set by professionals.
  3. 3. The intelligent early warning method for the photovoltaic power station equipment based on the multi-source data, according to claim 1, is characterized in that the step S300 comprises the following steps: Step S301, in a low irradiation standby period, the photovoltaic module approaches to an open circuit state, the average open circuit voltage V oc,avg of the photovoltaic module is calculated, then the average open circuit voltage V oc,avg is compared with the reference value voltage V oc,ref , the relative deviation of the voltage V oc,avg and the reference value voltage V oc,ref is normalized, and the insulation degradation coefficient is obtained The average open-circuit voltage is a direct-current voltage average value measured by the photovoltaic module in a low irradiation standby period, and the reference value voltage is a direct-current voltage average value calculated by the healthy photovoltaic module in the same environment; Step S302, in a linear working period, the behavior of a photovoltaic module is stable, the ratio PR raw of the average direct current power of the photovoltaic module to the current ideal power is calculated, and a performance attenuation index D is obtained, wherein the ideal power is the theoretical optimal output power of the photovoltaic module; step S303, in a saturation working period, a photovoltaic module works in a high current output interval, and the relative deviation between the average direct current voltage of the photovoltaic module and the average value of the string voltage is calculated to obtain a voltage mismatch factor VMF; analyzing the relative deviation between the average current of the photovoltaic module and the average value of the string current in the saturation working period to obtain a current consistency index CCI; the voltage and current consistency index is obtained by combining the voltage mismatch factor and the current consistency index, wherein F=k1×VMF+k2×CCI, and k1 and k2 are weighting coefficients, and k1+k2=1.
  4. 4. The intelligent early warning method for the photovoltaic power station equipment based on the multi-source data, which is disclosed in claim 1, is characterized in that the step S400 comprises the following steps: Step S401, counting the total operation time length of each photovoltaic module in a linear working period, comparing the total operation time length with the standard design life of the current photovoltaic module, and calculating the operation age factor of the current photovoltaic module; and step S402, counting the historical failure times of each photovoltaic module since the photovoltaic module is put into operation, and comparing the historical failure times with the design failure times, and calculating the failure density factor of the current photovoltaic module.
  5. 5. The intelligent early warning method for the photovoltaic power station equipment based on the multi-source data, which is disclosed in claim 1, is characterized in that the step S500 comprises the following steps: Step S501, after the daily operation is finished, calculating insulation degradation coefficients of the low irradiation standby period, the linear operation period and the saturated operation period The performance attenuation index D and the voltage and current consistency index F are fused by adopting a weighted geometric average method, and then the running age factor and the fault density factor are introduced to calculate the comprehensive degradation degree score S; And step S502, executing a dynamic grading early warning mechanism based on the comprehensive degradation degree score S.
  6. 6. The intelligent early warning method for the photovoltaic power station equipment based on the multi-source data, which is disclosed in claim 1, is characterized in that the step S100 comprises the following steps: Step S101, synchronously collecting meteorological and electrical multi-source data in a photovoltaic power station to construct an original data pool, wherein the meteorological multi-source data comprise plane irradiance I of a photovoltaic module and backboard temperature T, and the electrical multi-source data comprise direct current voltage, direct current and output power P of each photovoltaic group string and module; And S102, aligning the time stamps of all data streams through a clock synchronization protocol, filtering and removing abnormal data points, normalizing the actually measured electric power to a reference value under standard test conditions, and eliminating interference of environmental fluctuation on performance comparison.
  7. 7. The intelligent early warning system for the photovoltaic power station equipment based on the multi-source data is characterized by comprising a multi-source data acquisition and preprocessing module, a dynamic slope analysis and time interval division module, a time interval fault feature extraction module, a long-term state factor statistics module and a comprehensive evaluation and dynamic early warning module; The multi-source data acquisition and preprocessing module is responsible for synchronously acquiring meteorological data and electrical data from a photovoltaic power station on site, and performing time stamp alignment on all data streams through a clock synchronization protocol; The dynamic slope analysis and period division module calculates dynamic performance slopes for each component by analyzing the real-time relation between the output power and irradiance of each photovoltaic component, compares the dynamic performance slopes with theoretical slopes, divides the operation period into a low-irradiation standby period, a linear operation period and a saturated operation period, and is internally provided with emergency diagnosis logic, so that the component with serious performance degradation can be directly identified and marked, and the highest priority early warning is triggered; The time-division fault feature extraction module calculates the insulation degradation coefficient of the photovoltaic module in a low irradiation standby period, calculates the performance attenuation index of the photovoltaic module in a linear working period, and calculates the voltage mismatch factor and the current consistency index of the photovoltaic module in a saturated working period; the long-term state factor statistics module calculates an operation age factor by counting accumulated effective power generation time; After the daily operation is finished, the comprehensive evaluation and dynamic early warning module fuses fault characteristic indexes from three different working periods, wherein the fault characteristic indexes comprise insulation degradation coefficients, performance attenuation indexes, voltage mismatch factors and current consistency indexes, operational age factors and fault density factors are introduced, a weighted geometric average method is adopted to calculate a comprehensive degradation degree score S, and based on the comprehensive degradation degree score S, a dynamic grading early warning mechanism is executed by the system, early warning is divided into attention level, abnormal level and Yan Chongji, and different operation and maintenance response strategies are associated.
  8. 8. The intelligent early warning system of the photovoltaic power station equipment based on the multi-source data, which is characterized in that the dynamic slope analysis and time interval division module comprises a dynamic performance slope calculation unit, an intelligent working time interval division unit and a seriously deteriorated equipment identification unit; The dynamic performance slope calculation unit is used for calculating a real-time slope k (t) by fitting the relation between the normalized direct current power P (t) and irradiance I (t) of each photovoltaic module by adopting a linear regression method in a continuous sliding time window; The intelligent dividing unit of the working period automatically divides the working state of the component into three characteristic periods, namely a low irradiation standby period, a linear working period and a saturated working period, according to a preset coefficient threshold value by comparing the real-time slope k (t) with a theoretical slope k STC of the photovoltaic component under a standard condition; The serious degradation equipment identification unit is used for continuously monitoring in a linear working period, when the slope k (t) of the photovoltaic module is continuously lower than an emergency threshold k0 x k STC in linear working data points exceeding the theta percent ratio, the module is directly judged to be serious performance degradation, the system is used for skipping a subsequent comprehensive scoring process, and the highest priority maintenance early warning is immediately generated.
  9. 9. The intelligent early warning system of the photovoltaic power station equipment based on the multi-source data, which is characterized in that the time-period fault characteristic extraction module comprises an insulation degradation coefficient calculation unit, a performance attenuation index calculation unit and an electrical parameter mismatch detection unit; The insulation degradation coefficient calculating unit calculates an average open circuit voltage of the component in the low irradiation standby period and compares the average open circuit voltage with a health reference value, and calculates an insulation degradation coefficient gamma by normalizing the relative deviation; When the photovoltaic module is in the linear working period, the performance attenuation index calculation unit calculates the ratio of the actual measured power to the ideal power of the module to obtain an original performance ratio PR raw , and then performs secondary normalization on the original performance ratio PR raw and the reference performance ratio of the healthy module in the same environment to calculate a performance attenuation index D; When the photovoltaic module works in a saturated working period, the electrical parameter mismatch detection unit obtains a voltage mismatch factor VMF by calculating the relative deviation between the module voltage and the group string average voltage, obtains a current consistency index CCI by calculating the relative deviation between the module current and the group string average current, and fuses the two to be an electrical parameter consistency index F in a weighting way.
  10. 10. The intelligent early warning system of the photovoltaic power station equipment based on the multi-source data, which is characterized in that the long-term state factor statistics module comprises an operation age factor calculation unit and a fault density factor calculation unit; The operation age factor calculating unit counts the total operation time T oper of each photovoltaic module under the linear working period, compares the total operation time T oper with the standard design life T norm of the current photovoltaic module, and calculates an operation age factor tau; The fault density factor calculating unit traces back all historical fault records of each photovoltaic module since the photovoltaic module is put into operation, counts the total number of faults N fault , and calculates a fault density factor phi compared with the design number of faults N th .

Description

Photovoltaic power station equipment intelligent early warning system and method based on multi-source data Technical Field The invention relates to the technical field of intelligent operation and maintenance of photovoltaics, in particular to an intelligent early warning system and method for photovoltaic power station equipment based on multi-source data. Background Photovoltaic power generation plays a key role in achieving the "two carbon" goal as an important component of clean energy. With the continuous increase of the installed capacity and the continuous expansion of the deployment scale of the photovoltaic power station, the long-term reliable operation and efficient maintenance of power station equipment have become the core problems of industry attention. The photovoltaic module is used as a basic unit for power generation of a power station, is exposed to a complex natural environment for a long time, is inevitably influenced by multiple factors such as illumination, temperature, humidity, dust, aging and the like, and causes gradual attenuation of performance or various faults such as hot spot effect, wiring faults, insulation degradation, power mismatch and the like. At present, the operation and maintenance mode of the photovoltaic power station still depends on traditional modes such as regular inspection, post maintenance and the like, and the real-time, accurate sensing and intelligent research and judgment on the equipment state are lacking. The existing monitoring system can collect part of operation data, but is limited to single-dimensional electrical parameter monitoring, cannot fully fuse multi-source information, and lacks of targeted analysis of component behavior characteristics under different operation conditions. This results in early minor faults being difficult to find, potential degradation trends not being captured, and early warning often lags behind the occurrence of actual faults. Therefore, an intelligent early warning system and method which can integrate multi-source data, adapt to different operation conditions and comprehensively consider the historical state and real-time performance of equipment are needed, so that early diagnosis, accurate assessment and early warning of the health state of a photovoltaic module are realized, the initiative, predictability and intelligent level of operation and maintenance of a power station are improved, and the safe, efficient and long-term stable operation of the photovoltaic power station is ensured. Disclosure of Invention The invention aims to provide an intelligent early warning system and method for photovoltaic power station equipment based on multi-source data, which are used for solving the problems in the prior art. In order to achieve the above purpose, the present invention provides the following technical solutions: a photovoltaic power station equipment intelligent early warning method based on multi-source data is characterized by comprising the following steps: step 100, collecting meteorological and electrical multi-source data of a photovoltaic module in a photovoltaic power station, cleaning and aligning the multi-source data, and ensuring strict synchronization of signals of different sources on a time axis; Step S101, synchronously collecting meteorological and electrical multi-source data in a photovoltaic power station to construct an original data pool, wherein the meteorological multi-source data comprise plane irradiance I of a photovoltaic module and backboard temperature T, and the electrical multi-source data comprise direct current voltage, direct current and output power P of each photovoltaic group string and module; And S102, aligning the time stamps of all data streams through a clock synchronization protocol, filtering and removing abnormal data points, normalizing the actually measured electric power to a reference value under standard test conditions, and eliminating interference of environmental fluctuation on performance comparison. Step 200, the output characteristic of the photovoltaic module is nonlinear along with illumination intensity, the instantaneous energy conversion efficiency of the module is represented by analyzing the irradiance and the slope k of direct current power in the photovoltaic module, different physical characteristic time periods are divided based on the slope, serious imbalance equipment is identified, finer and more self-adaptive running state deconstruction is realized, a rapid channel is embedded, and catastrophic faults are blocked in real time; Step S201, calculating a dynamic performance slope k, namely calculating a real-time slope k (t) between normalized direct current power P (t) and irradiance I (t) of each photovoltaic module by using a linear regression method in a continuous time window t, wherein the slope k (t) represents the instantaneous conversion efficiency of the module in the current environment, and k (t) represents the slope between dir