CN113489459B - Photovoltaic power generation assembly fault detection and identification method based on digital twinning - Google Patents

Photovoltaic power generation assembly fault detection and identification method based on digital twinning Download PDF

Info

Publication number
CN113489459B
CN113489459B CN202110845731.6A CN202110845731A CN113489459B CN 113489459 B CN113489459 B CN 113489459B CN 202110845731 A CN202110845731 A CN 202110845731A CN 113489459 B CN113489459 B CN 113489459B
Authority
CN
China
Prior art keywords
fault
power generation
gamma
solar cell
photovoltaic power
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202110845731.6A
Other languages
Chinese (zh)
Other versions
CN113489459A (en
Inventor
周小杰
黄友锐
国海
徐善永
权悦
韩涛
张帝
胡福志
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Anhui University of Science and Technology
Original Assignee
Anhui University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Anhui University of Science and Technology filed Critical Anhui University of Science and Technology
Priority to CN202110845731.6A priority Critical patent/CN113489459B/en
Publication of CN113489459A publication Critical patent/CN113489459A/en
Application granted granted Critical
Publication of CN113489459B publication Critical patent/CN113489459B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02S—GENERATION OF ELECTRIC POWER BY CONVERSION OF INFRARED RADIATION, VISIBLE LIGHT OR ULTRAVIOLET LIGHT, e.g. USING PHOTOVOLTAIC [PV] MODULES
    • H02S50/00—Monitoring or testing of PV systems, e.g. load balancing or fault identification
    • H02S50/10—Testing of PV devices, e.g. of PV modules or single PV cells
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
    • Y02E10/00—Energy generation through renewable energy sources
    • Y02E10/50—Photovoltaic [PV] energy

Landscapes

  • Photovoltaic Devices (AREA)

Abstract

The invention discloses a photovoltaic power generation assembly fault detection and identification method based on digital twinning, which comprises the following steps: detecting and outputting characteristic quantity y (t) in a physical entity of the photovoltaic power generation component to be detected; the photovoltaic power generation assembly comprises a solar cell assembly and a DC-DC converter; constructing a digital twin body with the same physical entity structure as the photovoltaic power generation component to be detected, and calculating and outputting the measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body; calculating and outputting a residual vector gamma (t) according to the characteristic quantity y (t) and the measured characteristic quantity z (t); outputting a detection result according to the residual vector gamma (t); when the detection result has a fault, according to the residual error vector gamma (t) and the fault characteristic value fiCalculating and outputting L2Inner product; characteristic value f of faultiCalculating the luminance by the residual vector gamma (t) and the 2-norm of the residual vector gamma (t) | | gamma (t) |2Calculating to obtain; according to L2Inner product, output fault type. The method can detect whether the photovoltaic power generation assembly generates faults or not, and identify the type of the generated faults. High reliability and strong practicability.

Description

Photovoltaic power generation assembly fault detection and identification method based on digital twinning
Technical Field
The invention relates to the technical field of solar photovoltaic power generation, in particular to a digital twin-based photovoltaic power generation assembly fault detection and identification method.
Background
As environmental issues are receiving increased attention, the energy market is increasingly interested in photovoltaic power generation systems. A DC modular pv grid-connected power generation system is used by more and more people, as shown in fig. 1, in which a separate DC-DC converter is connected to each solar module, and then the output terminals of the DC-DC converters are connected in series to the input terminal of a DC-AC converter, through which energy is input into an AC grid. In the direct current modular photovoltaic grid-connected power generation system, the photovoltaic cell panel and the DC-DC converter adopted at the front stage are more, and the direct current modular photovoltaic grid-connected power generation system is difficult to identify and diagnose if a fault occurs.
Therefore, on the basis of the existing direct current modular photovoltaic grid-connected power generation system, how to provide a method for detecting and identifying faults of a photovoltaic power generation assembly based on digital twins so as to timely, efficiently and quickly diagnose and identify the faults generated by the photovoltaic power generation assembly becomes a problem to be solved by technical personnel in the field.
Disclosure of Invention
In view of the above problems, the present invention provides a method for detecting and identifying faults of a photovoltaic power generation assembly based on digital twinning, which can detect faults generated by the photovoltaic power generation assembly in time and accurately identify the types of the faults, and at least solves some technical problems.
The embodiment of the invention provides a photovoltaic power generation assembly fault detection and identification method based on digital twins, which comprises the following steps:
s1, detecting and outputting the characteristic quantity y (t) in the physical entity of the photovoltaic power generation assembly to be detected; the photovoltaic power generation assembly comprises a solar cell assembly and a DC-DC converter; the characteristic quantity y (t) is a characteristic output vector estimated value of the inductance current of the DC-DC converter, the capacitance voltage of the DC-DC converter, the maximum power point current of the solar cell module and the maximum power point voltage of the solar cell module in the physical entity;
s2, constructing a digital twin body with the same physical entity structure as the photovoltaic power generation component to be detected, and calculating and outputting the measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body; the measurement characteristic quantity z (t) is a characteristic output vector estimation value of DC-DC converter inductive current, DC-DC converter capacitance voltage, solar cell assembly maximum power point current and solar cell assembly maximum power point voltage in the digital twin body;
s3, calculating and outputting a residual vector gamma (t) according to the characteristic quantity y (t) and the measurement characteristic quantity z (t); outputting a detection result according to the residual error vector gamma (t);
s4, when the detection result has fault, the residual error vector gamma (t) calculated according to the step S3 and the fault characteristic value fiCalculating and outputting L2Inner product; the fault characteristic value fiCounting cells by the residual vector γ (t) and the 2-norm of the residual vector γ (t) | | γ (t) | Y2Calculating to obtain;
s5, according to the L2Inner product, output fault type.
Further, in the step S2, the calculation formula of the measured characteristic quantity z (t) of the photovoltaic power generation assembly in the digital twin body is:
Figure BDA0003180705740000021
Figure BDA0003180705740000022
k is an integer
Wherein k represents the current sampling period; g is the illumination intensity of the solar cell module in the digital twin body; t is the ambient temperature to which the solar cell module in the digital twin body is subjected; x [ k ]]A forward euler discretization equation representing a linear switch state space equation; I.C. Apv G,TThe estimated value of the output current of the solar cell component in the digital twin body under the condition of the maximum power is obtained; vpv G,TThe estimated value of the output voltage of the solar cell component in the digital twin body under the condition of the maximum power is obtained; e is an identity matrix of the electrical sensor gain; i is an identity matrix; snIs the identity matrix dimension; x [ k-1 ]]Is a state vector; u [ k-1 ]]Is an input vector;
Figure BDA0003180705740000031
Tsis a sampling period; a. theσ(t)And Bσ(t)Represents two states, Aσ(t)∈{A1,A2,K,Anm},Bσ(t)∈{B1,B2,K,Bnm}；
Figure BDA0003180705740000032
The illumination intensity in the digital twin is Gref＝1000W/m2Temperature of TrefUnder the condition of 25 ℃, when the maximum power is reached, the output current of the solar cell module is obtained;
Figure BDA0003180705740000033
the illumination intensity in the digital twin is GrefTemperature of TrefUnder the condition, when the maximum power is reached, the output voltage of the solar cell module is obtained; t is a unit ofref＝25℃；Gref＝1000W/m2；KiIs the current temperature coefficient; kpIs the power temperature coefficient.
Further, the step S3 includes:
s301, subtracting the characteristic quantity y (t) in the physical entity of the photovoltaic power generation component to be detected from the measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body to generate a residual vector gamma (t):
Figure BDA0003180705740000034
wherein the content of the first and second substances,
Figure BDA0003180705740000035
the current is the inductive current of a DC-DC converter in the digital twin body; i.e. iL(t) is the current through the inductor L in the physical entity;
Figure BDA0003180705740000036
the voltage is the capacitance voltage of a DC-DC converter in the digital twin body; v. ofc(t) is the output voltage of the DC-DC converter in the physical entity; i ispv G,TThe estimated value of the output current of the solar cell component in the digital twin body under the condition of the maximum power is obtained; vpv G,TThe estimated value of the output voltage of the solar cell component in the digital twin body under the condition of the maximum power is obtained; i.e. ipv(t) is the output current of the solar cell module in the physical entity; v. ofpv(t) is the output voltage of the solar cell module in the physical entity; gamma ray1(t)、γ2(t)、γ3(t)、γ4(t) forming a residual vector γ (t);
s302, calculating infinity norm gamma (t) | sweet wind according to the generated residual vector gamma (t)∞：
Figure BDA0003180705740000041
Wherein i represents the number of elements included in the residual vector γ (t);
s303, calculating ∞ -norm | | | gamma (t) | sweet wind∞And calculating a fault detection mark:
Figure BDA0003180705740000042
wherein Γ is a threshold for fault detection;
and S304, outputting a fault detection result according to the fault detection mark.
Further, in the step S4, L is calculated2The inner product is calculated by the formula:
Figure BDA0003180705740000043
wherein, W represents the window size of the inner product; t is time; gamma (t) is a residual vector; f. ofiIs a fault characteristic value;
Figure BDA0003180705740000044
represents L2Inner product; (ii) a Gamma rayT(t- τ) denotes γ (t) shifted in time by τ, which is an integral variable, and then transposed.
Further, in the step S4, the fault characteristic value fiCounting cells by residual vector gamma (t) and 2-norm of residual vector gamma (t) | | gamma (t) | non-calculation2Is calculated to obtain, calculateThe formula is as follows:
Figure BDA0003180705740000045
wherein γ (t) is a residual vector; | gamma (t) | non-conducting phosphor2Is a 2-norm; 2-norm | | gamma (t) | non-woven hair2The calculation formula of (c) is:
Figure BDA0003180705740000046
wherein i represents the number of elements included in the residual vector γ (t);
further, in the step S5, according to L2Inner product, output fault type, including:
fault identification is performed by calculating the following formula:
Figure BDA0003180705740000051
wherein, I represents a fault type; Λ is a threshold value of fault identification; gamma (t) is a residual vector; f. ofiIs the fault signature value, i ∈ (1, … 16).
The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:
the embodiment of the invention provides a photovoltaic power generation assembly fault detection and identification method based on digital twins, which comprises the following steps: detecting and outputting characteristic quantity y (t) in a physical entity of the photovoltaic power generation component to be detected; the photovoltaic power generation assembly comprises a solar cell assembly and a DC-DC converter; constructing a digital twin body with the same physical entity structure as that of the photovoltaic power generation component to be detected, and calculating and outputting a measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body; calculating and outputting a residual vector gamma (t) according to the characteristic quantity y (t) and the measured characteristic quantity z (t); outputting a detection result according to the residual vector gamma (t); when the detection result has a fault, according to the residual error vector gamma (t) and the fault characteristic value fiCalculating and outputting L2Inner product; characteristic value f of faultiCounting cells by residual vector gamma (t) and 2-norm of residual vector gamma (t) | | gamma (t) | non-calculation2Calculating to obtain; according toL2Inner product, output fault type. The method can detect whether the photovoltaic power generation assembly generates faults or not and identify the type of the generated faults. High flexibility, high efficiency and strong reliability.
Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a block diagram of a DC modular grid-connected PV power generation system of the prior art;
FIG. 2 is a flow chart of a method for detecting and identifying faults of a photovoltaic power generation assembly based on digital twinning according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a DC-DC converter provided by an embodiment of the present invention;
fig. 4 is a schematic diagram of a digital twin-based photovoltaic power generation assembly fault detection and identification method provided by an embodiment of the invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the invention provides a photovoltaic power generation assembly fault detection and identification method based on digital twins, which is shown in figure 2 and comprises the following steps:
s1, detecting and outputting the characteristic quantity y (t) in the physical entity of the photovoltaic power generation assembly to be detected; the photovoltaic power generation assembly comprises a solar cell assembly and a DC-DC converter; the characteristic quantity y (t) is a characteristic output vector estimated value of the inductance current of the DC-DC converter, the capacitance voltage of the DC-DC converter, the maximum power point current of the solar cell module and the maximum power point voltage of the solar cell module in the physical entity;
s2, constructing a digital twin body with the same physical entity structure as the photovoltaic power generation component to be detected, and calculating and outputting a measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body; measuring characteristic quantity z (t) which is a characteristic output vector estimated value of DC-DC converter inductive current, DC-DC converter capacitance voltage, solar cell assembly maximum power point current and solar cell assembly maximum power point voltage in the digital twin body;
s3, calculating and outputting a residual vector gamma (t) according to the characteristic quantity y (t) and the measured characteristic quantity z (t); outputting a detection result according to the residual vector gamma (t);
s4, when the detection result has fault, the residual error vector gamma (t) calculated according to the step S3 and the fault characteristic value fiCalculating and outputting L2Inner product; characteristic value f of faultiCounting cells by residual vector gamma (t) and 2-norm of residual vector gamma (t) | | gamma (t) | non-calculation2Calculating to obtain;
s5, according to L2Inner product, output fault type.
The method for detecting and identifying the faults of the photovoltaic power generation assembly based on the digital twin can detect whether the photovoltaic power generation assembly generates the faults or not and identify the types of the generated faults. Three types of fault types can be detected: whether the solar cell module has faults or not, whether the DC-DC converter has faults or not and whether the electric sensor has faults or not are distinguished and identified. The probability of these faults occurring in the photovoltaic power generation assembly and its constituent network depends on the operating conditions, system specifications and architecture. The method for detecting and identifying the faults of the photovoltaic power generation assembly is high in flexibility, efficiency and reliability.
The above steps are described in detail below:
specifically, referring to fig. 3, in the physical entity of the photovoltaic power generation module to be detected provided in step S1 of this embodiment, the DC-DC converter is a 4-switch buck-boost converter. The converter can operate in three modes: the buck-boost converter has the advantages of buck, boost and direct-through modes, and is higher in flexibility, efficiency and reliability than a buck-boost converter with only 2 switches. G, T respectively represents the illumination intensity and the ambient temperature of the solar cell module; v. ofpv(t) and ipv(t) respectively representing the output voltage and the output current of the solar cell module; cinAnd CoutThe capacitance of the input side and the capacitance of the output side of the DC-DC converter are respectively; l is filter inductance, R is equivalent internal resistance of filter inductance, iL(t) is the current through the inductor L; v. ofc(t) and io(t) output voltage and output current of the DC-DC converter, respectively; s1、S2、S3、S4There are 4 power electronic switches. And detecting the characteristic quantity y (t) in the physical entity of the photovoltaic power generation component to be detected by the electric sensor.
Specifically, step S2 is to construct a digital twin body having the same physical entity structure as the photovoltaic power generation component to be detected, and calculate and output a measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body, where the measurement characteristic quantity z (t) includes:
and (4) constructing a digital twin body with the same physical entity structure as the photovoltaic power generation component to be detected as shown in the figure 4. Wherein the digital twin is a digital simulation of a physical system. Analyzing, calculating and outputting the measurement characteristic quantity z (t) of the photovoltaic power generation assembly in the digital twin in real time:
Figure BDA0003180705740000081
in formula (1), x [ k ]]Is an estimated value of a DC-DC converter state vector; i ispv G,TIs an output current estimated value (which is a function of the illumination intensity G and the panel temperature T) of the solar cell component in the digital twin body under the condition of the maximum power)；Vpv G,TThe method comprises the following steps of obtaining an estimated value of output voltage (which is also a function of illumination intensity G and panel temperature T) of a solar cell module in a digital twin under the condition of maximum power; e is the identity matrix(s) of the gain of the electrical sensornRepresenting the identity matrix dimension).
In particular, the first element x [ k ] in z (t)]A forward Euler discretization equation for expressing a linear switch state space equation, which is applicable to any power converter with piecewise linear elements and a sampling period Ts。t＝kTsK is an integer, and k represents the current sampling period; x (kT)s)≡x[k]Is a state vector; u (kT)s)≡u[k]Is an input vector; continuous-time switching function σ (t), discretizing the continuous system, mapping time to an index set {1,2, K, nm }. Each index represents a specific model of the two-state given system (A)j,Bj),j∈{1,2,K,nm}。Aσ(t)∈{A1,A2,K,Anm},Bσ(t)∈{B1,B2,K,Bnm}, thus, in
Figure BDA0003180705740000082
In (1),
Figure BDA0003180705740000083
i is a unit matrix, and the unit matrix is,
Figure BDA0003180705740000084
in order to be a matrix of the system,
Figure BDA0003180705740000085
is a control matrix.
Further, a second element I in z (t)pv G,TThe output current estimation value of the solar cell module in the digital twin body under the condition of the maximum power is obtained, the illumination intensity is G, and the temperature of the cell plate is T:
Figure BDA0003180705740000091
in the formula (2), the first and second groups of the compound,
Figure BDA0003180705740000092
the illumination intensity in the digital twin is Gref＝1000W/m2Temperature of TrefUnder the condition of 25 ℃, when the maximum power is reached, the output current of the solar cell module is obtained; k isiIs the temperature coefficient of current.
Further, the last element V in z (t)pv G,TThe output voltage estimation value of the solar cell module in the digital twin body under the condition of the maximum power is obtained, the illumination intensity is G, and the temperature of the cell plate is T:
Figure BDA0003180705740000093
in the formula (3), the first and second groups,
Figure BDA0003180705740000094
at an illumination intensity of Gref＝1000W/m2Temperature of TrefThe maximum output power of the solar cell module under the condition of 25 ℃;
Figure BDA0003180705740000095
the illumination intensity in the digital twin is Gref＝1000W/m2Temperature of TrefUnder the condition of 25 ℃, when the maximum power is reached, the output voltage of the solar cell module is obtained; k ispIs the power temperature coefficient; k isiIs the current temperature coefficient; i ispv G,TThe estimated value of the output current of the solar cell component in the digital twin body under the condition of the maximum power is obtained; vpv G,TThe estimated value of the output voltage of the solar cell component in the digital twin body under the condition of the maximum power is obtained.
Further, the formulas (2) and (3) estimate the output current and voltage of the solar cell module at the maximum power point (i.e., the current and voltage output within the digital twin). These estimated values (i.e., the measured characteristic quantity z (T) of the photovoltaic power generation module in the digital twin) are compared with the actually measured values of the solar cell module under the actual G and T conditions (i.e., the characteristic quantity y (T) in the physical entity of the photovoltaic power generation module to be detected), and the estimated residual error (i.e., the residual vector γ (T)) of the digital twin is calculated.
Further, the output measurement characteristic quantity z (t) of the photovoltaic power generation assembly in the digital twin comprises inductance current of the DC-DC converter
Figure BDA0003180705740000101
Converter capacitor voltage
Figure BDA0003180705740000102
Maximum power point current I of solar cell modulepv G,TMaximum power point voltage V of solar cell modulepv G,T. In the non-fault case, the digital twin is the same output as the physical entity regardless of the input conditions. Therefore, the calculation formula of the measurement characteristic quantity z (t) of the photovoltaic power generation module in the digital twin at this time can further express the formula (1) as follows:
Figure BDA0003180705740000103
in the formula (4), x (kT)s)≡x[k]Is a state vector; u ((k-1) T)s)≡u[k-1]Is an input vector;
Figure BDA0003180705740000104
the value of the inductive current is the last sampling period;
Figure BDA0003180705740000105
the value of the voltage at two ends of the capacitor in the last sampling period is obtained; v. ofpv[k-1]The photovoltaic output voltage value is obtained in the last sampling period; i.e. io[k-1]Outputting a current value for the DC-DC converter in the last sampling period; r is the resistance value of the DC-DC converter in the digital twin; l is the inductance value of the DC-DC converter in the digital twin body; s1Is a switching function; s3Is a switching function; coutIs the output power of a DC-DC converter in a digital twin bodyCapacity value; t is a unit ofsIs a sampling period; i is2×2Is a 2-dimensional identity matrix;
Figure BDA0003180705740000106
and
Figure BDA0003180705740000107
is under standard test conditions (i.e. G)ref＝1000W/m2，TrefThe current and voltage values at maximum power of the solar cell module are output at 25 c, and these data are generally available in solar cell module data manuals.
Specifically, step S3 calculates and outputs residual vector γ (t) from feature y (t) and measured feature z (t); outputting a fault detection result according to the residual vector gamma (t), specifically comprising:
s301, subtracting the characteristic quantity y (t) in the physical entity of the photovoltaic power generation assembly to be detected from the measurement characteristic quantity z (t) of the photovoltaic power generation assembly in the digital twin, subtracting the steady-state output of the fault digital twin from the steady-state output of the normal digital twin, and generating a residual vector gamma (t):
Figure BDA0003180705740000111
in the formula (5), the first and second groups,
Figure BDA0003180705740000112
the current is the inductive current of a DC-DC converter in the digital twin body; i.e. iL(t) is the current in the physical entity through the inductor L;
Figure BDA0003180705740000113
the voltage is the capacitance voltage of a DC-DC converter in the digital twin body; v. ofc(t) is the output voltage of the DC-DC converter in the physical entity; i ispv G,TThe estimated value of the output current of the solar cell component in the digital twin body under the condition of the maximum power is obtained; vpv G,TThe estimated value of the output voltage of the solar cell component in the digital twin body under the condition of the maximum power is obtained; i.e. ipv(t) is physicalThe output current of the solar cell module in the entity; v. ofpv(t) is the output voltage of the solar cell module in the physical entity; gamma ray1(t)、γ2(t)、γ3(t)、γ4(t) constitutes a residual vector γ (t).
S302, calculating infinity norm gamma (t) | survival cell according to the generated residual vector gamma (t)∞：
Figure BDA0003180705740000114
Wherein i represents the number of elements included in the residual vector γ (t); . Infinity norm | gamma (t) | ventilation air grid∞Easy to calculate and can be used to detect failure events. In the fault-free case, the infinity norm is close to zero.
S303, calculating infinity norm gamma (t) | ventilation∞Calculating a fault detection flag indicating a fault according to the following logic:
Figure BDA0003180705740000121
in equation (6), Γ is a threshold for fault detection. Bias is generated due to analyzing the estimated residual, the discretized residual, and system noise (e.g., sensor). In order to cancel these variations and minimize the false alarm rate in the fault detection, it is necessary to select the threshold value Γ according to the actual situation.
And S304, outputting a detection result according to the fault detection mark. 1, in order to generate fault; and flag is 0, and no fault is generated.
Specifically, in step S4, when there is a fault in the detection result, the residual vector γ (t) calculated in step S3 and the fault feature value fiCalculating and outputting L2Inner product. The method comprises the following steps:
after a fault event, the residual vector γ (t) changes, indicating the fault type in the fault space in a unique direction. In order to determine the direction of the occurring fault in the fault space, the residual γ (t) is normalized to a unit vector f, let f be ξ (t) γ (t),
Figure BDA0003180705740000122
the unit vector f thus represents the unique direction of the fault event and is noted as the fault characteristic value fiWherein 2-norm | γ (t) | luminance2Indicating the magnitude of the fault (i.e. magnitude of fault)
Figure BDA0003180705740000123
) Thus, therefore, it is
Figure BDA0003180705740000124
Calculating a residual vector gamma (t) for each fault of interest, verifying the calculated gamma (t) through simulation and experiment, normalizing the gamma (t) and converting the normalized gamma (t) into a fault characteristic value fiAll fault characteristic values fiAnd forming a fault characteristic library.
Figure BDA0003180705740000125
Gamma (t) is a residual vector, | | gamma (t) | charging2Is a function of the 2-norm,
Figure BDA0003180705740000126
wherein i represents the number of elements included in the residual vector γ (t); .
Calculates and outputs L2Inner product:
Figure BDA0003180705740000127
in formula (7), W represents the window size for calculating the inner product; t is time; gamma (t) is a residual vector; f. ofiIs a fault characteristic value;
Figure BDA0003180705740000128
represents L2Inner product; gamma rayT(t- τ) denotes γ (t) shifted in time by τ and then transposed, τ being the integration variable. After a fault event occurs, the residual vector γ (t) represents the fault type in a unique direction in the fault space. Therefore, through fault detection, the residual vector γ (t) and all fault eigenvalues fi in the fault characteristic library are subjected to inner product calculation in real time. It represents the residual vector γ (t) at each fault eigenvalue fiAnd (6) projecting. The larger the inner product is, the closer the residual vector γ (t) is to the fault feature value fi, and the fault type can be determined by searching the largest inner product. Therefore, the fault identification method is to calculate the inner product and perform the maximum inner product search.
Specifically, step S5, according to L described above2Inner product, output fault type, including:
fault identification is performed by calculating the following equation:
Figure BDA0003180705740000131
in the formula (8), I represents the fault type, Λ is the threshold value of fault identification, γ (t) is the residual vector, fiFor the fault signature value, i ∈ (1, … 16).
The following lists the establishment of a fault feature library (including all fault feature values fi) by three specific examples, and develops a detailed description:
referring to fig. 4, according to the formula (1), the measurement characteristic quantity z (t) of the photovoltaic power generation assembly in the faultless digital twin body in the steady state is obtained by the following formula:
Figure BDA0003180705740000132
in formula (9), D3、D1Are respectively a switching device S3、S1Duty cycle of (d); i.e. io(t) is the output current of the DC-DC converter; r is the resistance in the digital twin shown in fig. 4; v. ofpv(t) is the output voltage of the solar cell module;
Figure BDA0003180705740000133
the illumination intensity in the digital twin is Gref＝1000W/m2Temperature of TrefUnder the condition of 25 ℃, when the maximum power is reached, the output current of the solar cell module is obtained;
Figure BDA0003180705740000134
in a digital twinningIllumination intensity of GrefTemperature of TrefUnder the condition, when the maximum power is reached, the output voltage of the solar cell module is obtained; gref＝1000W/m2；Tref＝25℃；KiIs the current temperature coefficient; kpIs the power temperature coefficient; g is the illumination intensity of the solar cell module; t is the temperature of the solar cell module cell panel; e is an identity matrix of the electrical sensor gain.
Example 1: fault feature calculation when the fault type of the solar cell module is i-1
The failure is due to the solar cell module being shielded (due to dust, etc.). It causes faulty panels
Figure BDA0003180705740000141
(light intensity is G)ref＝1000W/m2Temperature of TrefOutput current of solar cell module when maximum power is reached under 25 deg.c) is in Gref(Gref＝1000W/m2) Under the condition of
Figure BDA0003180705740000142
(at this time)
Figure BDA0003180705740000143
The change in (c) is negligible). Therefore, the characteristic quantity y (t) in the physical entity of the photovoltaic power generation component to be detected in a steady state is obtained as follows:
Figure BDA0003180705740000144
in the formula (10), io(t) is the output current of the DC-DC converter; d3、D1Are respectively a switching device S3、S1Duty cycle of (d); r is the resistance in the physical entity shown in fig. 4;
Figure BDA0003180705740000145
the illumination intensity in the digital twin is Gref＝1000W/m2Temperature controlDegree of TrefUnder the condition of 25 ℃, when the maximum power is reached, the output current of the solar cell module is obtained;
Figure BDA0003180705740000146
because the solar cell module is shielded, the solar cell module is tested under the standard test condition (the illumination intensity is G)ref＝1000W/m2Temperature of TrefThe variation of the output current when the solar cell module reaches the maximum power is 25 ℃;
Figure BDA0003180705740000147
the illumination intensity in the digital twin is GrefTemperature of TrefUnder the condition, when the maximum power is reached, the output voltage of the solar cell module is obtained; k isiIs the current temperature coefficient; k ispIs the power temperature coefficient; v. ofpv(t) is the output voltage of the solar cell module; gref＝1000W/m2；Tref＝25℃。
Specifically, the resistance value, the inductance value, and the capacitance value in the digital twin and the physical entity are the same.
In this case, the residual vector γ (t) is obtained by subtracting the equation (10) from the equation (9) by the residual vector γ (t) z (t) -y (t)
Figure BDA0003180705740000151
Divide it by | γ (t) | calculation2The normalized value f can be obtained and represents the fault amplitude
Figure BDA0003180705740000152
Obtaining a fault characteristic value f after normalization1＝[0,0,1,0]T. Similarly, the characteristic values of other failure types of the solar cell module listed in the following table can be obtained by subtracting the equation (10) from the equation (9).
TABLE 1 solar cell Module failure
Figure BDA0003180705740000153
Example 2: fault characteristic calculation when fault type i of DC-DC converter is 4
Referring to fig. 3, such as if the switching device S1An open circuit fault occurs, and this fault event will result in an open circuit in the output of the solar module. In this case, the residual vector γ (t) is z (t) -y (t), that is, the value obtained by subtracting the equation (10) from the equation (9), and y (t) is E [0,0,0, V ═ y (t)oc]T. Generating residual gamma (t), analyzing to obtain fault characteristic value f4And
Figure BDA0003180705740000161
similarly, the characteristic values of other fault types of the DC-DC converter listed in the following table can be obtained by calculation.
TABLE 2 DC-DC CONVERTER FAULT
Figure BDA0003180705740000162
Figure BDA0003180705740000171
In table IscShort-circuit current for the solar cell module; vocIs the open circuit voltage of the solar cell module; d1Being switching devices S1Duty cycle of (d); d3Being switching devices S3Of the duty cycle of (c).
Example 3: characteristic value calculation when the electrical sensor fault type is i-12
For measuring the inductor current iL(t) sensor drift amount Δ c considering gain12(t) drift and deviation of measured value from accurate value Δ e12(t) amount of deviation. In a measured physical system, the failure of such a sensor can be modeled as:
Figure BDA0003180705740000172
in formula (11), E is an electric sensorAn identity matrix of gains; Δ C (t) is the drift amount of the gain; i.e. iL(t) is the current flowing through the inductance L in the physical entity; v. ofc(t) is the DC-DC converter output voltage in the physical entity; i.e. ipv(t) is the output current of the solar cell module in the physical entity, vpv(t) is the output voltage of the solar cell module in the physical entity; Δ e (t) is the offset.
In this case, the residual γ (t) ═ z (t) -y (t), i.e., γ (t) ═ Δ c (t)12(t)iL(t)+Δe12(t),0,0,0]T. Therefore, the failure feature value f12＝[1,0,0,0]T，
Figure BDA0003180705740000173
Please note that the measured characteristic quantity of the photovoltaic power generation modules in the digital twin
Figure BDA0003180705740000174
And the characteristic quantity y (t) E [ i ] in the physical entity of the photovoltaic power generation component to be detectedL(t),vC(t),ipv(t),vpv(t)]The same is true in the absence of a fault. Similarly, the values characteristic of the other types of faults of the electrical sensor listed in the table below can be obtained by calculation.
TABLE 3 Electrical sensor failure
Figure BDA0003180705740000181
Through the analysis, a fault characteristic library is established, and fault characteristic values f of 16 faults of the photovoltaic power generation assembly are contained in the fault characteristic libraryi(including 3 types of solar cell module faults, 8 types of DC-DC converter faults and 5 types of electric sensor faults).
According to the method for detecting and identifying the faults of the photovoltaic power generation assembly based on the digital twins, provided by the embodiment of the invention, the digital twins of the solar cell assembly and the DC-DC converter are established, the output value of the digital twins is compared with the output value of the physical entity to obtain a residual vector, and the fault detection and the fault identification are realized by analyzing, calculating and evaluating the residual vector.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.

Claims (6)

1. A photovoltaic power generation assembly fault detection and identification method based on digital twinning is characterized by comprising the following steps:
s1, detecting and outputting the characteristic quantity y (t) in the physical entity of the photovoltaic power generation assembly to be detected; the photovoltaic power generation assembly comprises a solar cell assembly and a DC-DC converter; the characteristic quantity y (t) is a characteristic output vector estimated value of the inductance current of the DC-DC converter, the capacitance voltage of the DC-DC converter, the maximum power point current of the solar cell module and the maximum power point voltage of the solar cell module in the physical entity;
s2, constructing a digital twin body with the same physical entity structure as the photovoltaic power generation component to be detected, and calculating and outputting the measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body; the measurement characteristic quantity z (t) is a characteristic output vector estimation value of DC-DC converter inductive current, DC-DC converter capacitance voltage, solar cell module maximum power point current and solar cell module maximum power point voltage in the digital twin body;
s3, calculating and outputting a residual vector gamma (t) according to the characteristic quantity y (t) and the measurement characteristic quantity z (t); outputting a detection result according to the residual error vector gamma (t);
s4, when the detection result has fault, the residual error vector gamma (t) calculated according to the step S3 and the fault characteristic value fiCalculating and outputting L2Inner product; the fault characteristic value fiCalculating light by the residual vector gamma (t) and the 2-norm | gamma (t) | of the residual vector gamma (t)2Calculating to obtain;
s5, according to the L2Inner product, output fault type.
2. The method for detecting and identifying faults of photovoltaic power generation assemblies based on digital twins as claimed in claim 1, wherein in step S2, the measurement characteristic quantity z (t) of the photovoltaic power generation assemblies in the digital twins is calculated by the formula:
Figure FDA0003180705730000021
Figure FDA0003180705730000022
k is an integer
Wherein k represents the current sampling period; g is the illumination intensity of the solar cell module in the digital twin body; t is the ambient temperature to which the solar cell module in the digital twin body is subjected; x [ k ]]A forward euler discretization equation representing a linear switch state space equation; i ispv G,TThe estimated value of the output current of the solar cell component in the digital twin body under the condition of the maximum power is obtained; vpv G,TThe output voltage estimation value of the solar cell module in the digital twin body under the condition of maximum power is obtained; e is an identity matrix of the electrical sensor gain; i is an identity matrix; snIs the dimension of the identity matrix; x [ k-1 ]]Is a state vector; u [ k-1 ]]Is an input vector;
Figure FDA0003180705730000023
Tsis a sampling period; a. theσ(t)And Bσ(t)Represents two states, Aσ(t)∈{A1,A2,K,Anm},Bσ(t)∈{B1,B2,K,Bnm}；
Figure FDA0003180705730000024
The illumination intensity in the digital twin is Gref＝1000W/m2Temperature of TrefUnder the condition of 25 ℃, when the maximum power is reached, the output current of the solar cell module is obtained;
Figure FDA0003180705730000025
the illumination intensity in the digital twin is GrefTemperature of TrefUnder the condition, when the maximum power is reached, the output voltage of the solar cell module is obtained; t is a unit ofref＝25℃；Gref＝1000W/m2；KiIs the current temperature coefficient; kpIs the power temperature coefficient.
3. The digital twin-based photovoltaic power generation assembly fault detection and identification method according to claim 1, wherein the step S3 includes:
s301, subtracting the characteristic quantity y (t) in the physical entity of the photovoltaic power generation component to be detected from the measurement characteristic quantity z (t) of the photovoltaic power generation component in the digital twin body to generate a residual vector gamma (t):
Figure FDA0003180705730000031
wherein the content of the first and second substances,
Figure FDA0003180705730000032
the current is the inductive current of a DC-DC converter in the digital twin body; i.e. iL(t) is the current in the physical entity through the inductor L;
Figure FDA0003180705730000033
the voltage is the capacitance voltage of a DC-DC converter in the digital twin body; v. ofc(t) is the output voltage of the DC-DC converter in the physical entity; I.C. Apv G,TThe estimated value of the output current of the solar cell component in the digital twin body under the condition of the maximum power is obtained; vpv G,TThe output voltage estimation value of the solar cell module in the digital twin body under the condition of maximum power is obtained; i.e. ipv(t) is the output current of the solar cell module in the physical entity; v. ofpv(t) is the output voltage of the solar cell module in the physical entity; gamma ray1(t)、γ2(t)、γ3(t)、γ4(t) forming a residual vector γ (t);
s302, calculating infinity norm gamma (t) | sweet wind according to the generated residual vector gamma (t)∞：
Figure FDA0003180705730000034
Wherein i represents the number of elements included in the residual vector γ (t);
s303, calculating ∞ -norm | | | gamma (t) | sweet wind∞And calculating a fault detection mark:
Figure FDA0003180705730000035
wherein Γ is a threshold for fault detection;
and S304, outputting a fault detection result according to the fault detection mark.
4. The method for detecting and identifying faults of a photovoltaic power generation assembly based on digital twinning as claimed in claim 1, wherein in step S4, L is calculated2The inner product is calculated by the formula:
Figure FDA0003180705730000036
wherein W represents the window size for computing the inner product; t is time; gamma (t) is a residual vector; f. ofiIs a fault characteristic value;
Figure FDA0003180705730000044
represents L2Inner product; gamma rayT(t- τ) denotes γ (t) shifted in time by τ, which is an integral variable, and then transposed.
5. The method for detecting and identifying faults of a digital twin-based photovoltaic power generation assembly as claimed in claim 1, wherein in the step S4, the fault characteristic value f isiFrom residual vectorsGamma (t) and 2-norm of residual vector gamma (t) | | gamma (t) | survival of the eyes2And calculating to obtain the following formula:
Figure FDA0003180705730000041
wherein γ (t) is a residual vector; | gamma (t) | non-conducting phosphor2Is a 2-norm; 2-norm | | gamma (t) | non-woven hair2The calculation formula of (2) is as follows:
Figure FDA0003180705730000042
where i represents the number of elements included in the residual vector γ (t).
6. The method for detecting and identifying faults of a photovoltaic power generation assembly based on digital twinning as claimed in claim 1, wherein in step S5, according to L2Inner product, output fault type, including: fault identification is performed by calculating the following equation:
Figure FDA0003180705730000043
wherein I represents a fault type; Λ is a threshold value of fault identification; gamma (t) is a residual vector; f. ofiIs the fault signature value, i ∈ (1, … 16).
CN202110845731.6A 2021-07-26 2021-07-26 Photovoltaic power generation assembly fault detection and identification method based on digital twinning Active CN113489459B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110845731.6A CN113489459B (en) 2021-07-26 2021-07-26 Photovoltaic power generation assembly fault detection and identification method based on digital twinning

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110845731.6A CN113489459B (en) 2021-07-26 2021-07-26 Photovoltaic power generation assembly fault detection and identification method based on digital twinning

Publications (2)

Publication Number Publication Date
CN113489459A CN113489459A (en) 2021-10-08
CN113489459B true CN113489459B (en) 2022-07-19

Family

ID=77943731

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110845731.6A Active CN113489459B (en) 2021-07-26 2021-07-26 Photovoltaic power generation assembly fault detection and identification method based on digital twinning

Country Status (1)

Country Link
CN (1) CN113489459B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9002678B1 (en) * 2014-01-10 2015-04-07 King Fahd University Of Petroleum And Minerals Unified approach to detection and isolation of parametric faults using a kalman filter residual-based approach
JP2017162009A (en) * 2016-03-07 2017-09-14 日本電信電話株式会社 Device, method, and program for generating feature quantity
CN110530638A (en) * 2019-07-31 2019-12-03 西安交通大学 Based on number twin aeroplane engine main bearing damage check and diagnostic method
CN112018758A (en) * 2020-08-26 2020-12-01 国网重庆市电力公司电力科学研究院 Modeling method of high-proportion new energy-containing alternating current-direct current hybrid system based on digital twinning

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2007070247A2 (en) * 2005-11-29 2007-06-21 Venture Gain, L.L.C. Residual-based monitoring of human health
US20170286572A1 (en) * 2016-03-31 2017-10-05 General Electric Company Digital twin of twinned physical system

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9002678B1 (en) * 2014-01-10 2015-04-07 King Fahd University Of Petroleum And Minerals Unified approach to detection and isolation of parametric faults using a kalman filter residual-based approach
JP2017162009A (en) * 2016-03-07 2017-09-14 日本電信電話株式会社 Device, method, and program for generating feature quantity
CN110530638A (en) * 2019-07-31 2019-12-03 西安交通大学 Based on number twin aeroplane engine main bearing damage check and diagnostic method
CN112018758A (en) * 2020-08-26 2020-12-01 国网重庆市电力公司电力科学研究院 Modeling method of high-proportion new energy-containing alternating current-direct current hybrid system based on digital twinning

Also Published As

Publication number Publication date
CN113489459A (en) 2021-10-08

Similar Documents

Publication Publication Date Title
Kang et al. A multi-fault diagnostic method based on an interleaved voltage measurement topology for series connected battery packs
Yi et al. Fault detection for photovoltaic systems based on multi-resolution signal decomposition and fuzzy inference systems
Jain et al. A digital twin approach for fault diagnosis in distributed photovoltaic systems
Hu et al. Online two-section PV array fault diagnosis with optimized voltage sensor locations
Bastidas-Rodríguez et al. Model-based degradation analysis of photovoltaic modules through series resistance estimation
Mohagheghi et al. Condition monitoring of power electronic circuits using artificial neural networks
Spataru et al. Detection of increased series losses in PV arrays using Fuzzy Inference Systems
Jalil et al. An enhanced Cassie–Mayr-based approach for DC series arc modeling in PV systems
Abd el-Ghany et al. A new monitoring technique for fault detection and classification in PV systems based on rate of change of voltage-current trajectory
Soliman et al. Capacitance estimation for dc-link capacitors in a back-to-back converter based on artificial neural network algorithm
Jain et al. Fault diagnosis via PV panel-integrated power electronics
Gao et al. Novel PV fault diagnoses via SAE and improved multi-grained cascade forest with string voltage and currents measures
CA3131717A1 (en) Systems and methods for monitoring energy-related data in an electrical system
Xi et al. Online fault diagnosis for photovoltaic arrays based on fisher discrimination dictionary learning for sparse representation
Manjunath et al. On-line health monitoring of PV plants
Patil et al. Arc fault detection in DC microgrid using deep neural network
Zakir et al. Design and implementation of a fault detection method for a PV‐fed DC‐microgrid with power control mechanism
Ojetola et al. Testing machine learned fault detection and classification on a dc microgrid
Fan et al. A novel method of quantitative internal short circuit diagnosis based on charging electric quantity in fixed voltage window
Ahmad et al. An online technique for condition monitoring of capacitor in PV system
CN113489459B (en) Photovoltaic power generation assembly fault detection and identification method based on digital twinning
Chandrasekharan et al. Current indicator based fault detection algorithm for identification of faulty string in solar PV system
Wu et al. Research on short-circuit fault-diagnosis strategy of lithium-ion battery in an energy-storage system based on voltage cosine similarity
CN114172401B (en) NPC three-level inverter multi-class fault diagnosis method based on reduced order observer
CN113746132B (en) Photovoltaic power station based on cloud edge cooperation and control method thereof

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant