A method and system for online fault diagnosis of a photovoltaic module

Abstract

The application discloses a kind of photovoltaic module online fault diagnosis method and system thereof, including steps: S1, the PWM ripple current signal I pv ; S2, the PWM ripple current signal I pv It is filtered by conditioning circuit module;S3, real-time online extraction photovoltaic module ripple impedance Z i And phase φ i ;S4, based on the photovoltaic module ripple impedance Z i And phase φ i It is compared with the average of photovoltaic module reference ripple impedance Z ref And phase average φ ref Respectively, according to the contrast difference amount situation, the fault of photovoltaic module is diagnosed online;The application can be implemented without interfering with the normal operation of photovoltaic power generation system, without additional signal injection and high-bandwidth voltage sensor, can significantly reduce hardware complexity and engineering implementation cost, and the online fault diagnosis method provided by the application has the advantages that fault diagnosis sensitivity is high, and the system has strong anti-interference, can significantly reduce the influence of environmental factors on the diagnosis result.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic modules, and more specifically to an online fault diagnosis method for photovoltaic modules. This invention also relates to an online fault diagnosis system employing this online fault diagnosis method for photovoltaic modules. Background Technology

[0002] As the core component of a photovoltaic power generation system, the reliability of photovoltaic modules directly affects the system's power generation efficiency and safety. Existing fault diagnosis methods for photovoltaic modules mainly include the IV curve method, mathematical model method, infrared image analysis method, and intelligent diagnostic algorithms. The IV curve method diagnoses faults by measuring the current-voltage characteristic curve and can identify various fault types, but it requires disconnecting the system and cannot provide real-time diagnosis. It also has low sensitivity for faults such as aging, and is not applicable to catastrophic faults such as arcing. The mathematical model method compares physical models with actual data, providing a solid theoretical foundation, but the models are complex, parameter estimation is difficult, and it is susceptible to environmental interference. Infrared image analysis detects temperature distribution non-contactly, but the equipment is expensive, it cannot provide real-time diagnosis, and it is insensitive to early-stage faults. Intelligent diagnostic algorithms, with their powerful nonlinear mapping and feature mining capabilities, can diagnose faults without relying on precise physical models. However, they depend on artificial intelligence tools that use training data. Furthermore, they are system-specific, with much of the collected data only applicable to specific installation systems, and they have high requirements for embedded hardware resources, making practical application difficult. Existing fault diagnosis methods suffer from a dilemma in balancing real-time performance, sensitivity, and cost: insufficient real-time performance, with most methods requiring system interruption or relying on periodic detection, making millisecond-level online diagnosis impossible; and a conflict between cost and sensitivity, with high-precision methods (such as infrared imaging) being expensive and low-cost methods (such as IV curves) having low sensitivity.

[0003] Therefore, the applicant hopes to find a technical solution to address the above technical problems. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide an online fault diagnosis method and system for photovoltaic modules, which can be implemented without interfering with the normal operation of the photovoltaic power generation system, without the need for additional signal injection and high-bandwidth voltage sensors, and can significantly reduce hardware complexity and engineering implementation costs. Moreover, the online fault diagnosis method provided by this invention has the advantage of high fault diagnosis sensitivity.

[0005] The technical solution adopted in this invention is as follows:

[0006] A method for online fault diagnosis of photovoltaic modules, wherein a Boost circuit is provided between the photovoltaic module and the load, and the Boost circuit includes a DC-DC switching converter connected to the photovoltaic module. The online fault diagnosis method includes the following steps:

[0007] S1. A PWM ripple current is generated by a DC-DC switching converter, and the PWM ripple current signal I flowing through the photovoltaic module is acquired by a current sensor. pv ;

[0008] S2. The PWM ripple current signal I acquired in step S1 is processed by the conditioning circuit module. pv Perform filtering and conditioning to retain the PWM ripple component current signal i located at the switching frequency. pv-pwm ;

[0009] S3, the PWM ripple component current signal i is processed by the data acquisition module. pv-pwm Data is collected and then calculated by a host computer to extract the ripple impedance Z of the photovoltaic module in real time. i and its phase φ i ;

[0010] S4. Based on the real-time online extraction of photovoltaic module ripple impedance Z i and its phase φ i The average ripple impedance Z of the photovoltaic module reference ref and its phase mean φ ref The photovoltaic modules are compared separately, and the faults are diagnosed online based on the differences in the comparisons.

[0011] Preferably, the photovoltaic module includes a single photovoltaic module or a photovoltaic module string formed by connecting multiple photovoltaic modules in series. When multiple photovoltaic module strings are connected in parallel, online fault diagnosis is performed on each photovoltaic module string. And / or the switching frequency range of the DC-DC switching converter is 10-50kHz, and / or the conditioning circuit module adopts a passive bandpass filter with a passband range of 1kHz-100kHz, and / or the sampling frequency of the data acquisition module is greater than 200kHz.

[0012] Preferably, in step S2, the conditioning circuit module is used to filter out the PWM ripple current signal I. pv The DC component, low-frequency variation component, and high-frequency noise in the data.

[0013] Preferably, in step S3, the ripple impedance Z of the photovoltaic module i and its phase φ i The calculation process includes the following:

[0014] Estimate the frequency domain value V of the photovoltaic module terminal ripple voltage using the following formula. pv (jω):

[0015] ;

[0016] Among them, Vdc (jω) represents the frequency domain of the ripple voltage at the load terminal, jω is the imaginary part of the complex angular frequency in frequency domain analysis, where ω = 2πf, ω is the angular frequency, f is the frequency, L is the inductance value in the Boost circuit, and i pv (jω) represents the frequency domain value of the ripple current at the photovoltaic module terminal, which is the PWM ripple component current signal i. pv-pwm C in This refers to the input-side filter capacitor value between the photovoltaic module and the DC-DC switching converter.

[0017] Then, the ripple impedance Z of the photovoltaic module is calculated and extracted according to the following formula. i and its phase φ i :

[0018] ;

[0019] Among them, Z pv_pwm This is the ripple impedance Z of the photovoltaic module. i v pv_pwm This is the frequency domain value V of the ripple voltage at the photovoltaic module terminals. pv (jω), i pv_pwm This refers to the frequency domain value i of the ripple current at the photovoltaic module terminals. pv (jω), φ pv_pwm That is, the ripple impedance phase φ of the photovoltaic module. i .

[0020] Preferably, in step S3, the frequency domain ripple voltage V at the load terminal is estimated according to the following formula. dc (jω):

[0021] ;

[0022] Among them, V dc (s) is the frequency domain of the ripple voltage at the load end, V. dc (jω);V D φ is the DC voltage input to the load, D is the duty cycle of the DC-DC switching converter in the Boost circuit, and φ is the input DC voltage to the load. Vdc This represents the phase of the ripple voltage at the load end.

[0023] Preferably, in step S4, the average reference ripple impedance Z of the photovoltaic module ref The impedance value of a known reference healthy photovoltaic module can be used directly, or the following calculation process can be used:

[0024] ;

[0025] Among them, Z ref Z is the average amplitude of the reference ripple impedance of the photovoltaic module under normal conditions.ref Z k The ripple impedance Z of the photovoltaic module is extracted online in real time when the k-th photovoltaic module is in normal condition. i N is equal to or greater than 1, which is the number of photovoltaic modules in normal state participating in the calculation;

[0026] The photovoltaic module reference ripple impedance phase mean φ ref The impedance phase value of a known reference healthy photovoltaic module can be used directly, or the following calculation process can be used:

[0027] ;

[0028] Where, φ ref The mean amplitude of the reference ripple impedance phase of the photovoltaic module under normal conditions is φ. ref φ k The phase φ of the ripple impedance of the k-th photovoltaic module is extracted online in real time when the module is in normal condition. i .

[0029] Preferably, the relative change in impedance amplitude Δ|Zi| and the relative change in impedance phase Δφi of each photovoltaic module are calculated using the following formulas respectively;

[0030] ;

[0031] Furthermore, the standard deviation of the impedance magnitude σ(|Zi|) is calculated according to the following process:

[0032] First, calculate the arithmetic mean of the impedance amplitude of each photovoltaic module within the sampling window. ;

[0033] ;

[0034] Then, the standard deviation of the impedance amplitude σ(|Zi|) is calculated based on this arithmetic mean:

[0035] ;

[0036] Where j is the total number of discrete data points included in the sampling window, i.e., the sample size, and m is the discrete time index of the data points in the sampling window, m=1,2,3,…,j, Z m This is the calculated value of the photovoltaic module ripple impedance corresponding to the m-th discrete sampling point within the sampling window.

[0037] Preferably, in step S4, the online diagnosis of photovoltaic module faults based on the comparison difference includes the following process:

[0038] The presence of local anomalies is determined by comparing the differences in data. If a local anomaly is detected, it is then classified according to the fault's characteristic thresholds.

[0039] When the amplitude standard deviation σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc Furthermore, if the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th, it is determined that there is no fault.

[0040] When the impedance amplitude standard deviation σ(|Zi|) is found to exceed its preset standard deviation threshold σ arc When this occurs, it is determined to be an arc fault;

[0041] When the standard deviation of the impedance amplitude σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc However, if the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th and the relative change in impedance phase Δφi is not greater than 0, it is judged as an obstruction fault.

[0042] When the standard deviation of the impedance amplitude σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc However, if the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th and the relative change in impedance phase Δφi is greater than 0, it is judged as an aging fault.

[0043] The preset thresholds are adjusted based on system conditions and system parameters.

[0044] Preferably, a Boost circuit is provided between the photovoltaic module and the load, and the Boost circuit includes a DC-DC switching converter connected to the photovoltaic module; the photovoltaic module online fault diagnosis system adopts the photovoltaic module online fault diagnosis method described above, including:

[0045] The conditioning circuit module is connected between the current sensor and the data acquisition module;

[0046] The data acquisition module is connected between the conditioning circuit module and the host computer;

[0047] The host computer includes a ripple voltage estimation module, an impedance calculation module, and a fault diagnosis module.

[0048] The ripple voltage estimation module is used to estimate the ripple voltage at the photovoltaic module terminals in real time.

[0049] The impedance calculation module is used to extract the ripple impedance Z of the photovoltaic module in real time online. i and its phase φ i ;

[0050] The fault diagnosis module is used to diagnose faults in photovoltaic modules online based on the comparison difference.

[0051] The photovoltaic module includes a single photovoltaic module or a photovoltaic module string formed by connecting multiple photovoltaic modules in series.

[0052] Preferably, a Hall current sensor is installed on the connection cable between the photovoltaic module and the DC-DC switching converter; a filter capacitor is connected to the input side between the photovoltaic module and the DC-DC switching converter, and the filter capacitor is connected in parallel with the DC-DC switching converter.

[0053] The main advantages of this application are as follows: This method utilizes a current sensor (usually a built-in device in a photovoltaic power generation system) to acquire the inherent PWM ripple current signal of the DC-DC switching converter online, and then filters and conditions it through a conditioning circuit module to obtain the PWM ripple component current signal i located at the switching frequency. pv-pwm (Current ripple signal that can be used for subsequent impedance analysis), based on this characteristic signal, the ripple impedance Z of the photovoltaic module can be extracted online in real time. i and its phase φ i Finally, the extracted data is used as a measure of impedance characteristic changes to diagnose photovoltaic module faults online based on the comparative differences. This method can be implemented without interfering with the normal operation of the photovoltaic power generation system. Relying on the high switching frequency, it can achieve millisecond-level real-time online continuous monitoring without the need for additional signal injection and high-bandwidth voltage sensors, which can significantly reduce hardware complexity and engineering implementation costs. Moreover, since a photovoltaic module fault directly affects the high-frequency impedance dominated by the diffusion capacitance of the photovoltaic cell, thus causing significant impedance characteristic changes, the online fault diagnosis method provided by this invention has the advantage of high fault diagnosis sensitivity. In addition, the online fault diagnosis system for photovoltaic modules provided by this invention has strong anti-interference capabilities, which can significantly reduce the impact of environmental factors on the diagnostic results. Furthermore, the system is widely applicable to the detection of various common photovoltaic module fault types, including shading, aging, and arcing, demonstrating good versatility and reliability. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating the steps of the online fault diagnosis method for photovoltaic modules according to a specific embodiment of this application;

[0055] Figure 2 This is a schematic diagram of the photovoltaic power generation system architecture used in the online fault diagnosis method for photovoltaic modules according to a specific embodiment of this application;

[0056] Figure 3 This is a schematic diagram of the architecture of the photovoltaic module online fault diagnosis system according to a specific embodiment of this application;

[0057] Figure 4 This is a schematic diagram of the circuit structure of the conditioning circuit module according to a specific embodiment of this application;

[0058] Figure 5 This is a flowchart illustrating the online diagnosis of photovoltaic module faults based on the comparison difference under a specific implementation of this application;

[0059] Figure 6 This is the online fault diagnosis experimental system for photovoltaic modules built according to a specific embodiment of this application;

[0060] Figure 7 These are the experimental results of the first group of fault simulations in the specific embodiments of this application;

[0061] Figure 8 This is the experimental result of the second group of fault simulations in a specific embodiment of this application. Detailed Implementation

[0062] Please refer to the above. Figure 1 , Figure 2 and Figure 3 As shown, this embodiment proposes an online fault diagnosis method for photovoltaic modules, where the photovoltaic module and the load (i.e., the load) are connected. Figure 2 A Boost circuit is installed between the marked inverter and the power grid. The Boost circuit contains a DC-DC switching converter (i.e., a converter connected to the photovoltaic modules) Figure 2 The "DC-DC" designation, preferably, in this embodiment, refers to a photovoltaic module comprising a single photovoltaic module or a photovoltaic module string formed by connecting multiple photovoltaic modules in series. When multiple parallel photovoltaic module strings are provided, online fault diagnosis is performed on each photovoltaic module string separately. Please refer to [link to previous document]. Figure 2 As shown, the photovoltaic module consists of three parallel photovoltaic module strings, namely... Figure 2 The first photovoltaic module string PV1, the second photovoltaic module string PV2, and the third photovoltaic module string PV3 are marked; these photovoltaic module strings can all be used for online fault diagnosis using the scheme described below in this embodiment.

[0063] Preferably, please further combine Figure 3 As shown, this embodiment proposes an online fault diagnosis system for photovoltaic modules, including:

[0064] The conditioning circuit module is connected between the current sensor and the data acquisition module;

[0065] The data acquisition module is connected between the conditioning circuit module and the host computer;

[0066] The host computer includes a ripple voltage estimation module (i.e., Figure 3 The modules marked "Photovoltaic string ripple voltage estimation module", impedance calculation module and fault diagnosis module;

[0067] The ripple voltage estimation module is used to estimate the ripple voltage at the photovoltaic module terminals in real time (i.e., the voltage at the module terminals). Figure 3 The marked "v" pv_pwm (”);

[0068] The impedance calculation module is used to extract the ripple impedance Z of photovoltaic modules in real time online. i (that is) Figure 3 The marked "Photovoltaic string ripple impedance Z" pv ) and its phase φ i ;

[0069] The fault diagnosis module is used to diagnose photovoltaic module faults online based on the comparison difference.

[0070] Photovoltaic modules include single photovoltaic modules or photovoltaic modules connected in series to form a photovoltaic module string.

[0071] Preferably, in this embodiment, the photovoltaic module (i.e., Figure 3 The labeled "photovoltaic string" and DC-DC switching converter (i.e. Figure 3 A Hall current sensor is installed on the connecting cable between the photovoltaic module and the DC-DC switching converter (marked at "Duty Cycle D"); a filter capacitor C is connected to the input side between the photovoltaic module and the DC-DC switching converter. in The filter capacitor C in Connected in parallel with a DC-DC switching converter.

[0072] In this embodiment, the online fault diagnosis method includes the following steps:

[0073] S1. A PWM ripple current is generated by a DC-DC switching converter, and the PWM ripple current signal I flowing through the photovoltaic module is acquired by a current sensor. pv Preferably, in this embodiment, the switching frequency range of the DC-DC switching converter is 10-50kHz.

[0074] S2, via the conditioning circuit module (i.e.) Figure 2 The "conditioning circuit" (marked) applies the PWM ripple current signal I acquired in step S1. pv Perform filtering and conditioning to retain the PWM ripple component current signal i located at the switching frequency. pv-pwm Preferably, in this embodiment, the conditioning circuit module employs a passive bandpass filter with a passband range of 1kHz-100kHz, used to filter out the PWM ripple current signal I. pv The DC component, low-frequency variation component, and high-frequency noise in the filter; more preferably, the passive bandpass filter adopts a second-order passive bandpass filter circuit, please refer to [reference needed]. Figure 4As shown, a first capacitor C1 and a second resistor R2 are provided between the positive terminal IN and the positive terminal OUT, and a first resistor R1 is provided between the negative terminal IN and the first capacitor C1, and a second capacitor C2 is provided between the second resistor R2 and the negative terminal OUT; it should be noted that... Figure 4 The specific working parameters shown are only one example and are not intended to limit the implementation of this application.

[0075] S3, the PWM ripple component current signal i is processed by the data acquisition module. pv-pwm Data is collected and then calculated by a host computer to extract the ripple impedance Z of the photovoltaic module in real time. i and its phase φ i Preferably, in this embodiment, the sampling frequency of the data acquisition module is greater than 200kHz;

[0076] Preferably, in step S3, the ripple impedance Z of the photovoltaic module i and its phase φ i The calculation process includes the following:

[0077] Estimate the frequency domain value V of the photovoltaic module terminal ripple voltage using the following formula. pv (jω):

[0078] ;

[0079] Among them, V dc (jω) represents the frequency domain of the ripple voltage at the load terminal, jω is the imaginary part of the complex angular frequency in frequency domain analysis, where ω = 2πf, ω is the angular frequency, f is the frequency, L is the inductance value in the Boost circuit, and i pv (jω) represents the frequency domain value of the ripple current at the photovoltaic module terminal, which is the PWM ripple component current signal i. pv-pwm C in This refers to the input-side filter capacitor value between the photovoltaic module and the DC-DC switching converter.

[0080] Then, the ripple impedance Z of the photovoltaic module is calculated and extracted according to the following formula. i and its phase φ i :

[0081] ;

[0082] Among them, Z pv_pwm This is the ripple impedance Z of the photovoltaic module. i v pv_pwm This is the frequency domain value V of the ripple voltage at the photovoltaic module terminals. pv (jω), i pv_pwm This refers to the frequency domain value i of the ripple current at the photovoltaic module terminals. pv (jω), φpv_pwm That is, the ripple impedance phase φ of the photovoltaic module. i .

[0083] More preferably, in step S3, the frequency domain ripple voltage V at the load terminal is estimated according to the following formula. dc (jω):

[0084] ;

[0085] Among them, V dc (s) is the frequency domain of the ripple voltage at the load end, V. dc (jω);V D The DC voltage input to the load (i.e., Figure 3 The marked "V" dc ", D is the duty cycle of the DC-DC switching converter in the Boost circuit, φ Vdc This represents the phase of the ripple voltage at the load end.

[0086] S4. Based on the real-time online extraction of photovoltaic module ripple impedance Z i and its phase φ i The average ripple impedance Z of the photovoltaic module reference ref and its phase mean φ ref The photovoltaic modules are compared separately, and the faults are diagnosed online based on the differences in the comparisons.

[0087] Preferably, in step S4, the average reference ripple impedance Z of the photovoltaic module is... ref The calculation process is as follows:

[0088] ;

[0089] Among them, Z ref Z is the average amplitude of the reference ripple impedance of the photovoltaic module under normal conditions. ref Z k The ripple impedance Z of the photovoltaic module is extracted online in real time when the k-th photovoltaic module is in normal condition. i N is equal to or greater than 1, representing the number of photovoltaic modules in normal condition participating in the calculation; in other embodiments, the average ripple impedance Z of the photovoltaic modules is used as the reference. ref Alternatively, the impedance value of a known reference healthy photovoltaic module can be used directly;

[0090] Preferably, in step S4, the average phase value φ of the reference ripple impedance of the photovoltaic module ref The calculation process is as follows:

[0091] ;

[0092] Where, φref The mean amplitude of the reference ripple impedance phase of the photovoltaic module under normal conditions is φ. ref φ k The phase φ of the ripple impedance of the k-th photovoltaic module is extracted online in real time when the module is in normal condition. i In other embodiments, the photovoltaic module references the average phase value of the ripple impedance φ. ref Alternatively, the impedance phase value of a known reference healthy photovoltaic module can be used directly, or

[0093] Preferably, in step S4, the relative change in impedance amplitude Δ|Zi| and the relative change in impedance phase Δφi of each photovoltaic module are calculated using the following formulas;

[0094] ;

[0095] Furthermore, the standard deviation of the impedance magnitude σ(|Zi|) is calculated according to the following process:

[0096] First, calculate the arithmetic mean of the impedance amplitude of each photovoltaic module within the sampling window. ;

[0097] ;

[0098] Then, the standard deviation of the impedance amplitude σ(|Zi|) is calculated based on this arithmetic mean:

[0099] ;

[0100] Where j is the total number of discrete data points included in the sampling window, i.e., the sample size, and m is the discrete time index of the data points in the sampling window, m=1,2,3,…,j, Z m This is the calculated value of the photovoltaic module ripple impedance corresponding to the m-th discrete sampling point within the sampling window;

[0101] The relative change in impedance amplitude Δ|Zi|, the relative change in impedance phase Δφi, and the standard deviation of impedance amplitude σ(|Zi|) obtained from the above calculations can be used as the comparison difference in the embodiments of this application.

[0102] Preferably, please refer to [see also] Figure 5 As shown, in step S4, the online diagnosis of photovoltaic module faults based on the comparison difference includes the following process:

[0103] Based on the comparison of differences, determine whether there are any local anomalies (i.e., Figure 5 The "Differential Calculation: Δ|Zi|, Δφi, σ(|Zi|)" shown above, if a local anomaly is detected, is then classified according to the fault's characteristic threshold; among which,

[0104] When the amplitude standard deviation σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc And the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th (i.e., ... Figure 5 The figure "σ(|Zi|)<σ" is shown. arc When Δ|Zi| ∈ [-th, +th] is true, it is determined that there is no fault;

[0105] When the impedance amplitude standard deviation σ(|Zi|) is found to exceed its preset standard deviation threshold σ arc (that is) Figure 5 The figure "σ(|Zi|)≥σ" is shown. arc When the condition is "yes", it is determined to be an arc fault;

[0106] When the standard deviation of the impedance amplitude σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc However, the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th, and the relative change in impedance phase Δφi is not greater than 0 (i.e., it is...). Figure 5 The figure "σ(|Zi|)≥σ" is shown. arc When "no" and "Δ|Zi|≥th" are yes, and "Δφi>0" are no", it is determined to be an occlusion fault;

[0107] When the standard deviation of the impedance amplitude σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc However, the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th and the relative change in impedance phase Δφi is greater than 0 (i.e., it is...). Figure 5 The figure "σ(|Zi|)≥σ" is shown. arc When "no" and "Δ|Zi|≥th" are yes, and "Δφi>0" are yes, it is determined to be an aging fault;

[0108] The preset thresholds are adjusted and set according to system conditions and system parameters, and this embodiment does not impose any special restrictions.

[0109] To enable those skilled in the art to better understand the technical solutions of this invention, based on the above implementation schemes, the following specific embodiments will be proposed in conjunction with the accompanying drawings of the embodiments of this invention. Obviously, the described embodiments are merely some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this invention.

[0110] Based on the above implementation scheme, the embodiments of this application specifically construct as follows: Figure 6The online fault diagnosis experimental system for photovoltaic modules shown is as follows:

[0111] Photovoltaic strings (i.e., Figure 6 The "photovoltaic modules" shown: 3 x 5W photovoltaic panels ( Figure 7 and Figure 8 (Represented by PV1, PV2, and PV3 respectively), where STC parameter (Standard Test Conditions): open circuit voltage V oc =21.67V, short-circuit current I sc =0.32A;

[0112] Boost circuit, DC-DC switching converter, with a switching frequency of 25kHz, is... Figure 6 The "boost converter" shown is controlled by a TI TMS320F28335 DSP controller.

[0113] Conditioning circuit (i.e., "conditioning circuit module"): Second-order passive bandpass filter, passband 1kHz-100kHz;

[0114] Data acquisition module: PICO virtual oscilloscope, sampling frequency of 500kHz, which transmits the acquired signal to the host computer.

[0115] The following two sets of fault simulations were performed on each PV1, PV2, and PV3 photovoltaic panel, and online fault diagnosis was conducted using the method provided in the above embodiments of this application, as detailed below:

[0116] First set of fault simulation process:

[0117] During the shading failure experiment, two shading sheets were placed sequentially on the surface of PV3 photovoltaic panel, while the other two photovoltaic panels, PV1 and PV2, were left unshaded. The experiment was conducted at an air temperature of T=25℃ and an irradiance of G=900W / m². 2 Conducted in an environment;

[0118] The results of the fault simulation experiment are as follows Figure 7 As shown, when t=1.8s, the first shading sheet is placed, and the shading area of ​​the PV3 photovoltaic panel is approximately 50% of that of a single cell in the module. The impedance amplitude Z of the shaded PV3 photovoltaic panel string is... pv_pwm Compared to a normal string, it will increase, and its corresponding phase φ pv_pwm It suddenly decreases and then tends to stabilize; when t=4.6s, the second shading sheet is placed, and the cumulative shading ratio of the PV3 photovoltaic panel is 100%. The impedance amplitude Z of the shaded PV3 photovoltaic panel string is... pv_pwm As it continues to increase, its corresponding phase φ pv_pwmThen it continues to decrease; when the standard deviation of the impedance amplitude σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc Furthermore, the impedance phase shows a decreasing trend; therefore, it can be determined that the PV3 photovoltaic panel photovoltaic string has experienced a shading fault.

[0119] The second set of fault simulation procedures: The series resistance of the PV3 photovoltaic panel was increased by manually adjusting the sliding potentiometer knob, thereby simulating the aging process of the PV3 photovoltaic panel. The PV1 and PV2 photovoltaic panels were in normal condition. The experiment was conducted at an air temperature of T=25℃ and an irradiance of G=900W / m². 2 Conducted in an environment;

[0120] The results of the fault simulation experiment are as follows Figure 8 As shown in the figure, it can be seen that as the aging of the PV3 photovoltaic panel increases, the impedance amplitude Z of the PV3 photovoltaic panel string decreases. pv_pwm Compared to normal string configurations, the impedance phase φ gradually increases. pv_pwm Compared to the gradual increase in normal string counts, it can be determined that the PV3 photovoltaic panel string has experienced aging failure.

[0121] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0122] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for online fault diagnosis of photovoltaic modules, characterized in that, A Boost circuit is provided between the photovoltaic module and the load. The Boost circuit includes a DC-DC switching converter connected to the photovoltaic module. The online fault diagnosis method includes the following steps: S1, generating PWM ripple current through the DC-DC switching converter, collecting the PWM ripple current signal I flowing through the photovoltaic module through the current sensor pv ; S2, filtering the PWM ripple current signal I collected in step S1 through a conditioning circuit module pv filtering and conditioning, retaining the PWM ripple component current signal i at the switching frequency pv-pwm ; S3, the PWM ripple component current signal i pv-pwm The data acquisition module is used to collect the current signal, and then the upper computer is used for calculation, and the photovoltaic module ripple impedance Z is extracted in real time online i And the phase φ i ; S4. Based on the real-time online extraction of photovoltaic module ripple impedance Z i and its phase φ i The average ripple impedance Z of the photovoltaic module reference ref and its phase mean φ ref The photovoltaic modules are compared separately, and the faults are diagnosed online based on the differences in the comparisons.

2. The online fault diagnosis method for photovoltaic modules according to claim 1, characterized in that, The photovoltaic module includes a single photovoltaic module or a photovoltaic module string formed by connecting multiple photovoltaic modules in series. When multiple photovoltaic module strings are connected in parallel, online fault diagnosis is performed on each photovoltaic module string separately. And / or the switching frequency range of the DC-DC switching converter is 10-50kHz, and / or the conditioning circuit module adopts a passive bandpass filter with a passband range of 1kHz-100kHz, and / or the sampling frequency of the data acquisition module is greater than 200kHz.

3. The online fault diagnosis method for photovoltaic modules according to claim 1 or 2, characterized in that, In step S2, the conditioning circuit module is used to filter out the PWM ripple current signal I. pv The DC component, low-frequency variation component, and high-frequency noise in the data.

4. The online fault diagnosis method for photovoltaic modules according to claim 1, characterized in that, In step S3, the ripple impedance Z of the photovoltaic module i and its phase φ i The calculation process includes the following: Estimate the frequency domain value V of the photovoltaic module terminal ripple voltage using the following formula. pv (jω): ； Among them, V dc (jω) represents the frequency domain of the ripple voltage at the load terminal, jω is the imaginary part of the complex angular frequency in frequency domain analysis, where ω = 2πf, ω is the angular frequency, f is the frequency, L is the inductance value in the Boost circuit, and i pv (jω) represents the frequency domain value of the ripple current at the photovoltaic module terminal, which is the PWM ripple component current signal i. pv-pwm C in This refers to the input-side filter capacitor value between the photovoltaic module and the DC-DC switching converter. Then, the ripple impedance Z of the photovoltaic module is calculated and extracted according to the following formula. i and its phase φ i : ； Among them, Z pv_pwm This is the ripple impedance Z of the photovoltaic module. i v pv_pwm This is the frequency domain value V of the ripple voltage at the photovoltaic module terminals. pv (jω), i pv_pwm This refers to the frequency domain value i of the ripple current at the photovoltaic module terminals. pv (jω), φ pv_pwm That is, the ripple impedance phase φ of the photovoltaic module. i .

5. The online fault diagnosis method for photovoltaic modules according to claim 4, characterized in that, In step S3, the frequency domain ripple voltage V at the load terminal is estimated according to the following formula. dc (jω): ； Among them, V dc (s) is the frequency domain of the ripple voltage at the load end, V. dc (jω);V D φ is the DC voltage input to the load, D is the duty cycle of the DC-DC switching converter in the Boost circuit, and φ is the input DC voltage to the load. Vdc This represents the phase of the ripple voltage at the load end.

6. The online fault diagnosis method for photovoltaic modules according to claim 1, characterized in that, In step S4, the average reference ripple impedance Z of the photovoltaic module ref The impedance value of a known reference healthy photovoltaic module can be used directly, or the following calculation process can be used: ； Among them, Z ref Z is the average amplitude of the reference ripple impedance of the photovoltaic module under normal conditions. ref Z k The ripple impedance Z of the photovoltaic module is extracted online in real time when the k-th photovoltaic module is in normal condition. i N is equal to or greater than 1, which is the number of photovoltaic modules in normal state participating in the calculation; The photovoltaic module reference ripple impedance phase mean φ ref The impedance phase value of a known reference healthy photovoltaic module can be used directly, or the following calculation process can be used: ； Where, φ ref The mean amplitude of the reference ripple impedance phase of the photovoltaic module under normal conditions is φ. ref φ k The phase φ of the ripple impedance of the k-th photovoltaic module is extracted online in real time when the module is in normal condition. i .

7. The online fault diagnosis method for photovoltaic modules according to claim 4, characterized in that, The relative change in impedance amplitude Δ|Zi| and the relative change in impedance phase Δφi of each photovoltaic module are calculated using the following formulas respectively; ； Furthermore, the standard deviation of the impedance magnitude σ(|Zi|) is calculated according to the following process: First, calculate the arithmetic mean of the impedance amplitude of each photovoltaic module within the sampling window. ; ； Then, the standard deviation of the impedance amplitude σ(|Zi|) is calculated based on this arithmetic mean: ； Where j is the total number of discrete data points included in the sampling window, i.e., the sample size, and m is the discrete time index of the data points in the sampling window, m=1,2,3,…,j, Z m This is the calculated value of the photovoltaic module ripple impedance corresponding to the m-th discrete sampling point within the sampling window.

8. The online fault diagnosis method for photovoltaic modules according to claim 1, characterized in that, In step S4, the online diagnosis of photovoltaic module faults based on the comparison difference includes the following process: The presence of local anomalies is determined by comparing the differences in data. If a local anomaly is detected, it is then classified according to the fault's characteristic thresholds. When the amplitude standard deviation σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc Furthermore, if the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th, it is determined that there is no fault. When the impedance amplitude standard deviation σ(|Zi|) is found to exceed its preset standard deviation threshold σ arc When this occurs, it is determined to be an arc fault; When the standard deviation of the impedance amplitude σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc However, if the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th and the relative change in impedance phase Δφi is not greater than 0, it is judged as an obstruction fault. When the standard deviation of the impedance amplitude σ(|Zi|) is found to be less than its preset standard deviation threshold σ arc However, if the relative change in impedance amplitude Δ|Zi| exceeds its preset change threshold th and the relative change in impedance phase Δφi is greater than 0, it is judged as an aging fault. The preset thresholds are adjusted based on system conditions and system parameters.

9. An online fault diagnosis system for photovoltaic modules, characterized in that, A Boost circuit is provided between the photovoltaic module and the load, and the Boost circuit includes a DC-DC switching converter connected to the photovoltaic module; the photovoltaic module online fault diagnosis system adopts the photovoltaic module online fault diagnosis method as described in any one of claims 1-8, including: The conditioning circuit module is connected between the current sensor and the data acquisition module; The data acquisition module is connected between the conditioning circuit module and the host computer; The host computer includes a ripple voltage estimation module, an impedance calculation module, and a fault diagnosis module. The ripple voltage estimation module is used to estimate the ripple voltage at the photovoltaic module terminals in real time. The impedance calculation module is used to extract the ripple impedance Z of the photovoltaic module in real time online. i and its phase φ i ; The fault diagnosis module is used to diagnose faults in photovoltaic modules online based on the comparison difference. The photovoltaic module includes a single photovoltaic module or a photovoltaic module string formed by connecting multiple photovoltaic modules in series.

10. The photovoltaic module online fault diagnosis system according to claim 9, characterized in that, A Hall current sensor is installed on the connection cable between the photovoltaic module and the DC-DC switching converter; a filter capacitor is connected to the input side of the photovoltaic module and the DC-DC switching converter, and the filter capacitor is connected in parallel with the DC-DC switching converter.

Images