Photovoltaic device temperature anomaly prediction method and device based on digital twin model

Abstract

The application relates to the field of photovoltaic equipment anomaly early warning, and provides a photovoltaic equipment temperature anomaly prediction method and device based on a digital twin model, the method comprises the following steps: acquiring environmental acoustic signals, meteorological data and measured temperature data of a photovoltaic equipment; performing acoustic spectrum feature extraction according to the environmental acoustic signals to obtain a virtual environmental parameter set; sending the meteorological data and the virtual environmental parameter set to a preset digital twin model for processing to obtain an initial temperature prediction value set; determining a temperature prediction value distribution map according to the initial temperature prediction value set and the measured temperature data of the photovoltaic equipment, comparing a temperature prediction result with an abnormal temperature threshold, and obtaining a temperature anomaly risk point; and processing the temperature prediction value distribution map, the temperature anomaly risk point and the meteorological data to obtain photovoltaic equipment temperature anomaly early warning information, so that the accuracy and foresight of photovoltaic equipment temperature anomaly prediction are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of photovoltaic device anomaly early warning, in particular to a photovoltaic device temperature anomaly prediction method and device based on a digital twin model. BACKGROUND

[0002] In the field of photovoltaic power station operation and maintenance, accurate monitoring and fault early warning of the device operating state are the key to ensuring power generation efficiency and safety. Among them, the abnormal rise of the temperature of the photovoltaic module is the main cause of the induction of hot spot effect, power attenuation and even fire risk, so accurate prediction and abnormal early warning of the module temperature are particularly important. At present, the mainstream method in this field mainly relies on the deployment of temperature sensors for direct measurement, or estimates the temperature based on limited meteorological data through a simplified thermodynamic model. However, the existing technical solutions generally have inherent defects such as insufficient perception of the module surface micro-environment, overly single and idealized model input parameters, and inability to quantify prediction uncertainty, resulting in lagging early warning results, high false alarm rate or missed alarm risk, which makes it difficult to meet the operation and maintenance needs of modern large-scale photovoltaic power stations for high-precision, forward-looking prediction and intelligent diagnosis of temperature anomalies. SUMMARY

[0003] The purpose of the present application is to provide a photovoltaic device temperature anomaly prediction method and device based on a digital twin model to improve the above problems.

[0004] To achieve the above purpose, the embodiments of the present application provide the following technical solutions:

[0005] On the one hand, the embodiments of the present application provide a photovoltaic device temperature anomaly prediction method based on a digital twin model, which comprises:

[0006] acquiring environmental acoustic signals, meteorological data and measured temperature data of a photovoltaic device;

[0007] extracting acoustic spectrum features from the environmental acoustic signals to obtain a virtual environmental parameter set, the virtual environmental parameter set comprising a virtual wind speed distribution map and a virtual stain distribution map;

[0008] sending the meteorological data and the virtual environmental parameter set to a preset digital twin model for processing to obtain an initial temperature prediction value set;

[0009] determining a temperature prediction value distribution map according to the initial temperature prediction value set and the measured temperature data of the photovoltaic device, and comparing the temperature prediction result with an abnormal temperature threshold to obtain a temperature anomaly risk point;

[0010] processing the temperature prediction value distribution map, the temperature anomaly risk point and the meteorological data to obtain photovoltaic device temperature anomaly early warning information.

[0011] In a second aspect, the embodiments of the present application provide a photovoltaic device temperature anomaly prediction device based on a digital twin model, the device comprising:

[0012] An acquisition module is configured to acquire an environmental acoustic signal, meteorological data, and measured temperature data of a photovoltaic device.

[0013] A first processing module is configured to perform acoustic spectrum feature extraction based on the environmental acoustic signal to obtain a virtual environmental parameter set, the virtual environmental parameter set comprising a virtual wind speed distribution map and a virtual stain distribution map.

[0014] A second processing module is configured to send the meteorological data and the virtual environmental parameter set to a preset digital twin model for processing to obtain an initial temperature prediction value set.

[0015] A third processing module is configured to determine a temperature prediction value distribution map based on the initial temperature prediction value set and the measured temperature data of the photovoltaic device, and compare the temperature prediction result with an abnormal temperature threshold to obtain a temperature anomaly risk point.

[0016] A fourth processing module is configured to process the temperature prediction value distribution map, the temperature anomaly risk point, and the meteorological data to obtain photovoltaic device temperature anomaly warning information.

[0017] In a third aspect, the embodiments of the present application provide a photovoltaic device temperature anomaly prediction device based on a digital twin model, the device comprising a memory and a processor. The memory is configured to store a computer program, and the processor is configured to execute the computer program to implement the steps of the photovoltaic device temperature anomaly prediction method based on a digital twin model.

[0018] In a fourth aspect, the embodiments of the present application provide a readable storage medium, the readable storage medium storing a computer program, the computer program being executed by a processor to implement the steps of the photovoltaic device temperature anomaly prediction method based on a digital twin model.

[0019] The present application has the following beneficial effects:

[0020] The present application obtains environmental acoustic signals, meteorological data and measured temperature data of a photovoltaic device, extracts acoustic spectrum features based on the environmental acoustic signals to obtain a virtual environmental parameter set containing a virtual wind speed distribution map and a virtual stain distribution map, inputs the meteorological data and the virtual environmental parameter set into a preset digital twin model to obtain an initial temperature prediction value set, determines a temperature prediction value distribution map by combining the initial temperature prediction value set and the measured temperature data, and compares the temperature prediction value distribution map with an abnormal temperature threshold to obtain a temperature abnormality risk point, and finally generates photovoltaic device temperature abnormality warning information based on the temperature prediction value distribution map, the temperature abnormality risk point and the meteorological data, effectively making up for the defects of the prior art, such as insufficient perception of the surface microenvironment of the photovoltaic device, single idealization of model input parameters and inability to quantitatively predict uncertainty, improving the accuracy and forward-looking nature of photovoltaic device temperature abnormality prediction, and reducing the lag, false alarm rate and missed alarm risk of the warning result.

[0021] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent from the description, or can be learned by practice of the present application according to the embodiments. The objects and other advantages of the present application can be achieved and obtained by the structure particularly pointed out in the written description and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0023] Figure 1 The flow chart of the photovoltaic device temperature abnormality prediction method of the digital twin model described in the embodiments of the present application.

[0024] Figure 2 The schematic diagram of the photovoltaic device temperature abnormality prediction device structure of the digital twin model described in the embodiments of the present application.

[0025] In the figure, the annotations are: 800, photovoltaic device temperature abnormality prediction device of digital twin model; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component. DETAILED DESCRIPTION

[0026] In order to make the purposes, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work are within the scope of protection of the present application.

[0027] It should be noted that similar reference numerals and letters represent similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", and the like are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0028] Embodiment 1

[0029] The present embodiment provides a photovoltaic device temperature anomaly prediction method of a digital twin model. It can be understood that in the present embodiment, a scene can be laid out, for example: in the operation and maintenance of a large photovoltaic power station, the scene of early warning of temperature anomalies of photovoltaic devices through acquisition of on-site environmental acoustic signals, meteorological data such as air temperature and illumination, and measured temperature of photovoltaic panels.

[0030] Referring to Figure 1 , the present method includes steps S1-S5.

[0031] Step S1, acquiring environmental acoustic signals, meteorological data, and measured temperature data of photovoltaic devices;

[0032] This step is different from the traditional single dependence on meteorological data or sparse temperature sensors, and introduces the information dimension of environmental acoustic signals. In the broad and complex environment of a photovoltaic power station, wind, rain, dust, and the vibration of the components themselves will all produce unique sound characteristics. Acquisition of these signals can capture the dynamic changes of the microclimate on the surface of the components, laying a data foundation for subsequent non-contact and large-scale inversion of key environmental parameters, thereby overcoming the bottleneck of traditional point measurement that is difficult to capture spatial heterogeneity.

[0033] Step S2, performing acoustic spectrum feature extraction according to the environmental acoustic signals to obtain a virtual environmental parameter set, the virtual environmental parameter set including a virtual wind speed distribution map and a virtual stain distribution map;

[0034] The step converts the acoustic phenomenon of the physical world into a virtual environment field in the digital space, utilizes the law of sound propagation in the turbulent boundary layer to invert the wind speed, and estimates the dirt thickness based on the attenuation characteristics of the interaction between the sound wave and the dust particles, thereby realizing real-time monitoring of the wind speed and dirt on the surface of the component and generating a high-resolution environmental map that is difficult to obtain by a traditional sensor network.

[0035] The step S2 further comprises steps S21-S24, which specifically include:

[0036] Step S21, performing enhancement processing on the environmental acoustic signal to obtain a feature vector of enhanced acoustic features;

[0037] An independent component analysis (ICA) algorithm is used to perform blind source separation processing on the environmental acoustic signal collected by the acoustic sensor network, and the environmental wind sound is effectively stripped from the mixed signal as a dominant independent sound source. Subsequently, the signal mainly containing particle impact and component vibration information after separation is subjected to wavelet packet transform, and the energy features of 2kHz-8kHz are extracted to obtain a feature vector of enhanced acoustic features. The step is aimed at the characteristics of strong wind noise in the photovoltaic array area and weak target acoustic features, and through signal separation guided by physical mechanism, the signal-to-noise ratio of the acoustic features related to the surface state of the component is significantly improved, thereby laying a clean signal foundation for subsequent accurate inversion.

[0038] Step S22, inverting the wind speed on the surface of the photovoltaic device according to the feature vector of the enhanced acoustic features to obtain a virtual wind speed distribution map;

[0039] The step S22 further comprises steps S221-S224, which specifically include:

[0040] Step S221, performing acoustic coherence feature extraction processing on the feature vector of the enhanced acoustic features to obtain a frequency band energy matrix, wherein the frequency band energy matrix comprises spatial coherence corresponding to a frequency band;

[0041] The step firstly selects a specific high-frequency band, i.e., 2kHz-8kHz, which is sensitive to the change of turbulence scale and less disturbed by environmental noise, based on the acoustic characteristics generated by the turbulent boundary layer on the surface of the photovoltaic panel. Then, the coherence of the signals of the microphones at different positions in the acoustic sensor network in the frequency band is calculated by using the cross-power spectral density analysis method. The coherence represents the linear correlation degree of the signals at different positions in a statistical sense, and the calculation is based on the ratio of the cross-power spectral density to the self-power spectral density obtained by Fourier transform. By analyzing the spatial coherence pattern of the sensor array signals in the characteristic frequency band, a frequency band energy matrix containing the spatial coherence information of each frequency band is constructed. By using the physical characteristic that the acoustic signals excited at different positions by the same turbulent structure passing through the surface of the photovoltaic panel have high coherence, the non-related environmental background noise is effectively suppressed, and the robust acoustic features highly correlated with the surface friction wind speed are extracted.

[0042] Step S222, forward modeling is performed according to the frequency band energy matrix to obtain a forward model;

[0043] The step is based on the improved Curle acoustic analogy theory to construct a physical model. The classical Curle acoustic analogy regards the surface of an object as a sound source and considers that sound is generated by the unsteady pressure fluctuations on the surface of the object. For this specific scenario of photovoltaic modules, the model simplifies the photovoltaic panel into a rigid flat plate of finite size and considers the turbulent fluctuating pressure of the boundary layer as the main sound source. The modeling process specifically includes: defining the geometric size and spatial orientation of the flat plate, setting the surface friction wind speed As a key input parameter, the surface friction wind speed related boundary layer pressure fluctuation statistical model is solved, and the theoretical sound pressure level distribution radiated by these pressure fluctuations at the sensor positions, especially in the 2kHz-8kHz band, is calculated by integration. The final forward model is a deterministic function, which establishes a physical mapping relationship from the fluid mechanics parameter of the surface friction wind speed to the predicted sound pressure level distribution.

[0044] Step S223, the surface friction wind speed is solved according to the forward model and the frequency band energy matrix to obtain a surface friction wind speed field;

[0045] The step takes the forward model as a forward operator, and assumes that the model error and the measurement error obey Gaussian distribution, thereby constructing a likelihood function of the inversion problem, which describes the probability of observing the current frequency band energy matrix under the condition of a given surface friction wind speed. Secondly, the spatial smoothness of the surface friction wind speed is introduced as a priori constraint, that is, the wind speed values of adjacent spatial points should not change dramatically. Then, the Markov Chain Monte Carlo sampling method is used to perform random walk in the parameter space to explore and sample the posterior probability distribution. Markov Chain Monte Carlo generates a Markov chain, and the stationary distribution of the Markov chain is the posterior distribution of the to-be-solved parameter. Finally, by analyzing a large number of sampling points, the posterior probability distribution of the surface friction wind speed field can be obtained, and the maximum value (maximum a posteriori estimation) of the posterior probability distribution is the most likely surface friction wind speed field.

[0046] Step S224, determining a virtual wind speed distribution map according to the surface friction wind speed field.

[0047] In this step, the basic fluid mechanics parameters obtained by inversion are converted into standard height wind speed which is more intuitive and directly usable in engineering. The specific conversion process is as follows:

[0048]

[0049] In the above formula, is the target wind speed to be solved, which represents the average wind speed at a distance of height from the component surface; represents the surface friction wind speed; represents the Karman constant, which is usually taken as 0.4; represents the vertical height of the target wind speed to be solved; represents the surface roughness length, which is used to quantify the aerodynamic roughness of the ground surface. According to the underlying surface conditions of the power station site, it is judged or selected by table lookup, such as 0.0001-0.001 for cement ground and 0.01-0.05 for low grassland. Through this formula, the surface friction wind speed at each position is converted into the average wind speed at a standard height from the component surface. Finally, the spatial interpolation algorithm is used to interpolate and extrapolate the wind speed estimates at these discrete sensor points to generate a continuous spatial two-dimensional virtual wind speed distribution map covering the entire photovoltaic array.

[0050] Step S23, predicting the thickness of the dirt on the surface of the photovoltaic device according to the virtual wind speed distribution map and the feature vector of the enhanced acoustic feature, to obtain a virtual dirt distribution map;

[0051] The step S23 further includes steps S231-S234, which specifically include:

[0052] Step S231, according to the virtual wind speed distribution and the feature vector of the enhanced acoustic feature, secondary separation processing of particle impact sound and wind speed background sound is performed to obtain a sound energy feature;

[0053] Since the wind speed is the main factor affecting the intensity of the environmental background noise, the virtual wind speed distribution is taken as the explanatory variable, and the feature vector of the enhanced acoustic feature is taken as the response variable, and a linear regression model between the wind speed and the broadband acoustic signal energy is established by a partial least squares regression algorithm. The algorithm maximizes the covariance between the explanatory variable and the response variable by finding a set of latent variables, thereby fitting the acoustic component most related to the wind speed. After subtracting the fitting value from the original acoustic signal, the residual signal obtained is the feature sound energy mainly generated by the dust particle impact and friction, which removes the wind speed related noise. This way can effectively strip the background interference and highlight the acoustic signal sensitive to stains in view of the dynamic change characteristics of the photovoltaic power station environmental noise with wind speed.

[0054] Step S232, according to the sound energy feature and the virtual wind speed distribution, a normalization processing is performed to obtain a wind speed normalized particle impact sound energy distribution;

[0055] The normalization function of this step is specifically:

[0056]

[0057] In the above formula, represents the wind speed normalized particle impact sound energy; represents the measured particle impact sound energy; K represents a proportional coefficient; U represents the wind speed value of the corresponding sensor position obtained from the virtual wind speed distribution. Through this calculation, the sound energy measured at different wind speeds is uniformly converted to the equivalent sound energy at the same standard wind speed.

[0058] Step S233, according to the wind speed normalized particle impact sound energy distribution, an inversion processing is performed to obtain a stain thickness information;

[0059] In this step, the dust layer on the surface of the photovoltaic device is modeled as a porous medium layer with a specific acoustic impedance. Then, the transfer matrix method of sound wave propagation in multilayer medium is applied to calculate the theoretical attenuation of the sound wave after penetrating the dust layer. The transfer matrix method represents each layer of medium and its boundary conditions by a matrix, and calculates the total propagation characteristics of the sound wave on the whole path by matrix multiplication. By matching the theoretical attenuation corresponding to the dust layer of different thickness with the normalized sound energy attenuation (which is obtained by comparing with the reference sound energy in the clean state of the region), the least squares optimization algorithm is used to find the optimal solution, that is, the equivalent stain thickness corresponding to each sensor point can be inverted.

[0060] Step S234, spatial interpolation processing is performed according to the stain thickness information to obtain a virtual stain distribution map.

[0061] Step S24, a virtual environment parameter set is constructed according to the virtual wind speed distribution map and the virtual stain distribution map.

[0062] In this step, the obtained two-dimensional field data is subjected to feature compression and fusion by principal component analysis. The principal component analysis algorithm can extract a principal component feature vector that best represents the comprehensive microclimate state of the component surface from the two highly correlated parameters of wind speed and stain thickness (for example, high-stain areas are often accompanied by low wind speed). The finally generated virtual environment parameter set is a multi-dimensional, grid-based data structure that not only contains the original field information, but also generates a low-dimensional, efficient feature representation that can more comprehensively reflect the real operating environment of the component, providing a high-quality, integrated driving input for the subsequent digital twin model.

[0063] Step S3, the meteorological data and the virtual environment parameter set are sent to a preset digital twin model for processing to obtain an initial temperature prediction value set;

[0064] This step performs parallel simulation by constructing a group of digital twin models with different parameters, with each model representing a possible physical state. The output of the model group is no longer a certain temperature value, but a prediction set containing probability distribution information. This method is particularly suitable for the dispersed nature of component performance in photovoltaic power stations, making the prediction result self-contained with a measure of its own credibility, providing more abundant basis for risk assessment.

[0065] The step S3 further includes steps S31-S37, which specifically include:

[0066] Step S31, finite element modeling is performed based on the material properties and structural dimensions of the photovoltaic equipment to obtain a finite element model;

[0067] Establishing a finite element model of photovoltaic equipment is a well-known technical solution to those skilled in the art, and therefore will not be described here.

[0068] Step S32, posterior distribution estimation processing is performed according to the finite element model and historical operation data to obtain the posterior probability distribution of the key thermophysical parameters;

[0069] In this step, the finite element model is taken as the forward model, and the key thermophysical parameters are taken as unknown quantities to be calibrated, and reasonable prior probability distributions are set for them. The irradiance, ambient temperature and wind speed monitored during the historical operation are taken as model inputs, and the component temperature measured at the corresponding time is taken as observation data. Through the Markov chain Monte Carlo sampling method, random walk is performed in the parameter space to explore which parameter combination can make the model output best match the historical observation data, so as to obtain the posterior probability distribution of the key thermophysical parameters. It should be noted that the key thermophysical parameters include but are not limited to the thermal conductivity of the back plate and the specific heat capacity of the packaging material.

[0070] Step S33, according to the posterior probability distribution of the key thermophysical parameters, a model population sampling process based on optimal Latin hypercube experimental design is performed to obtain a virtual component model instance set;

[0071] In this step, optimal Latin hypercube sampling is an experimental design method that can uniformly fill the entire space with the least number of sample points in a multi-dimensional parameter space (each dimension corresponds to a parameter to be sampled). Here, the sampling space is the posterior distribution of each parameter. Through optimal Latin hypercube sampling, hundreds to thousands of different parameter combinations can be systematically generated, and each parameter combination instantiates a component model with different parameters. The collection of these model instances constitutes a population model that can fully represent the performance dispersion of the component caused by manufacturing tolerances, material batch differences and different degrees of aging.

[0072] Step S34, constructing a digital twin model based on the virtual component model instance set;

[0073] Step S35, according to the digital twin model, the meteorological data and the virtual environmental parameter set, solving the body temperature field evolution sequence of each model;

[0074] In this step, the meteorological forecast data (irradiance, ambient temperature) for a future period of time and the virtual wind speed distribution map and virtual stain distribution map included in the virtual environmental parameter set are taken as unified input conditions to simultaneously drive the constructed digital twin model to perform transient thermal simulation. Each model independently solves its energy conservation equation, where the virtual wind speed distribution map is taken as the boundary condition for convective heat dissipation, and the virtual stain distribution map modifies the solar radiation heat absorption by affecting the optical properties of the component surface. Through parallel computing, the temperature distribution of each model instance at each time step in the future is quickly obtained, forming its unique temperature field evolution sequence.

[0075] Step S36, according to the body temperature field evolution sequence of each model, performing population output statistical aggregation processing to obtain the set mean field and set variance field of the temperature prediction;

[0076] In this step, for each location point in the power station, at each prediction time point, the predicted temperature values of all model individuals at the point are taken as a statistical sample set. The arithmetic mean of the set is calculated to obtain a set mean field, representing the most likely temperature prediction considered by the model population. At the same time, the variance of the set is calculated to obtain a set variance field, quantifying the degree of prediction disagreement due to model parameter uncertainty. A large variance area means that the prediction result is unreliable and the risk is high; a small variance area means that the prediction confidence is high.

[0077] Step S37, probabilistic prediction set construction processing is performed according to the set mean field and the set variance field of the temperature prediction to obtain an initial temperature prediction value set and a distribution description thereof.

[0078] This step converts the determined prediction value into a probabilistic prediction. It is assumed that at each location and each time, the predicted temperature of the model population obeys a probability distribution with the set mean as the expectation and the set variance as the variance, so that the final output is no longer a single temperature value, but a probability distribution. A specific embodiment is that the prediction result can be expressed as a 90% possibility of the temperature of a certain photovoltaic device at a certain point falling between 55°C and 62°C.

[0079] Step S4, determining a temperature prediction value distribution map according to the initial temperature prediction value set and the measured temperature data of the photovoltaic device, and comparing the temperature prediction result with an abnormal temperature threshold to obtain a temperature abnormal risk point;

[0080] The step S4 further includes steps S41-S44, which specifically include:

[0081] Step S41, performing weight dynamic calculation according to the initial temperature prediction value set and the measured temperature data of the photovoltaic device to obtain a dynamic adaptive weight coefficient;

[0082] In this step, the specific calculation process of the dynamic adaptive weight coefficient includes:

[0083]

[0084] In the above formula, and respectively represent the dynamic adaptive weight coefficient of the i-th twin model at the current time t and the last time t-1; represents a smoothing factor, and the value range is 0.9-0.99; represents the prediction residual of the i-th twin model at time t; The instantaneous performance evaluation item for the current prediction residual is that when the model predicts very accurately at the current time, this item is close to 1, and a positive adjustment is made to the weight, and when the prediction error is large, this item is close to 0, and the weight of the model is significantly reduced. The dynamic weight allocation using the formula enables the system to adapt to environmental changes and photovoltaic device aging, automatically optimize the model with stable performance, and enable the digital twin system to have online self-learning capability.

[0085] Step S42, weighted fusion processing is performed according to the dynamic adaptive weight coefficient and the initial temperature prediction value set to obtain a temperature prediction value distribution map;

[0086] Step S43, dynamic abnormal threshold calculation is performed according to the temperature prediction value distribution map, the meteorological data and the rated parameters of the photovoltaic device to obtain a dynamic abnormal temperature threshold distribution map;

[0087] In this step, the dynamic abnormal threshold calculation process specifically includes:

[0088]

[0089] In the above formula, represents the dynamic abnormal temperature threshold; represents the ambient temperature; represents the solar irradiance; represents the solar radiation absorption rate of the photovoltaic device surface; represents the photoelectric conversion efficiency; represents the comprehensive convection and radiation heat transfer coefficient; represents the statistical safety margin. It should be noted that is the sum of the convection heat transfer coefficient and the radiation heat transfer coefficient, wherein the convection heat transfer coefficient is determined by , a and b are empirical constants whose values are related to the flow state and the geometric characteristics of the photovoltaic device surface; the radiation heat transfer coefficient is determined by , represents the Stefan-Boltzmann constant, represents the emissivity of the photovoltaic device surface, represents the characteristic temperature, which is the average of the absolute temperature of the photovoltaic device surface and the absolute temperature of the environment.

[0090] It can be understood that this step calculates the upper limit of the temperature that the photovoltaic device can reach under normal conditions under the current irradiation, wind speed and ambient temperature. When there is strong light at noon, the threshold value will automatically increase; when it is early in the morning or the wind speed is large, the threshold value will decrease. This dynamic threshold effectively overcomes the shortcomings of high false alarm rate or missed report of the fixed threshold under varying working conditions.

[0091] Step S44, determining temperature anomaly risk points according to the dynamic anomaly temperature threshold distribution map and the temperature prediction value distribution map.

[0092] It can be understood that the dynamic anomaly temperature threshold distribution map and the temperature prediction value distribution map are compared element by element, and the spatial points whose predicted temperature values exceed the local dynamic threshold are preliminarily marked as temperature anomaly risk points. In addition, this step introduces a filtering rule: only when the predicted temperature exceeds the threshold for more than 3 time steps can it be confirmed as a valid risk point, so as to avoid misjudgment caused by transient fluctuations.

[0093] Step S5, processing according to the temperature prediction value distribution map, the temperature anomaly risk points and the meteorological data to obtain photovoltaic device temperature anomaly early warning information.

[0094] The step S5 further includes steps S51-S54, which specifically include:

[0095] Step S51, correcting processing according to the temperature prediction value distribution map and the meteorological data to obtain a corrected temperature prediction distribution map;

[0096] This step corrects the prediction result by introducing an equivalent sky cold source model and dew point temperature. The specific process is as follows: cloud cover and dew point temperature are obtained from meteorological data. The equivalent sky cold source model regards the sky as a cold source with an equivalent radiation temperature, which is lower than the ambient temperature, especially at night. Through model calculation, the long-wave radiation heat exchange between the component surface and the sky is calculated, and the prediction temperature dominated by solar radiation and convection is corrected, especially at night or in the absence of sunlight. This correction can more accurately reflect the actual cooling process of the component. At the same time, it is judged whether the component surface temperature is close to or lower than the dew point temperature. If it is close, it means that dew may occur on the surface, and the evaporation of dew moisture will absorb a large amount of heat, producing additional cooling effect. By introducing a non-steady-state mass transfer model to quantify this evaporation heat absorption process, the predicted temperature is further corrected. This step compensates for the shortcomings of traditional thermal models that only focus on convection and solar radiation, significantly improving the temperature prediction accuracy in the morning, at night or in high humidity conditions, and in other specific scenarios.

[0097] It should be noted that the equivalent sky cold source model is specifically:

[0098]

[0099] In the above formula, represents the net radiation heat exchange rate; represents the Stefan-Boltzmann constant; represents the emissivity of the photovoltaic device surface; represents the thermodynamic temperature of the surface of the photovoltaic device; represents the effective radiative temperature of the sky. It is not a constant value, and it is significantly affected by factors such as water vapor content and cloud cover in the atmosphere, which is the key to adapting the model to complex weather, and the specific calculation process is as follows:

[0100]

[0101] In the above formula, represents the effective emission rate of the sky under clear sky conditions, which is calculated by , wherein m and n represent empirical coefficients, is the water vapor pressure; and n represent empirical coefficients for cloud cover correction, which are determined by regression analysis of local historical meteorological data and radiation observation data; C represents total cloud cover.

[0102] Step S52, determining a temperature anomaly region in the corrected temperature prediction distribution map;

[0103] Step S53, performing anomaly diagnosis according to the temperature anomaly region and the virtual stain distribution map to obtain a diagnosis result;

[0104] In this step, the temperature anomaly region and the virtual stain distribution map are spatially superimposed and analyzed, and the proportion of pixel points in the temperature anomaly region that also belong to the high stain region and the proportion of the high stain region that appears in the temperature anomaly region are calculated. If both are greater than 80%, it can be inferred that the possibility of the anomaly being caused by stain accumulation leading to local hot spots is extremely high.

[0105] Step S54, generating photovoltaic device temperature anomaly warning information according to the diagnosis result.

[0106] In this step, the photovoltaic device temperature anomaly warning information is a structured report, and its content includes the geographic location or number of the abnormal photovoltaic device, the risk level, the predicted temperature value, the possible cause, etc. Through the warning information, a closed loop from prediction, diagnosis to decision support is realized.

[0107] Embodiment 2:

[0108] The embodiment provides a photovoltaic device temperature anomaly prediction device of a digital twin model, and the device comprises an acquisition module, a first processing module, a second processing module, a third processing module and a fourth processing module, and specifically comprises:

[0109] The acquisition module is used to acquire an environmental acoustic signal, meteorological data and measured temperature data of a photovoltaic device.

[0110] The first processing module is configured to perform acoustic spectrum feature extraction on the environmental acoustic signal to obtain a virtual environmental parameter set, wherein the virtual environmental parameter set comprises a virtual wind speed distribution map and a virtual stain distribution map.

[0111] The second processing module is configured to send the meteorological data and the virtual environmental parameter set to a preset digital twin model for processing to obtain an initial temperature prediction value set.

[0112] The third processing module is configured to determine a temperature prediction value distribution map according to the initial temperature prediction value set and measured temperature data of the photovoltaic device, compare the temperature prediction result with an abnormal temperature threshold, and obtain a temperature abnormality risk point.

[0113] The fourth processing module is configured to process the temperature prediction value distribution map, the temperature abnormality risk point and the meteorological data to obtain photovoltaic device temperature abnormality early warning information.

[0114] In one specific embodiment of the present disclosure, the first processing module further comprises a first processing unit, a second processing unit, a third processing unit and a fourth processing unit, which specifically comprise:

[0115] The first processing unit is configured to perform enhancement processing on the environmental acoustic signal to obtain a feature vector of enhanced acoustic features.

[0116] The second processing unit is configured to perform inversion on the surface wind speed of the photovoltaic device according to the feature vector of enhanced acoustic features to obtain a virtual wind speed distribution map.

[0117] The third processing unit is configured to predict the surface stain thickness of the photovoltaic device according to the virtual wind speed distribution map and the feature vector of enhanced acoustic features to obtain a virtual stain distribution map.

[0118] The fourth processing unit is configured to construct a virtual environmental parameter set according to the virtual wind speed distribution map and the virtual stain distribution map.

[0119] In one specific embodiment of the present disclosure, the second processing unit further comprises a fifth processing unit, a sixth processing unit, a seventh processing unit and an eighth processing unit, which specifically comprise:

[0120] The fifth processing unit is configured to perform acoustic coherence feature extraction processing on the feature vector of enhanced acoustic features to obtain a frequency band energy matrix, wherein the frequency band energy matrix comprises a spatial coherence corresponding to a frequency band.

[0121] The sixth processing unit is configured to perform forward modeling according to the frequency band energy matrix to obtain a forward model.

[0122] a seventh processing unit, configured to solve surface friction wind speed according to the forward model and the frequency band energy matrix to obtain a surface friction wind speed field;

[0123] an eighth processing unit, configured to determine a virtual wind speed distribution map according to the surface friction wind speed field.

[0124] In one specific embodiment of the present disclosure, the third processing unit further includes a ninth processing unit, a tenth processing unit, an eleventh processing unit, and a twelfth processing unit, which specifically include:

[0125] the ninth processing unit is configured to perform secondary separation processing of particle impact sound and wind speed background sound according to the virtual wind speed distribution map and the feature vector of the enhanced acoustic feature to obtain a sound energy feature;

[0126] the tenth processing unit is configured to perform normalization processing according to the sound energy feature and the virtual wind speed distribution map to obtain a particle impact sound energy distribution after wind speed normalization;

[0127] the eleventh processing unit is configured to perform inversion processing according to the particle impact sound energy distribution after wind speed normalization to obtain stain thickness information;

[0128] the twelfth processing unit is configured to perform spatial interpolation processing according to the stain thickness information to obtain a virtual stain distribution map.

[0129] It should be noted that, as for the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment related to the method, and will not be described in detail here.

[0130] Embodiment 3:

[0131] Corresponding to the above method embodiment, the present embodiment also provides a digital twin model photovoltaic device temperature anomaly prediction device. The digital twin model photovoltaic device temperature anomaly prediction device described below can be mutually corresponding to the digital twin model photovoltaic device temperature anomaly prediction method described above.

[0132] Figure 2 is a block diagram of a digital twin model photovoltaic device temperature anomaly prediction device 800 according to an exemplary embodiment. As Figure 2 shown, the digital twin model photovoltaic device temperature anomaly prediction device 800 can include a processor 801, a memory 802. The digital twin model photovoltaic device temperature anomaly prediction device 800 can also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0133] The processor 801 is configured to control overall operation of the photovoltaic device temperature anomaly prediction device 800 of the digital twin model, so as to complete all or part of the steps in the photovoltaic device temperature anomaly prediction method of the digital twin model described above. The memory 802 is configured to store various types of data to support the operation of the photovoltaic device temperature anomaly prediction device 800 of the digital twin model. For example, the data can include instructions for operating any application or method on the photovoltaic device temperature anomaly prediction device 800 of the digital twin model, and application-related data, such as contact data, sent and received messages, pictures, audio, video, and the like. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 803 can include a screen and an audio component. The screen can be a touch screen, for example, and the audio component is configured to output and / or input audio signals. For example, the audio component can include a microphone configured to receive external audio signals. The received audio signals can be further stored in the memory 802 or transmitted through the communication component 805. The audio component also includes at least one speaker configured to output audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, which can be a keyboard, a mouse, a button, and the like. The buttons can be virtual buttons or physical buttons. The communication component 805 is configured to perform wired or wireless communication between the photovoltaic device temperature anomaly prediction device 800 of the digital twin model and other devices. The wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include a Wi-Fi module, a Bluetooth module, an NFC module.

[0134] In an example embodiment, the photovoltaic device temperature anomaly prediction device 800 of the digital twin model can be implemented by one or more Application Specific Integrated Circuit (ASIC), Digital Signal Processor (DSP), Digital Signal Processing Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic elements for performing the above-mentioned photovoltaic device temperature anomaly prediction method of the digital twin model.

[0135] In another example embodiment, a computer readable storage medium including program instructions is also provided, which when executed by a processor, implement the steps of the above-mentioned photovoltaic device temperature anomaly prediction method of the digital twin model. For example, the computer readable storage medium can be the above-mentioned memory 802 including program instructions, which can be executed by the processor 801 of the photovoltaic device temperature anomaly prediction device 800 of the digital twin model to complete the above-mentioned photovoltaic device temperature anomaly prediction method of the digital twin model.

[0136] Embodiment 4:

[0137] Corresponding to the above method embodiments, in this embodiment, a readable storage medium is also provided, which can be referred to each other below described a readable storage medium and above described a photovoltaic device temperature anomaly prediction method of the digital twin model.

[0138] A readable storage medium, the readable storage medium stores a computer program, the computer program is executed by a processor to implement the steps of the above-mentioned photovoltaic device temperature anomaly prediction method of the digital twin model of the method embodiments.

[0139] The readable storage medium can be specifically a U disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and various readable storage media that can store program codes.

[0140] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0141] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0141] The above merely describes the preferred embodiments of the present application, and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for predicting temperature anomalies in photovoltaic equipment based on a digital twin model, characterized in that, include: Acquire environmental acoustic signals, meteorological data, and measured temperature data of photovoltaic equipment; Based on the environmental acoustic signal, acoustic spectrum features are extracted to obtain a virtual environmental parameter set, which includes a virtual wind speed distribution map and a virtual stain distribution map. The meteorological data and the set of virtual environmental parameters are sent to a preset digital twin model for processing to obtain an initial set of temperature prediction values; The temperature prediction result is determined based on the initial temperature prediction value set and the measured temperature data of the photovoltaic equipment, and the temperature prediction result is compared with the abnormal temperature threshold to obtain the temperature abnormality risk point. Based on the temperature prediction distribution map, the temperature anomaly risk points, and the meteorological data, an early warning information for abnormal temperature of photovoltaic equipment is obtained. Specifically, acoustic spectral features are extracted based on the environmental acoustic signals to obtain a virtual environment parameter set, including: The environmental acoustic signal is enhanced to obtain a feature vector with enhanced acoustic features; Based on the feature vector of the enhanced acoustic features, the surface wind speed of the photovoltaic device is inverted to obtain a virtual wind speed distribution map. The thickness of the surface stains on the photovoltaic equipment is predicted based on the virtual wind speed distribution map and the feature vector of the enhanced acoustic features, thus obtaining the virtual stain distribution map; A virtual environmental parameter set is constructed based on the virtual wind speed distribution map and the virtual stain distribution map; The prediction of the surface dirt thickness of photovoltaic equipment based on the virtual wind speed distribution map and the feature vector of the enhanced acoustic features includes: Based on the virtual wind speed distribution map and the feature vector of the enhanced acoustic features, a secondary separation process is performed between particle impact sound and wind speed background sound to obtain acoustic energy features. Based on the acoustic energy characteristics and the virtual wind speed distribution map, normalization processing is performed to obtain the particle impact acoustic energy distribution after wind speed normalization. The stain thickness information is obtained by inversion processing based on the particle impact acoustic energy distribution after wind speed normalization. Spatial interpolation is performed based on the stain thickness information to obtain a virtual stain distribution map.

2. The method for predicting temperature anomalies in photovoltaic equipment based on a digital twin model according to claim 1, characterized in that, The surface wind speed of the photovoltaic device is inverted based on the feature vector of the enhanced acoustic features, including: The feature vectors of the enhanced acoustic features are subjected to acoustic coherence feature extraction processing to obtain a frequency band energy matrix, wherein the frequency band energy matrix includes the spatial coherence corresponding to the frequency band; Forward modeling is performed based on the frequency band energy matrix to obtain the forward model; The surface friction wind speed field is obtained by solving the surface friction wind speed according to the forward model and the frequency band energy matrix. A virtual wind speed distribution map is determined based on the surface friction wind speed field.

3. The method for predicting temperature anomalies in photovoltaic equipment based on a digital twin model according to claim 1, characterized in that, The meteorological data and the virtual environmental parameter set are sent to a preset digital twin model for processing, including: Finite element modeling is performed based on the material properties and structural dimensions of photovoltaic equipment to obtain the finite element model. Based on the finite element model and historical operating data, posterior distribution estimation is performed to obtain the posterior probability distribution of key thermophysical parameters; Based on the posterior probability distribution of the key thermophysical parameters, a model population sampling process based on the optimal Latin hypercube experimental design is performed to obtain a set of virtual component model instances. A digital twin model is constructed based on the set of virtual component model instances; The temperature field evolution sequence of each model is obtained by solving the problem based on the digital twin model, the meteorological data, and the virtual environment parameter set. Based on the evolution sequence of the body temperature field of each model, statistical aggregation of the group output is performed to obtain the ensemble mean field and ensemble variance field of the temperature prediction. Based on the set mean field and set variance field of the temperature prediction, a probabilistic prediction set construction process is performed to obtain the initial temperature prediction value set and its distribution description.

4. The method for predicting temperature anomalies in photovoltaic equipment based on a digital twin model according to claim 1, characterized in that, The temperature prediction result is determined based on the initial set of predicted temperature values ​​and the measured temperature data of the photovoltaic device, and the temperature prediction result is compared with the abnormal temperature threshold, including: Based on the initial temperature prediction set and the measured temperature data of the photovoltaic equipment, dynamic weight calculation is performed to obtain dynamic adaptive weight coefficients. A temperature prediction distribution map is obtained by performing a weighted fusion process based on the dynamic adaptive weighting coefficients and the initial temperature prediction value set. Based on the temperature prediction distribution map, the meteorological data, and the rated parameters of the photovoltaic equipment, a dynamic anomaly threshold is calculated to obtain a dynamic anomaly temperature threshold distribution map. Temperature anomaly risk points are determined based on the dynamic abnormal temperature threshold distribution map and the temperature prediction value distribution map.

5. A photovoltaic equipment temperature anomaly prediction device based on a digital twin model, characterized in that, include: The acquisition module is used to acquire environmental acoustic signals, meteorological data, and measured temperature data of photovoltaic equipment. The first processing module is used to extract acoustic spectrum features based on the environmental acoustic signal to obtain a virtual environmental parameter set, which includes a virtual wind speed distribution map and a virtual stain distribution map. The second processing module is used to send the meteorological data and the virtual environment parameter set to a preset digital twin model for processing to obtain an initial temperature prediction value set; The third processing module is used to determine the temperature prediction value distribution map based on the initial temperature prediction value set and the measured temperature data of the photovoltaic equipment, and compare the temperature prediction results with the abnormal temperature threshold to obtain the temperature abnormality risk points. The fourth processing module is used to process the temperature prediction distribution map, the temperature anomaly risk points, and the meteorological data to obtain photovoltaic equipment temperature anomaly early warning information. The first processing module includes: The first processing unit is used to enhance the environmental acoustic signal to obtain a feature vector with enhanced acoustic features. The second processing unit is used to invert the surface wind speed of the photovoltaic device based on the feature vector of the enhanced acoustic features to obtain a virtual wind speed distribution map. The third processing unit is used to predict the thickness of the stains on the surface of the photovoltaic equipment based on the virtual wind speed distribution map and the feature vector of the enhanced acoustic features, so as to obtain the virtual stain distribution map. The fourth processing unit is used to construct a virtual environmental parameter set based on the virtual wind speed distribution map and the virtual stain distribution map; The third processing unit includes: The ninth processing unit is used to perform secondary separation processing of particle impact sound and wind speed background sound based on the virtual wind speed distribution map and the feature vector of the enhanced acoustic features to obtain acoustic energy features; The tenth processing unit is used to perform normalization processing based on the acoustic energy characteristics and the virtual wind speed distribution map to obtain the particle impact acoustic energy distribution after wind speed normalization. The eleventh processing unit is used to perform inversion processing based on the particle impact acoustic energy distribution after wind speed normalization to obtain stain thickness information. The twelfth processing unit is used to perform spatial interpolation processing based on the stain thickness information to obtain a virtual stain distribution map.

6. The photovoltaic equipment temperature anomaly prediction device based on a digital twin model according to claim 5, characterized in that, The second processing unit includes: The fifth processing unit is used to perform acoustic coherence feature extraction processing on the feature vector of the enhanced acoustic features to obtain a frequency band energy matrix, wherein the frequency band energy matrix includes the spatial coherence corresponding to the frequency band. The sixth processing unit is used to perform forward modeling based on the frequency band energy matrix to obtain a forward model; The seventh processing unit is used to solve the surface friction wind speed according to the forward model and the frequency band energy matrix to obtain the surface friction wind speed field. The eighth processing unit is used to determine a virtual wind speed distribution map based on the surface friction wind speed field.

Images