Big data-based photovoltaic power station fault detection method, system and storage medium

Abstract

The application discloses a photovoltaic power station fault detection method and system based on big data and a storage medium, relates to the technical field of photovoltaic power station fault detection, and comprises the following steps: through communication connection with various sensors, collecting power parameters and environmental parameters of each sub-string, and performing deep analysis on the data according to Fourier transform, trend change analysis and other technical means, extracting the key indicators of harmonic distortion, conversion monitoring values and resistance monitoring values as three judging elements; based on the three judging elements, comprehensively analyzing, identifying the sub-strings with faults, and executing corresponding safety strategies accordingly; realizing intelligent monitoring and preventive maintenance of the photovoltaic power station, taking preventive measures before the faults occur, reducing the probability and influence range of the faults, thereby prolonging the service life of the equipment and improving the overall operation efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power station fault detection, and in particular to a photovoltaic power station fault detection method, system and storage medium based on big data. Background Art

[0002] Photovoltaic power stations are facilities that convert solar energy into electricity, especially centralized photovoltaic power stations. They are often built in areas with ample sunlight and open areas, such as deserts, mountains, and plains. Because photovoltaic power stations have the environmental advantages of reducing fossil energy consumption and carbon emissions, they play a key role in the promotion of clean energy. Therefore, fault detection in photovoltaic power stations is particularly important to ensure safe operation and stable power output.

[0003] Centralized photovoltaic power plants typically occupy very large areas. Traditional fault detection in photovoltaic power plants often relies on a single detection method, such as manual inspections or simple monitoring systems. This results in low accuracy and efficiency in fault detection. In particular, when a fault occurs, timely maintenance and resolution are often not possible, leading to the spread of the problem or a decrease in power generation efficiency, affecting stable output. Summary of the Invention

[0004] The present invention provides a photovoltaic power station fault detection method, system and storage medium based on big data, which are used to solve the above technical problems.

[0005] A first aspect of the present invention provides a photovoltaic power station fault detection method based on big data, comprising the following steps:

[0006] Step 1: Divide the PV power station into several areas. Each area is equipped with an inverter. The PV panels within the area are divided into several substrings. The substring PV panels are connected in parallel. The PV panels within a substring are connected in series. The PV panels in each substring are numbered and marked on the PV power station map.

[0007] Step 2: Communicate with various sensors installed in each area to collect power parameters and environmental parameters of each substring in each area. Based on this, in-depth analysis is performed on the inverter working status, light energy conversion status, and resistance status of each substring in the area to obtain harmonic distortion, conversion monitoring value, and resistance monitoring value. This can be used to determine the three key factors for judging the performance of each substring photovoltaic panel in the area. Specifically, the three key factors are: harmonic distortion of the inverter, conversion monitoring value, and resistance monitoring value. The power parameters include voltage signal V(t), current signal I(t), and power generation; the environmental parameters include ambient temperature, ambient humidity, and irradiance.

[0008] Step 3: Perform a comprehensive analysis based on the three judgment factors of each sub-string photovoltaic panel in the area to accurately determine the health status of each sub-string photovoltaic panel, identify the sub-string photovoltaic panel with faults, and implement the corresponding safety strategy accordingly.

[0009] Optionally, the specific steps for comprehensively analyzing the three factors to identify the faulty substring of photovoltaic panels are as follows:

[0010] 3-1: Query the three key factors for determining the quality of each PV panel string in the area: the inverter's harmonic distortion, conversion monitoring value, and resistance monitoring value. If the harmonic distortion is greater than or equal to the set harmonic threshold, the PV panels in the area are stopped, the area is marked as a risk zone, and the risk zone number and location are output. The risk zone number and location are sent to the corresponding engineer. If the harmonic distortion is less than the set harmonic threshold, execute 3-2.

[0011] 3-2: Normalize the harmonic distortion VI, conversion monitoring value KZ, and resistance monitoring value RD and take their values. Calculate and analyze the values ​​using a formula to obtain the abnormal value CY. The specific Pythagorean theorem calculation formula is:

[0012]

[0013] Among them, λ1 and λ2 are the set proportional constants respectively;

[0014] 3-3: Compare and analyze the abnormal value with the set abnormal interval. When the abnormal value is greater than the maximum value in the set abnormal interval, the substring photovoltaic panel is recorded as a high-risk substring; when the abnormal value is in the set abnormal interval, the substring photovoltaic panel is recorded as a medium-risk substring; when the abnormal value is less than the minimum value in the set abnormal interval, the substring photovoltaic panel is recorded as a light-risk substring; count the number of high-risk substrings, medium-risk substrings and light-risk substrings in the area respectively, and record them as G1, G2 and G3 respectively; if G3≥G1+G2, the high-risk substring in the area is recorded as a faulty substring, the faulty substring is controlled to stop working, and the number and location of the faulty substring are output, and the number and location of the faulty substring are sent to the corresponding engineer; otherwise, the area is recorded as an abnormal area, the abnormal area is controlled to stop working, and the number and location of the abnormal area are output, thereby integrating the number, location and abnormal values ​​of each substring photovoltaic panel into the abnormal information of the abnormal area, and the abnormal information is sent to the corresponding engineer;

[0015] 3-4: Repeat steps 3-1 to 3-3 until all areas of the photovoltaic power station are fault-diagnosed and the corresponding safety policies are output.

[0016] Optionally, the monitoring and analysis steps for the inverter working status are:

[0017] The voltage and current signals of the inverter are retrieved, and the sampling frequency is often much higher than the output signal frequency, such as several thousand hertz or higher, to ensure that high-order harmonics can be accurately captured; through Fourier transform (FFT)

[0018] The time domain signal is converted into a frequency domain signal, and the voltage harmonic expression and each current harmonic expression are obtained accordingly. The specific voltage harmonic expression is:

[0019] V(t)=V1·sin(ωt+θ1)+V2·sin(2ωt+θ2)+V3·sin(3ωt+θ3)+…+V J ·sin(Jωt+θ J )

[0020] Where j = 1, 2, 3, ..., J; J is a positive integer, J represents the highest order of voltage harmonics, j represents the number of one of the voltage harmonics; V1 is the voltage amplitude of the fundamental wave, V2, V3, ..., V J is the voltage amplitude of each harmonic, corresponding to the 2nd, 3rd, up to the Jth voltage harmonic; ω is the angular frequency of the fundamental frequency;

[0021] θ1, θ2, θ3, ..., θ J is the phase angle of each voltage harmonic;

[0022] The specific current harmonic expression is:

[0023] I(t)=I1·sin(ωt+φ1)+I2·sin(2ωt+φ2)+I3·sin(3ωt+φ3)+…+I F ·sin(Fωt+φ F )

[0024] Where f = 1, 2, 3, ..., F; F is a positive integer, F represents the highest order of current harmonics, f represents the order number of one of the current harmonics; I1 is the current amplitude of the fundamental wave, I2, I3, ..., I F is the current amplitude of each harmonic, corresponding to the 2nd, 3rd, and up to the Fth current harmonic; ω is the angular frequency of the fundamental frequency; φφ1, φ2, φ3, ..., φ F is the phase angle of each current harmonic;

[0025] Extract the amplitude V of each harmonic based on the voltage harmonic expression and current harmonic expression j and I f , and perform formulaic calculation and analysis to obtain the harmonic distortion VI; the specific calculation formula is:

[0026]

[0027] where a V 、a I The specific values ​​are set by those skilled in the art according to actual needs, for example, a V The value is 0.35, a I The value is 0.65.

[0028] Optionally, the steps for monitoring and analyzing the conversion status are:

[0029] Retrieve the irradiance and power generation corresponding to each substring in the area at each collection time; normalize the irradiance and power generation and take their values, divide the values ​​to calculate the conversion rate, and thus obtain the conversion rate corresponding to each collection time, recorded as Kb, where b = 1, 2, 3, ..., B; B is a positive integer, B represents the total number of collection times, and b represents the sequence number of any collection time; construct a two-dimensional rectangular coordinate system with time as the horizontal coordinate and the conversion rate as the vertical coordinate, input the conversion rate into the coordinate axis according to its corresponding collection time, and record the position of the conversion rate in the coordinate axis as the conversion point; use a smooth curve to connect each conversion point in sequence to obtain a curve graph of the conversion rate changing with time; draw a tangent line at each conversion point, use data fitting to calculate the tangent slope and record it as the conversion slope; sum the conversion slopes greater than zero to calculate the conversion increase value, and sum the conversion slopes less than zero and take the absolute value to calculate the conversion decrease value;

[0030] The conversion rate Kb, conversion slope Zb, conversion increase value H1 and conversion decrease value H2 are normalized and their values ​​are taken. The values ​​are calculated and analyzed in a formula to obtain the conversion monitoring value KZ. The specific calculation formula is:

[0031]

[0032] Among them, σ1, σ2, and σ3 are the set proportional constants, is the mean of the conversion slopes at each conversion point.

[0033] Optionally, the monitoring and analysis steps for the resistance status are:

[0034] Set each substring photovoltaic panel to correspond to a standard environmental parameter, where the standard environmental parameters include standard ambient temperature and standard ambient humidity;

[0035] The insulation resistance Rb 绝缘 , series impedance Rb 阻抗 , ambient temperature Tb, ambient humidity Sb, standard ambient temperature AT and standard ambient humidity AS and normalize them and take their values, perform formula calculation and analysis on the values ​​to obtain the resistance state value PR; the specific calculation formula is:

[0036]

[0037] Among them, β1, β2, β3, and β4 are the set proportional constants, from which the resistance state value PRb corresponding to each acquisition moment can be obtained;

[0038] A two-dimensional rectangular coordinate system is constructed with time as the abscissa and resistance state value as the ordinate. The resistance state value is input into the coordinate axis according to its corresponding acquisition time, and the position of the resistance state value in the coordinate axis is recorded as a resistance point. A smooth curve is used to connect each resistance point in sequence to obtain a resistance state value variation curve over time; a tangent line is drawn at each resistance point, and the slope of the tangent line is obtained by data fitting and recorded as the resistance slope; the resistance slopes greater than zero are summed to obtain the state increase value, and the resistance slopes less than zero are summed and then the absolute value is taken to obtain the state decrease value;

[0039] The resistance state value PRb, resistance slope Db, state increase value H3 and state decrease value H4 at each acquisition moment are normalized and their values ​​are taken. The values ​​are calculated and analyzed in a formula to obtain the resistance monitoring value RD. The specific calculation formula is:

[0040]

[0041] Among them, α1, α2, and α3 are the set proportional constants, is the mean resistance slope of each resistance point.

[0042] A second aspect of the present invention provides a photovoltaic power station fault detection system based on big data, comprising: a server, a monitoring and analysis module, and a fault output module;

[0043] The server divides the photovoltaic power station into regions and collects information. The specific information collected is power parameters and environmental parameters. The power parameters include voltage signal V(t), current signal I(t), and power generation; the environmental parameters include ambient temperature, ambient humidity, and irradiance.

[0044] The monitoring and analysis module performs feature analysis on the inverter status, conversion status, and resistance status of each substring photovoltaic panel in each area based on power parameters and environmental parameters to extract the three judgment factors. The three judgment factors are: the harmonic distortion of the inverter, the conversion monitoring value, and the resistance monitoring value.

[0045] The fault output module performs a comprehensive analysis based on the three judgment factors of each sub-string photovoltaic panel in the area to accurately determine the health status of each sub-string photovoltaic panel, identify the sub-string photovoltaic panel with a fault, and implement the corresponding safety strategy accordingly; specifically:

[0046] 3-1: Query the three key factors for determining the quality of each PV panel string in the area: the inverter's harmonic distortion, conversion monitoring value, and resistance monitoring value. If the harmonic distortion is greater than or equal to the set harmonic threshold, the PV panels in the area are stopped, the area is marked as a risk zone, and the risk zone number and location are output. The risk zone number and location are sent to the corresponding engineer. If the harmonic distortion is less than the set harmonic threshold, execute 3-2.

[0047] 3-2: Normalize the harmonic distortion VI, conversion monitoring value KZ, and resistance monitoring value RD and take their values. Calculate and analyze the values ​​using a formula to obtain the abnormal value CY. The specific Pythagorean theorem calculation formula is:

[0048]

[0049] Among them, λ1 and λ2 are the set proportional constants respectively;

[0050] 3-3: Compare and analyze the abnormal value with the set abnormal interval. When the abnormal value is greater than the maximum value in the set abnormal interval, the substring photovoltaic panel is recorded as a high-risk substring; when the abnormal value is in the set abnormal interval, the substring photovoltaic panel is recorded as a medium-risk substring; when the abnormal value is less than the minimum value in the set abnormal interval, the substring photovoltaic panel is recorded as a light-risk substring; count the number of high-risk substrings, medium-risk substrings and light-risk substrings in the area respectively, and record them as G1, G2 and G3 respectively; if G3≥G1+G2, the high-risk substring in the area is recorded as a faulty substring, the faulty substring is controlled to stop working, and the number and location of the faulty substring are output, and the number and location of the faulty substring are sent to the corresponding engineer; otherwise, the area is recorded as an abnormal area, the abnormal area is controlled to stop working, and the number and location of the abnormal area are output, thereby integrating the number, location and abnormal values ​​of each substring photovoltaic panel into the abnormal information of the abnormal area, and the abnormal information is sent to the corresponding engineer;

[0051] 3-4: Repeat steps 3-1 to 3-3 until all areas of the photovoltaic power station are fault-diagnosed and the corresponding safety policies are output.

[0052] A third aspect of the present invention provides a computer-readable storage medium having instructions stored therein, which, when executed on a computer, enables the computer to execute the above-mentioned photovoltaic power station fault detection method based on big data.

[0053] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0054] 1. By numbering each region and substring and marking the location of each substring on the map, it provides convenience for subsequent fault location and maintenance; when a substring photovoltaic panel fails, engineers can quickly locate the problem according to the number and location information, thereby shortening the maintenance time and improving the maintenance efficiency;

[0055] 2. By communicating with various sensors, collecting power parameters and environmental parameters of each substring, and using Fourier transform, trend change analysis and other technical means to deeply analyze the data, extracting the key indicators of harmonic distortion, conversion monitoring value and resistance monitoring value as the three elements of judgment, it can realize accurate monitoring and analysis of the inverter state of the region, the conversion state of the photovoltaic panel and the resistance state; This data-driven analysis method can more accurately reflect the working state of the photovoltaic panel, and provide strong support for subsequent fault diagnosis and preventive maintenance;

[0056] 3. Based on the comprehensive analysis of the three elements of judgment of each substring photovoltaic panel, the health status of each substring can be accurately judged, and the substring with fault can be identified, and the corresponding safety strategy is executed accordingly; Specifically, when the harmonic distortion exceeds the set threshold, the photovoltaic panel of the related region can be immediately controlled to stop working to prevent further damage to the equipment and system caused by harmonic exceeding; In addition, by classifying the risk level of the substring photovoltaic panel, the corresponding safety strategy can be taken for the substring photovoltaic panel with different risk levels, further improving the safety of the system; Realize the intelligent monitoring and preventive maintenance of photovoltaic power station, can take preventive measures before the fault occurs, reduce the probability and influence range of the fault, thereby prolong the service life of the equipment and improve the overall operation efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0057] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiment description, the following drawings are not deliberately drawn according to the actual size, etc. Proportion, the focus is to show the main idea of the present application.

[0058] Figure 1 The principle diagram of the present application;

[0059] Figure 2 The system module connection diagram of the present application. DETAILED DESCRIPTION

[0060] Embodiments of the present invention provide an energy scheduling management method, apparatus, device and storage medium. The terms "first", "second", "third", "fourth", etc. (if any) in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or that are inherent to these processes, methods, products or devices.

[0061] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 In an embodiment of the present invention, a photovoltaic power station fault detection method based on big data includes the following steps:

[0062] Step 1: Since the centralized photovoltaic power station occupies a vast area and contains a huge number of photovoltaic panels, the entire photovoltaic power station is divided into several areas. Each area is numbered m, where m = 1, 2, 3, ..., M; M is a positive integer, M represents the total number of areas in the entire photovoltaic power station, and m represents the serial number of any area. Each area is equipped with an inverter. Compared with configuring one or several high-power centralized inverters for the entire photovoltaic power station, this configuration can effectively disperse risks and reduce the impact of single-point failures; the photovoltaic panels in each area use multiple The photovoltaic panels in the area are first divided into several groups (also called substrings), and the photovoltaic panels in each group are connected in series to form a series group. The photovoltaic panels in the area can be divided into several substrings, and each substring is numbered as n, where n = 1, 2, 3, ..., N; N is a positive integer, N represents the total number of substrings in the area, and n represents the sequence number of any substring. The position of each substring is recorded as Lmn and marked on the map of the photovoltaic power station, which helps to quickly locate and promptly maintain faulty photovoltaic panels in the future.

[0063] By numbering each area and substring and marking the location of each substring on a map, subsequent fault location and maintenance are facilitated. When a substring PV panel fails, engineers can quickly locate the problem based on the number and location information, thus shortening maintenance time and improving maintenance efficiency.

[0064] Step 2: Communicate with various sensors installed in each area to collect power parameters and environmental parameters for each substring in each area. Based on this, in-depth analysis is performed on the inverter operating status, light energy conversion status, and resistance status of each substring in the area to obtain harmonic distortion, conversion monitoring values, and resistance monitoring values. The power parameters include voltage signal V(t), current signal I(t), and power generation.

[0065] Environmental parameters include ambient temperature, ambient humidity and irradiance;

[0066] 2-1: Monitoring and analysis of inverter:

[0067] The voltage and current signals of the inverter are retrieved, and the sampling frequency is often much higher than the output signal frequency, such as several thousand hertz or higher, to ensure that high-order harmonics can be accurately captured; through Fourier transform (FFT)

[0068] The time domain signal is converted into a frequency domain signal, and the voltage harmonic expression and each current harmonic expression are obtained accordingly. The specific voltage harmonic expression is:

[0069] V(t)=V1·sin(ωt+θ1)+V2·sin(2ωt+θ2)+V3·sin(3ωt+θ3)+…+V J ·sin(Jωt+θ J )

[0070] Where j = 1, 2, 3, ..., J; J is a positive integer, J represents the highest order of voltage harmonics, j represents the number of one of the voltage harmonics; V1 is the voltage amplitude of the fundamental wave, V2, V3, ..., V J is the voltage amplitude of each harmonic, corresponding to the 2nd, 3rd, up to the Jth voltage harmonic; ω is the angular frequency of the fundamental frequency;

[0071] θ1, θ2, θ3, ..., θ J is the phase angle of each voltage harmonic;

[0072] The specific current harmonic expression is:

[0073] I(t)=I1·sin(ωt+φ1)+I2·sin(2ωt+φ2)+I3·sin(3ωt+φ3)+…+I F ·sin(Fωt+φ F )

[0074] Where f = 1, 2, 3, ..., F; F is a positive integer, F represents the highest order of current harmonics, f represents the order number of one of the current harmonics; I1 is the current amplitude of the fundamental wave, I2, I3, ..., I Fis the current amplitude of each harmonic, corresponding to the 2nd, 3rd, up to the Fth current harmonic; ω is the angular frequency of the fundamental frequency;

[0075] φφ1,φ2,φ3,…,φ F is the phase angle of each current harmonic;

[0076] Extract the amplitude V of each harmonic based on the voltage harmonic expression and current harmonic expression j and I f , and perform formulaic calculation and analysis to obtain the harmonic distortion VI; the specific calculation formula is:

[0077]

[0078] where a V 、a I The specific values ​​are set by those skilled in the art according to actual needs, for example, a V The value is 0.35, a I The value is 0.65;

[0079] It should be noted that abnormal harmonic values ​​usually indicate a fault in the inverter;

[0080] 2-2: Monitoring and analysis of sub-string photovoltaic panels:

[0081] Retrieve the irradiance and power generation corresponding to each substring in the area at each collection time; normalize the irradiance and power generation and take their values, divide the values ​​to calculate the conversion rate, and thus obtain the conversion rate corresponding to each collection time, which is recorded as Kb, where b = 1, 2, 3, ..., B; B is a positive integer, B represents the total number of collection times, and b represents the sequence number of any collection time; construct a two-dimensional rectangular coordinate system with time as the horizontal coordinate and the conversion rate as the vertical coordinate, input the conversion rate into the coordinate axis according to its corresponding collection time, and record the position of the conversion rate in the coordinate axis as the conversion point; use a smooth curve to connect each conversion point in turn to obtain the curve of the conversion rate changing with time. A line graph is drawn; a tangent line is drawn at each conversion point, and the slope of the tangent line is obtained by data fitting and recorded as the conversion slope Zb; it should be noted that when the conversion slope is greater than zero, it indicates that the conversion rate is increasing, and when the conversion slope is less than zero, it indicates that the conversion rate is decreasing; the conversion slopes greater than zero are summed to obtain a conversion increase value recorded as H1, and the conversion slopes less than zero are summed and taken to obtain an absolute value to obtain a conversion decrease value recorded as H2; the conversion rate Kb, the conversion slope Zb, the conversion increase value H1, and the conversion decrease value H2 are normalized and their values ​​are taken, and the values ​​are analyzed and calculated formulaically to obtain the conversion monitoring value KZ. The specific calculation formula is:

[0082]

[0083] wherein σ1, σ2, σ3 are the set proportionality constants, respectively, is the average of the conversion slope of each conversion point; it can be seen from the formula that the greater the conversion increase value and the smaller the conversion decrease value, the greater the conversion monitoring value; when the conversion rate is unstable, the greater the risk of failure of the photovoltaic panel, the smaller the conversion monitoring value;

[0084] 2-3: Resistance monitoring analysis of sub-string photovoltaic panels:

[0085] Let the insulation resistance, series impedance, ambient temperature and ambient humidity corresponding to each collection time be Rb 绝缘 , Rb 阻抗 , Tb and Sb, respectively; set a standard ambient parameter for each sub-string photovoltaic panel, wherein the standard ambient parameter includes a standard ambient temperature and a standard ambient humidity, and is denoted by AT and AS, respectively; it should be noted that the standard ambient parameter refers to the environmental condition that has the least impact on the photovoltaic panel, which is determined by the manufacturing process and manufacturing materials of the photovoltaic panel; the environment has a greater impact on the photovoltaic panel, especially the insulation resistance and series impedance are more sensitive to the environment; humidity infiltration will reduce the insulation performance of the photovoltaic panel material and increase the risk of current leakage, especially in harsh weather or high humidity environment; high temperature will exacerbate the oxidation or corrosion of the internal metal connecting parts of the photovoltaic panel, causing poor contact, thereby increasing the impedance and reducing the output power; long-term thermal expansion and contraction may also damage the solder joints and wiring of the photovoltaic panel;

[0086] It should be noted that the insulation resistance refers to the insulation level of the electrical part (such as wire, junction box, etc.) in the photovoltaic module or system to the ground or grounding, the higher the insulation resistance, the better the insulation of the system, which can effectively prevent faults such as leakage and short circuit; the series impedance refers to the total resistance of the circuit between the photovoltaic module and the cable, including the contact resistance and the internal resistance of the photovoltaic cell, the lower the series impedance, the smaller the current loss, and the higher the power generation efficiency; but if the impedance increases (such as poor contact, solder joint problem, etc.), it will increase the current loss of the system and reduce the power output. Generally, by analyzing the change of series impedance, local faults of the module or connecting parts can be detected;

[0087] The insulation resistance Rb 绝缘 , series impedance Rb 阻抗 , ambient temperature Tb, ambient humidity Sb, standard ambient temperature AT and standard ambient humidity AS are normalized and their values are taken, and the values are calculated and analyzed by formula to obtain the resistance state value PR; the specific calculation formula is:

[0088]

[0089] Where β1, β2, β3, and β4 are the set proportional constants, respectively. From this, we can obtain the resistance state value PRb corresponding to each acquisition moment. The formula shows that when the insulation resistance is smaller and the series impedance is larger, the resistance state value is larger, which means that the risk of failure of the photovoltaic panel is greater. When the ambient temperature and humidity deviate from the standard environmental parameters, it means that the environmental conditions at that time have a greater impact on the photovoltaic panel, and the resistance state value is larger.

[0090] A two-dimensional rectangular coordinate system is constructed with time as the abscissa and the resistance state value as the ordinate. The resistance state value is input into the coordinate axis according to its corresponding acquisition time, and the position of the resistance state value in the coordinate axis is recorded as a resistance point. A smooth curve is used to connect each resistance point in sequence to obtain a resistance state value change curve with time; a tangent line is drawn at each resistance point, and the tangent slope is obtained by data fitting and recorded as the resistance slope, recorded as Db; the resistance slopes greater than zero are summed to obtain a state increase value, recorded as H3, and the resistance slopes less than zero are summed and then taken as the absolute value to obtain a state decrease value, recorded as H4; the resistance state value PRb, resistance slope Db, state increase value H3 and state decrease value H4 at each acquisition time are normalized and their values ​​are taken, and the values ​​are formulaically calculated and analyzed to obtain the resistance monitoring value RD. The specific calculation formula is:

[0091]

[0092] Among them, α1, α2, and α3 are the set proportional constants, is the average resistance slope of each resistance point; from the formula, it can be seen that when the resistance state value is larger, the resistance state value increasing trend is greater than the resistance state value decreasing trend, which means that the resistance state of the photovoltaic panel is worse and the fault risk is greater; the larger the resistance monitoring value; when the resistance state value is more volatile (i.e. The larger the value), the greater the resistance monitoring value;

[0093] 2-4: From this, we can determine the three factors for judging the performance of each sub-string of photovoltaic panels in the area. Specifically, the three factors are: harmonic distortion of the inverter, conversion monitoring value, and resistance monitoring value.

[0094] By communicating with various sensors, the power parameters and environmental parameters of each substring are collected, and the data is deeply analyzed based on technical means such as Fourier transform and trend change analysis. The key indicators of harmonic distortion, conversion monitoring value and resistance monitoring value are extracted as the three elements of judgment. It can accurately monitor and analyze the inverter status of the region and the conversion status and resistance status of the photovoltaic panels. This data-driven analysis method can more accurately reflect the working status of the photovoltaic panels and provide strong support for subsequent fault diagnosis and preventive maintenance.

[0095] Step 3: Perform a comprehensive analysis based on the three judgment factors of each sub-string photovoltaic panel in the area to accurately determine the health status of each sub-string photovoltaic panel, identify the sub-string photovoltaic panel with faults, and implement corresponding safety policies accordingly to achieve efficient intelligent monitoring and preventive maintenance; specifically:

[0096] 3-1: The system retrieves the three key elements for determining the performance of each PV panel string in the area: the inverter's harmonic distortion, conversion monitoring value, and resistance monitoring value. The system then compares the harmonic distortion with the set harmonic threshold. If the harmonic distortion is greater than or equal to the set harmonic threshold, the system's power quality has exceeded the safe range, posing a significant risk of equipment damage or reduced efficiency. Therefore, to ensure safety and equipment reliability, the PV panels in the area are shut down to prevent further impacts caused by excessive harmonics. The area is then marked as a risk zone, and its number and location are output. The risk zone number and location are then sent to the appropriate engineer for prompt repair. If the harmonic distortion is less than the set harmonic threshold, the system proceeds to 3-2.

[0097] 3-2: Normalize the harmonic distortion VI, conversion monitoring value KZ, and resistance monitoring value RD and take their values. Calculate and analyze the values ​​using a formula to obtain the abnormal value CY. The specific Pythagorean theorem calculation formula is:

[0098]

[0099] Among them, λ1 and λ2 are the set proportional constants respectively;

[0100] 3-3: Compare and analyze the abnormal value with the set abnormal interval. When the abnormal value is greater than the maximum value in the set abnormal interval, the substring photovoltaic panel is recorded as a high-risk substring; when the abnormal value is in the set abnormal interval, the substring photovoltaic panel is recorded as a medium-risk substring; when the abnormal value is less than the minimum value in the set abnormal interval, the substring photovoltaic panel is recorded as a light-risk substring; count the number of high-risk substrings, medium-risk substrings and light-risk substrings in the area respectively, and record them as G1, G2 and G3 respectively; if G3 ≥ G1 + G2, the area is marked as high-risk substrings, medium-risk substrings and light-risk substrings. High-risk substrings within the domain are recorded as faulty substrings, which are then stopped from working. The faulty substring's number and location are output and sent to the corresponding engineer to facilitate timely repair of the abnormal substring. Otherwise, the area is recorded as an abnormal area, and the abnormal area is stopped from working. The number and location of the abnormal area are output, and the abnormal values ​​of each substring's photovoltaic panels are integrated into abnormal information of the abnormal area. The abnormal information is then sent to the corresponding engineer to facilitate timely maintenance of each substring in the abnormal area.

[0101] 3-4: Repeat the above steps until all areas of the photovoltaic power station are fault-diagnosed and the corresponding safety policies are output;

[0102] Based on a comprehensive analysis of the three elements of each sub-string photovoltaic panel, the health status of each sub-string can be accurately judged, the sub-string with faults can be identified, and the corresponding safety strategy can be implemented accordingly; specifically: when the harmonic distortion is detected to exceed the set threshold, the photovoltaic panels in the relevant area can be immediately stopped to prevent the excessive harmonics from causing further damage to the equipment and system; in addition, by grading the risk level of the sub-string photovoltaic panels, corresponding safety strategies can be adopted for sub-string photovoltaic panels of different risk levels, further improving the safety of the system; realizing intelligent monitoring and preventive maintenance of photovoltaic power stations, preventive measures can be taken before faults occur, reducing the probability and impact range of faults, thereby extending the service life of equipment and improving overall operational efficiency.

[0103] See also Figure 2 In an embodiment of the present invention, a photovoltaic power station fault detection system based on big data includes: a server, a monitoring and analysis module, and a fault output module;

[0104] The server divides the photovoltaic power station into regions and collects information, specifically:

[0105] Since centralized photovoltaic power stations occupy a vast area and contain a huge number of photovoltaic panels, the entire photovoltaic power station is divided into several areas, each of which is numbered m, where m = 1, 2, 3, ..., M; M is a positive integer, M represents the total number of areas in the entire photovoltaic power station, and m represents the sequence number of any area. Each area is equipped with an inverter. Compared with configuring one or several high-power centralized inverters for the entire photovoltaic power station, this configuration can effectively disperse risks and reduce the impact of single-point failures; the photovoltaic panels in each area are connected in series using multiple groups. The photovoltaic panels in the area are first divided into several groups (also called substrings), and the photovoltaic panels in each group are connected in series to form a series group. The photovoltaic panels in the area can be divided into several substrings, and each substring is numbered as n, where n = 1, 2, 3, ..., N; N is a positive integer, N represents the total number of substrings in the area, and n represents the sequence number of any substring. The position of each substring is recorded as Lmn and marked on the map of the photovoltaic power station, which helps to quickly locate and promptly maintain faulty photovoltaic panels in the future.

[0106] By communicating with various sensors installed in various areas, the power parameters and environmental parameters of each substring in each area are collected. The power parameters include voltage signal V(t), current signal I(t), and power generation; the environmental parameters include ambient temperature, ambient humidity, and irradiance.

[0107] The monitoring analysis module respectively performs feature analysis on the inverter state, the conversion state and the resistance state of each sub-string photovoltaic panel in each region based on the power parameters and the environmental parameters to extract three judgment elements; the three judgment elements are specifically: the harmonic distortion degree of the inverter, the conversion monitoring value and the resistance monitoring value.

[0108] The fault output module performs comprehensive analysis on the three judgment elements of each sub-string photovoltaic panel in the region to accurately judge the health state of each sub-string photovoltaic panel, identify the sub-string photovoltaic panel with fault, and execute the corresponding safety strategy accordingly; specifically:

[0109] 3-1: retrieve the three judgment elements of each sub-string photovoltaic panel in the region: the harmonic distortion degree of the inverter, the conversion monitoring value and the resistance monitoring value; if the harmonic distortion degree is greater than or equal to the set harmonic threshold, then control the photovoltaic panel in the region to stop working, and record the region as a risk region, output the number and position of the risk region; send the number and position of the risk region to the corresponding engineer; if the harmonic distortion degree is less than the set harmonic threshold, then execute 3-2;

[0110] 3-2: normalize the harmonic distortion degree VI, the conversion monitoring value KZ and the resistance monitoring value RD and take their values, and perform formulaic calculation and analysis on the values to obtain an abnormal value CY; the specific Pythagorean theorem calculation formula is:

[0111]

[0112] wherein λ1 and λ2 are respectively set proportional constants;

[0113] 3-3: compare and analyze the abnormal value with the set abnormal interval, when the abnormal value is greater than the maximum value in the set abnormal interval, then record the sub-string photovoltaic panel as a high-risk sub-string; when the abnormal value is in the set abnormal interval, then record the sub-string photovoltaic panel as a moderate-risk sub-string; when the abnormal value is less than the minimum value in the set abnormal interval, then record the sub-string photovoltaic panel as a low-risk sub-string; respectively count the number of high-risk sub-strings, moderate-risk sub-strings and low-risk sub-strings in the region, and record them as G1, G2 and G3 respectively; if G3≥G1+G2, then record the high-risk sub-strings in the region as fault sub-strings, control the fault sub-strings to stop working, and output the number and position of the fault sub-strings, send the number and position of the fault sub-strings to the corresponding engineer; otherwise, record the region as an abnormal region, control the abnormal region to stop working; and output the number and position of the abnormal region, thereby integrating the number, position and abnormal value of each sub-string photovoltaic panel in the abnormal region into abnormal information of the abnormal region, and sending the abnormal information to the corresponding engineer;

[0114] 3-4: Repeat steps 3-1 to 3-3 until all areas of the photovoltaic power station are fault-diagnosed and the corresponding safety policies are output.

[0115] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the steps of the photovoltaic power station fault detection method based on big data.

[0116] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), magnetic disk or optical disk, and other media that can store program code.

[0118] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that the technical solutions described in the above embodiments can still be modified, or some of the technical features thereof can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A photovoltaic power station fault detection method based on big data, characterized in that: The following steps are involved: Step 1: Divide the PV power station into several areas. Each area is equipped with an inverter. The PV panels within the area are divided into several substrings. The substring PV panels are connected in parallel. The PV panels within a substring are connected in series. The PV panels in each substring are numbered and marked on the PV power station map. Step 2: Communicate with various sensors installed in each area to collect power parameters and environmental parameters for each substring in each area. Based on this, in-depth analysis is performed on the inverter operating status, light energy conversion status, and resistance status of each substring in the area to obtain harmonic distortion, conversion monitoring value, and resistance monitoring value. This can be used to determine the three key factors for judging the performance of each substring photovoltaic panel in the area: the harmonic distortion of the inverter, the conversion monitoring value, and the resistance monitoring value. The power parameters include the voltage signal V(t), the current signal I(t), and the power generation; the environmental parameters include the ambient temperature, ambient humidity, and irradiance. Step 3: Perform a comprehensive analysis based on the three judgment factors of each sub-string photovoltaic panel in the area to accurately determine the health status of each sub-string photovoltaic panel, identify the sub-string photovoltaic panel with faults, and implement the corresponding safety strategy accordingly; the specific steps are as follows: 3-1: Query the three key factors for determining the quality of each PV panel string in the area: the inverter's harmonic distortion, conversion monitoring value, and resistance monitoring value. If the harmonic distortion is greater than or equal to the set harmonic threshold, the PV panels in the area are stopped, the area is marked as a risk zone, and the risk zone number and location are output. The risk zone number and location are sent to the corresponding engineer. If the harmonic distortion is less than the set harmonic threshold, execute 3-2. 3-2: Normalize the harmonic distortion, conversion monitoring value, and resistance monitoring value and take their values. Perform formula calculation and analysis on the values ​​to obtain abnormal values. 3-3: Compare and analyze the abnormal value with the set abnormal interval. When the abnormal value is greater than the maximum value in the set abnormal interval, the photovoltaic panel substring is marked as a high-risk substring; when the abnormal value is within the set abnormal interval, the photovoltaic panel substring is marked as a moderate-risk substring; When the outlier value is less than the minimum value in the set outlier interval, the photovoltaic panel substring is marked as a mild risk substring; the number of high-risk substrings, medium-risk substrings, and mild-risk substrings in the area is counted and recorded as G1, G2, and G3, respectively; if G3 ≥ G1 + G2, the high-risk substring in the area is marked as a faulty substring, the faulty substring is stopped, and the faulty substring number and location are output and sent to the corresponding engineer; Otherwise, the area is marked as an abnormal area, and the control area stops working; and the number and location of the abnormal area are output. Thus, the number, location, and abnormal values ​​of each substring photovoltaic panel of the abnormal area are integrated into the abnormal information of the abnormal area, and the abnormal information is sent to the corresponding engineer; 3-4: Repeat steps 3-1 to 3-3 until all areas of the photovoltaic power station are fault-diagnosed and the corresponding safety policies are output.

2. The photovoltaic power station fault detection method based on big data according to claim 1 is characterized in that: The monitoring and analysis steps for the inverter working status are as follows: The voltage signal V(t) and current signal I(t) of the inverter are retrieved, and the time domain signals are converted into frequency domain signals through Fourier transform, and the voltage harmonic expression and each subharmonic expression of the current are obtained accordingly; Extract the amplitude V of each harmonic based on the voltage harmonic expression and current harmonic expression j and I f And the harmonic distortion will be obtained by formula calculation and analysis.

3. The photovoltaic power station fault detection method based on big data according to claim 2, characterized in that: The steps for monitoring and analyzing the conversion status are as follows: Retrieve the irradiance and power generation corresponding to each substring in the area at each collection time; normalize the irradiance and power generation and take their numerical values, divide the numerical values ​​to calculate the conversion rate, and thus obtain the conversion rate corresponding to each collection time; construct a two-dimensional rectangular coordinate system with time as the horizontal coordinate and the conversion rate as the vertical coordinate, input the conversion rate into the coordinate axis according to its corresponding collection time, and record the position of the conversion rate in the coordinate axis as the conversion point; use a smooth curve to connect each conversion point in sequence to obtain a curve graph of the conversion rate changing with time; draw a tangent line at each conversion point, use data fitting to calculate the tangent slope and record it as the conversion slope; sum the conversion slopes greater than zero to calculate the conversion increase value, and sum the conversion slopes less than zero and take the absolute value to calculate the conversion decrease value; The conversion rate, conversion slope, conversion increase value and conversion decrease value are normalized and their numerical values ​​are taken, and the numerical values ​​are subjected to formulaic calculation and analysis to obtain the conversion monitoring value.

4. The photovoltaic power station fault detection method based on big data according to claim 3 is characterized in that: The monitoring and analysis steps for resistance status are as follows: Set each substring photovoltaic panel to correspond to a standard environmental parameter, where the standard environmental parameters include standard ambient temperature and standard ambient humidity; The insulation resistance, series impedance, ambient temperature and ambient humidity corresponding to each acquisition moment are normalized and their values ​​are taken, and the values ​​are calculated and analyzed in a formula to obtain the resistance state value; A two-dimensional rectangular coordinate system is constructed with time as the abscissa and resistance state value as the ordinate. The resistance state value is input into the coordinate axis according to its corresponding acquisition time, and the position of the resistance state value in the coordinate axis is recorded as a resistance point. A smooth curve is used to connect each resistance point in sequence to obtain a resistance state value variation curve over time; a tangent line is drawn at each resistance point, and the slope of the tangent line is obtained by data fitting and recorded as the resistance slope; the resistance slopes greater than zero are summed to obtain the state increase value, and the resistance slopes less than zero are summed and then the absolute value is taken to obtain the state decrease value; The resistance state value, resistance slope, state increase value and state decrease value at each acquisition moment are normalized and their values ​​are taken, and the values ​​are calculated and analyzed in a formula to obtain the resistance monitoring value.

5. Photovoltaic power station fault detection system based on big data, characterized by: Applied to the photovoltaic power station fault detection method based on big data as claimed in any one of claims 1 to 4, the system comprises: a server, a monitoring and analysis module and a fault output module; The server divides the photovoltaic power station into regions and collects information. The specific information collected is power parameters and environmental parameters. Power parameters include voltage signal V(t), current signal I(t), and power generation; environmental parameters include ambient temperature, ambient humidity, and irradiance. The monitoring and analysis module performs feature analysis on the inverter status, conversion status, and resistance status of each substring photovoltaic panel in each area based on power parameters and environmental parameters to extract the three judgment factors. The three judgment factors are: the harmonic distortion of the inverter, the conversion monitoring value, and the resistance monitoring value. The fault output module performs a comprehensive analysis based on the three judgment factors of each sub-string photovoltaic panel in the area to accurately determine the health status of each sub-string photovoltaic panel, identify the sub-string photovoltaic panel with a fault, and implement the corresponding safety strategy accordingly; specifically: 3-1: Query the three key factors for determining the quality of each PV panel string in the area: the inverter's harmonic distortion, conversion monitoring value, and resistance monitoring value. If the harmonic distortion is greater than or equal to the set harmonic threshold, the PV panels in the area are stopped, the area is marked as a risk zone, and the risk zone number and location are output. The risk zone number and location are sent to the corresponding engineer. If the harmonic distortion is less than the set harmonic threshold, execute 3-2. 3-2: Normalize the harmonic distortion, conversion monitoring value, and resistance monitoring value and take their values. Perform formula calculation and analysis on the values ​​to obtain abnormal values. 3-3: Compare and analyze the abnormal value with the set abnormal interval. When the abnormal value is greater than the maximum value in the set abnormal interval, the substring photovoltaic panel is recorded as a high-risk substring; when the abnormal value is in the set abnormal interval, the substring photovoltaic panel is recorded as a medium-risk substring; when the abnormal value is less than the minimum value in the set abnormal interval, the substring photovoltaic panel is recorded as a light-risk substring; count the number of high-risk substrings, medium-risk substrings and light-risk substrings in the area respectively, and record them as G1, G2 and G3 respectively; if G3≥G1+G2, the high-risk substring in the area is recorded as a faulty substring, the faulty substring is controlled to stop working, and the number and location of the faulty substring are output, and the number and location of the faulty substring are sent to the corresponding engineer; otherwise, the area is recorded as an abnormal area, the abnormal area is controlled to stop working, and the number and location of the abnormal area are output, thereby integrating the number, location and abnormal values ​​of each substring photovoltaic panel into the abnormal information of the abnormal area, and the abnormal information is sent to the corresponding engineer; 3-4: Repeat steps 3-1 to 3-3 until all areas of the photovoltaic power station are fault-diagnosed and the corresponding safety policies are output.

6. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the photovoltaic power station fault detection method based on big data according to any one of claims 1 to 4 is implemented.

Images