Intelligent temperature regulation device and method for photovoltaic cell assembly

Abstract

The present invention relates to an intelligent temperature regulating device and method for a photovoltaic cell assembly, and belongs to the technical field of photovoltaic panel temperature regulation. It includes a data acquisition module, a data processing module, a temperature prediction module, a spray control module and a liquid spray module; the spray control module is used to control the liquid spray module to complete the cleaning and cooling operations by executing a cleaning strategy and a cooling strategy; the data processing module is used to save the collected electric energy data and environmental monitoring data and establish a database, and use the data in the database to construct a training sample for temperature prediction. The present invention executes a cleaning strategy by comparing the predicted temperature of the photovoltaic assembly with the measured temperature, and then sprays and cools the photovoltaic assembly according to the measured temperature, thereby avoiding blind cooling operations; at the same time, the decision-making cleaning and cooling operations reduce the corrosion of the surface of the photovoltaic assembly by water and dust, thereby improving the power generation efficiency and safety of the photovoltaic assembly.

Description

Technical Field

[0001] The invention belongs to the technical field of photovoltaic panel temperature regulation, and in particular relates to an intelligent temperature regulation device and method for a photovoltaic cell assembly. Background Art

[0002] There are three main forms of solar energy consumption absorbed by photovoltaic modules, of which about 20% is converted into electrical energy, about 80% is converted into heat energy, and the remaining small part is reflected back into the air. The conversion of absorbed solar energy into heat energy will cause the temperature of photovoltaic modules to rise, which in turn affects the power generation efficiency of the modules. At present, photovoltaic module cooling technology is mainly divided into air cooling and water cooling. Air cooling technology mainly uses the temperature difference between air and photovoltaic modules to produce heat exchange to achieve the purpose of cooling; the main technical means include increasing ventilation ducts, installing fins, and introducing fans, etc., thereby increasing the air cooling heat exchange coefficient or heat exchange area to promote the cooling of photovoltaic modules. Water cooling technology mainly uses water film cooling, using water as the cooling medium, and using nozzles or drip pipes and other devices to spray water directly on the surface of photovoltaic modules, relying on direct contact between photovoltaic modules and water for heat exchange.

[0003] Air cooling technology is widely used in photovoltaic buildings due to its advantages such as low investment cost and simple structure. However, the cooling effect of air cooling technology is limited. For mechanical ventilation air cooling technology, there is still a balance problem between fan energy consumption and photoelectric conversion efficiency improvement. Compared with air cooling technology, water cooling technology has a better cooling effect and can also clean photovoltaic modules. However, the water cooling process inevitably causes water loss and causes certain corrosion to photovoltaic modules. At the same time, in the process of water cooling operation control, decisions are made only through temperature comparison, without considering the operating status of photovoltaic modules. Blind water cooling decisions result in waste of water resources and mediocre results. Summary of the invention

[0004] In order to solve the above-mentioned problems existing in the prior art, the present invention provides an intelligent temperature adjustment device and method for photovoltaic cell modules, which executes a cleaning strategy by comparing the predicted temperature of the photovoltaic module with the actual temperature, and then sprays and cools the photovoltaic module according to the actual temperature, thereby avoiding blind cooling operations and saving water resources; at the same time, the decision-making cleaning and cooling operations reduce the corrosion of the surface of the photovoltaic module by water and dust, thereby improving the power generation efficiency and safety of the photovoltaic module.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A first aspect of the present disclosure provides an intelligent temperature regulating device for a photovoltaic cell assembly, comprising a data acquisition module, a data processing module, a temperature prediction module, an injection control module and a liquid injection module;

[0007] The injection control module is used to control the liquid injection module to complete the cleaning and cooling operations by executing the cleaning strategy and the cooling strategy;

[0008] The cleaning strategy includes:

[0009] Set the temperature difference threshold of the measuring point, obtain the temperature data of multi-point measurement of the temperature sensor array, and establish the temperature field distribution. When a measuring point with a temperature difference greater than the temperature difference threshold appears in the temperature field, it is judged as a hot spot failure, otherwise the temperature distribution is normal;

[0010] Set the PV module temperature judgment threshold, use the predicted temperature of the PV module as the reference data, and the measured temperature of the PV module as the measured data. Calculate the absolute difference between the measured data and the reference data and compare it with the temperature judgment threshold. When the absolute difference exceeds the temperature judgment threshold, it is determined that the module state is abnormal, otherwise it is normal.

[0011] When the judgment result is that the temperature distribution is normal and the component state is abnormal, a jet cleaning command is generated;

[0012] When the judgment result is a hot spot fault and the component status is normal, a fault warning is generated;

[0013] When the judgment result is a hot spot failure and an abnormal component state, a jet cleaning command is generated;

[0014] The cooling strategies include:

[0015] Set a temperature adjustment threshold. When the actual temperature data of the photovoltaic module is greater than the temperature adjustment threshold, set the injection pressure and start the liquid injection module according to the wind direction and wind speed.

[0016] When the cleaning operation is completed before the cooling operation is performed, the spray water collection operation is performed for recycling and reducing water resource loss.

[0017] As a preferred technical solution of the present invention, the injection control module includes:

[0018] A microcontroller to which is connected at least one module having the following functions:

[0019] A data acquisition module for collecting monitoring data;

[0020] A data processing module for storing monitoring data;

[0021] A temperature prediction module for predicting the temperature of photovoltaic modules;

[0022] Wireless transmission module for data transmission;

[0023] Liquid jetting module for performing cleaning and cooling operations.

[0024] As a preferred technical solution of the present invention, the data acquisition module is used to collect real-time electric energy data and environmental monitoring data of the photovoltaic power generation system, and transmit the data to the data processing module through a communication device, wherein the electric energy data includes current, voltage and power, and the environmental monitoring data includes the ambient temperature and humidity, irradiance, cloud observation data, wind speed and direction and temperature of the photovoltaic components.

[0025] As a preferred technical solution of the present invention, the data processing module is used to save the collected power data and environmental monitoring data and establish a database, and use the data in the database to construct training samples for temperature prediction.

[0026] As a preferred technical solution of the present invention, the processing flow of the data processing module includes the following steps:

[0027] Select prediction influencing factors: classify the data in the database, including independent variables and dependent variables, conduct correlation analysis on the independent variables and dependent variables, and select independent variables with non-zero correlation as prediction factors; the independent variables include the ambient temperature and humidity, irradiance and wind speed and direction of the working environment of the photovoltaic modules, and the dependent variables include the temperature of the photovoltaic modules;

[0028] Solve the prior temperature of the photovoltaic module: Use the output characteristic curve and irradiance of the photovoltaic module to calculate the temperature of the photovoltaic module as the prior temperature input of the temperature prediction model;

[0029] Calculate the prior temperature T x The formula is:

[0030] ;

[0031] In the formula, is the modified Fréchet distance, G is the irradiance, a , b is a constant, and the calculation formula is as follows:

[0032] ;

[0033] ;

[0034] Where, d F1 With d F2 , T1 and T2 are the Fréchet distance and temperature at different temperatures under the same irradiance; d F3 With d F4 , G3 and G4 are the Fréchet distance and irradiance at the same temperature and different irradiances;

[0035] Corrected Fréchet distance The calculation formula is as follows:

[0036] ;

[0037] Where, d F is the Fréchet distance, is the penalty factor, is the relative overshoot, and the calculation formula is:

[0038] ;

[0039] In the formula, U P1 , U P2 are the maximum peak values ​​of characteristic curve L1 and characteristic curve L2 respectively;

[0040] Establish training samples for temperature prediction: collect sample data based on the screened prediction influencing factors and solve the prior temperature of the sample data to construct training samples for the temperature prediction model.

[0041] As a preferred technical solution of the present invention, the temperature prediction module is used to predict the temperature of the photovoltaic module by constructing a temperature prediction model using the screened sample data of prediction influencing factors and the prior temperature of the sample data;

[0042] The temperature prediction process comprises the following steps:

[0043] Establish a temperature prediction model: use the selected prediction factor type data and the corresponding prior temperature as input variables, the PV module temperature as output, and set the number of hidden layer nodes to establish a neural network model;

[0044] Training the temperature prediction model: Normalize the input variables and set the error target to 10 -8 Train the neural network of photovoltaic module temperature prediction model;

[0045] Input data to predict temperature: Input the collected data into the trained prediction model according to the data type of the model input to obtain the output temperature.

[0046] As a preferred technical solution of the present invention, the liquid spraying module is used to perform cleaning operations and cooling operations by spraying liquid, and the liquid spraying module includes at least one module having the following functions:

[0047] Relay module for controlling injection action;

[0048] A water pump module for completing the jetting action;

[0049] Control module for controlling injection pressure.

[0050] A second aspect of the present disclosure provides an intelligent temperature adjustment method for a photovoltaic cell assembly, which is applied to an intelligent temperature adjustment device for a photovoltaic cell assembly as described above, and comprises the following steps:

[0051] Collect monitoring data: collect real-time power data and environmental monitoring data of the photovoltaic power generation system, and transmit the data to the data processing module through communication equipment;

[0052] Data preprocessing: By saving the collected power data and environmental monitoring data and establishing a database, the data in the database is used to construct training samples for temperature prediction;

[0053] PV panel temperature prediction: Based on the established temperature prediction training samples, a temperature prediction model is established for training, and the collected data is input into the trained prediction model to obtain the temperature output;

[0054] Execute cleaning strategy: control the liquid injection module to complete the cleaning operation by executing the cleaning strategy;

[0055] Execute the cooling strategy: set the temperature adjustment threshold. When the measured temperature data of the photovoltaic module is greater than the temperature adjustment threshold, set the injection pressure and start the liquid injection module according to the wind direction and wind speed.

[0056] Furthermore, the data preprocessing comprises the following steps:

[0057] Select prediction influencing factors: classify the data in the database, including independent variables X and dependent variables Y, conduct correlation analysis on independent variables X and dependent variables Y, and select independent variables with non-zero correlation as prediction factors;

[0058] The specific selection process includes data conversion and correlation coefficient calculation:

[0059] k =( u - w ) / δ ;

[0060] In the formula, k represents the data after mean normalization. u represents the average value of the original data, w represents the normalized data, δ Represents the variance of the original data set;

[0061] r=cov (X,Y) / σxσy ;

[0062] In the formula, r represents the Person correlation coefficient, cov represents the covariance, σ represents standard deviation;

[0063] Solve the prior temperature of the photovoltaic module: Use the output characteristic curve and irradiance of the photovoltaic module to calculate the temperature of the photovoltaic module, which is used as the prior temperature input of the temperature prediction model. Other environmental factors are added to the temperature prediction model to correct the prior temperature.

[0064] Establish training samples for temperature prediction: collect sample data based on the screened prediction influencing factors and solve the prior temperature of the sample data to construct training samples for the temperature prediction model.

[0065] Further, the method of solving the prior temperature of the photovoltaic module includes the following steps:

[0066] Construct input vector: Construct a feature vector representing the characteristic curve of the photovoltaic module through the collected power data: Z=[(0, I sc ), (V m , I m ), (V oc ,0)];

[0067] Among them, (0, I sc ) short-circuit current point, (V m , I m ) is the maximum power point, (V oc , 0) is the open circuit voltage point;

[0068] Calculate the Fréchet distance d F : For two characteristic curves L1 and L2 with lengths N1 and N2 respectively, let i ∈(0, N1], j ∈(0, N2] are two points of the two curves, P ( i )=( x i , y i )and q ( j )=( x j , y j ) is the coordinate of the point on the curve, and the Fréchet distance d is solved. F :

[0069] i ≠1, j ≠1;

[0070] In the formula, d[ P ( i ), q ( j )]for i ,j The direct distance between two points is calculated as:

[0071] ;

[0072] Modified Fréchet distance d F : Relative overshoot through the specified curve Modified Fréchet distance, calculated as follows:

[0073] ;

[0074] In the formula, is the modified Fréchet distance, is the penalty factor, is the relative overshoot, and the calculation formula is:

[0075] ;

[0076] In the formula, U P1 , U P2 are the maximum peak values ​​of characteristic curve L1 and characteristic curve L2 respectively;

[0077] Solving the prior temperature: Select a set of data under standard test conditions as a benchmark and solve the prior temperature T x :

[0078] ;

[0079] Where G is the irradiance, a , b is a constant, and the calculation formula is as follows:

[0080] ;

[0081] ;

[0082] Where, d F1 With d F2 , T1 and T2 are the Fréchet distance and temperature at different temperatures under the same irradiance; d F3 With d F4 , G3 and G4 are the Fréchet distance and irradiance at the same temperature but different irradiances.

[0083] The beneficial effects of the present invention are:

[0084] The present invention first eliminates the abnormal state of the photovoltaic module by comparing the predicted temperature of the photovoltaic module with the actual temperature, and then sprays and cools the photovoltaic module according to the actual temperature, making the temperature adjustment strategy more accurate and effective. At the same time, through the cleaning decision of the photovoltaic module, the problem of abnormal temperature change of the photovoltaic module is solved, and the power generation efficiency and safety of the photovoltaic module are improved.

[0085] In the temperature prediction process of photovoltaic modules, the present invention ensures the validity of model prediction results by adding a priori temperature and improves the accuracy of temperature prediction; at the same time, the difference in output characteristics of photovoltaic modules under two environmental conditions is evaluated by similarity, and a relationship equation with temperature and irradiance is established to calculate the priori temperature, taking into account the output characteristic curve of the photovoltaic module, so that the priori temperature is more in line with the target object, and then the Fréchet distance is corrected for correlation, so as to improve the accuracy of the Fréchet distance and the global correlation of the characteristic curve of the photovoltaic module. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0087] Figure 1 A schematic structural diagram of an intelligent temperature regulating device for a photovoltaic cell assembly provided by an embodiment of the present invention;

[0088] Figure 2 A schematic diagram of the steps of an intelligent temperature adjustment method for a photovoltaic cell assembly provided by an embodiment of the present invention;

[0089] Figure 3 A schematic diagram of the steps of data preprocessing provided by an embodiment of the present invention;

[0090] Figure 4 A schematic diagram of the steps for solving the prior temperature of a photovoltaic module provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0091] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0092] Photovoltaic power generation is currently the main way to utilize solar energy. The core components of photovoltaic power generation are photovoltaic cells based on single crystal or polycrystalline silicon. The maximum photoelectric conversion efficiency is about 30%, and the actual conversion efficiency is 10% to 26%. Most of the remaining solar energy is lost in the form of heat energy, causing the battery temperature to rise. The power generation efficiency of photovoltaic modules is greatly affected by their own temperature, and the power generation efficiency decreases as the operating temperature of the module increases. At the same time, the temperature of photovoltaic modules is also affected by environmental factors such as solar irradiance, dust and meteorological conditions, such as photovoltaic cell failure caused by hot spot effect. Therefore, taking effective cooling methods to reduce the temperature of photovoltaic modules can improve their photoelectric conversion efficiency.

[0093] At least one embodiment of the present disclosure provides an intelligent temperature regulating device for a photovoltaic cell assembly, such as Figure 1 As shown, it includes a data acquisition module, a data processing module, a temperature prediction module, an injection control module and a liquid injection module.

[0094] The data acquisition module is used to collect real-time power data and environmental monitoring data of the photovoltaic power generation system, and transmit the data to the data processing module through the communication equipment, wherein the power data includes current, voltage and power, etc., and the environmental monitoring data includes the ambient temperature and humidity, irradiance, cloud observation data, wind speed and direction and temperature of the photovoltaic components.

[0095] The data acquisition module includes an electric energy data acquisition module and an environmental data monitoring module. It is understandable that data acquisition acquires various data in the power distribution system in real time through sensors and other equipment. Electric energy data is acquired through current sensors and voltage sensors, and environmental monitoring data is detected through temperature and humidity sensors, radiometers, and wind speed and direction measuring instruments or acquired through meteorological data.

[0096] The data processing module is used to save the collected electric energy data and environmental monitoring data and establish a database, and use the data in the database to construct training samples for temperature prediction.

[0097] The processing flow of the data processing module includes the following steps:

[0098] Select prediction influencing factors: Classify the data in the database, including independent variables and dependent variables, conduct correlation analysis on the independent variables and dependent variables, and select independent variables with non-zero correlation as prediction factors. The independent variables include the ambient temperature and humidity, irradiance, wind speed and direction of the working environment of the photovoltaic module, and the dependent variable includes the temperature of the photovoltaic module.

[0099] Solve the prior temperature of the PV module: Use the output characteristic curve and irradiance of the PV module to calculate the temperature of the PV module as the prior temperature input of the temperature prediction model.

[0100] Calculate the prior temperature Tx The formula is:

[0101] ;

[0102] In the formula, is the modified Fréchet distance, G is the irradiance, a , b is a constant, and the calculation formula is as follows:

[0103] ;

[0104] ;

[0105] Where, d F1 With d F2 , T1 and T2 are the Fréchet distance and temperature at different temperatures under the same irradiance; d F3 With d F4 , G3 and G4 are the Fréchet distance and irradiance at the same temperature but different irradiances.

[0106] Corrected Fréchet distance The calculation formula is as follows:

[0107] ;

[0108] Where, d F is the Fréchet distance, is the penalty factor, is the relative overshoot, and the calculation formula is:

[0109] ;

[0110] In the formula, U P1 , U P2 They are the maximum peak values ​​of characteristic curve L1 and characteristic curve L2 respectively.

[0111] Establish training samples for temperature prediction: collect sample data based on the screened prediction influencing factors and solve the prior temperature of the sample data to construct training samples for the temperature prediction model.

[0112] In this embodiment, the temperature of the photovoltaic module is solved by the operating state and irradiance of the photovoltaic module, which is used as prior knowledge in subsequent temperature prediction, thereby ensuring the validity of the model prediction result and improving the accuracy of the prediction result.

[0113] The temperature prediction module is used to predict the temperature of the photovoltaic module by constructing a temperature prediction model using the screened sample data of the prediction influencing factors and the priori temperature of the sample data.

[0114] The temperature prediction process comprises the following steps:

[0115] Establish a temperature prediction model: use the screened prediction factor type data and the corresponding prior temperature as input variables, the PV module temperature as output, and set the number of hidden layer nodes to establish a neural network model.

[0116] Training the temperature prediction model: Normalize the input variables and set the error target to 10 -8 Train the neural network of photovoltaic module temperature prediction model.

[0117] Input data to predict temperature: Input the collected data into the trained prediction model according to the data type of the model input to obtain the output temperature.

[0118] It is understandable that the temperature is predicted by the model, and other influencing factors are added to the analysis of the calculated results of the PV module temperature, that is, the calculated prior temperature is corrected. The two promote each other and the accuracy of the obtained PV module temperature.

[0119] The injection control module is used to control the liquid injection module to complete the cleaning and cooling operations by executing the cleaning strategy and the cooling strategy. The injection control module includes:

[0120] A microcontroller to which is connected at least one module having the following functions:

[0121] A data acquisition module for collecting monitoring data;

[0122] A data processing module for storing monitoring data;

[0123] A temperature prediction module for predicting the temperature of photovoltaic modules;

[0124] Wireless transmission module for data transmission;

[0125] Liquid jetting module for performing cleaning and cooling operations.

[0126] The cleaning strategy includes:

[0127] Set the temperature difference threshold of the measurement point, obtain the temperature data of multi-point measurement of the FBG temperature sensor array, and establish the temperature field distribution. When a measurement point with a temperature difference greater than the temperature difference threshold appears in the temperature field, it is judged as a hot spot fault, otherwise the temperature distribution is normal;

[0128] Set the PV module temperature judgment threshold, use the predicted temperature of the PV module as the reference data, and the measured temperature of the PV module as the measured data. Calculate the absolute difference between the measured data and the reference data and compare it with the temperature judgment threshold. When the absolute difference exceeds the temperature judgment threshold, it is determined that the module state is abnormal, otherwise it is normal.

[0129] When the judgment result is that the temperature distribution is normal and the component state is abnormal, a jet cleaning command is generated;

[0130] When the judgment result is a hot spot fault and the component status is normal, a fault warning is generated;

[0131] When the judgment result is a hot spot failure and the component status is abnormal, a spray cleaning command is generated.

[0132] The cooling strategies include:

[0133] Set a temperature adjustment threshold. When the actual temperature data of the photovoltaic module is greater than the temperature adjustment threshold, set the injection pressure and start the liquid injection module according to the wind direction and wind speed.

[0134] When the cleaning operation is completed before the cooling operation is performed, the spray water collection operation is performed for recycling and reducing water resource loss.

[0135] It is understandable that dust accumulation on the surface of the photovoltaic panel increases the heat transfer resistance of the photovoltaic panel, affecting the heat dissipation of the photovoltaic panel. Therefore, in this embodiment, a cleaning strategy is executed before executing the photovoltaic component cooling strategy.

[0136] The liquid spraying module is used to perform cleaning and cooling operations by spraying liquid, and the liquid spraying module includes at least one module having the following functions:

[0137] Relay module for controlling injection action;

[0138] A water pump module for completing the jetting action;

[0139] Control module for controlling injection pressure.

[0140] At least one embodiment of the present disclosure also provides an intelligent temperature regulation method for a photovoltaic cell assembly, such as Figure 2 As shown, the following steps are included:

[0141] S1. Collect monitoring data: collect real-time power data and environmental monitoring data of the photovoltaic power generation system, and transmit the data to the data processing module through communication equipment. The power data includes current, voltage and power, and the environmental monitoring data includes the ambient temperature and humidity, irradiance, cloud observation data, wind speed and direction, and temperature of the photovoltaic components.

[0142] It should be noted that data collection uses sensors and other equipment to obtain various data in the power distribution system in real time. Electric energy data is collected through current sensors and voltage sensors, and environmental monitoring data is detected by temperature and humidity sensors, radiometers, and wind speed and direction meters or obtained through meteorological data. Among them, the temperature detection on the surface of photovoltaic modules is carried out by setting up an FBG temperature sensor array for multi-point measurement.

[0143] As an emerging optical passive device, FBG temperature sensor has advantages that traditional electrical sensors do not have, such as small size, corrosion resistance, low cost, easy networking, anti-electromagnetic interference, and strong radiation ability. In addition, multiple gratings with different wavelengths can be connected in series on a single optical fiber to achieve distributed measurement. In this embodiment, an FBG sensor array is arranged on the surface of the photovoltaic module to perform multi-point temperature measurement to obtain the distribution of the temperature field on the surface of the photovoltaic module.

[0144] S2. Data preprocessing: By saving the collected power data and environmental monitoring data and establishing a database, the data in the database are used to construct training samples for temperature prediction, such as Figure 3 As shown, the specific steps include:

[0145] S21. Select prediction influencing factors: Classify the data in the database, including independent variables X and dependent variables Y, perform correlation analysis on independent variables X and dependent variables Y, and select independent variables with non-zero correlation as prediction factors. The independent variable X includes the ambient temperature and humidity, irradiance, wind speed and direction of the working environment of the photovoltaic module, and the dependent variable Y includes the temperature of the photovoltaic module.

[0146] The specific selection process includes data conversion and correlation coefficient calculation:

[0147] k =( u - w ) / δ ;

[0148] In the formula, k represents the data after mean normalization. u represents the average value of the original data, w represents the normalized data, δ Represents the variance of the original data set.

[0149] r=cov (X,Y) / σxσy ;

[0150] In the formula, r represents the Person correlation coefficient, cov represents the covariance, σ Represents standard deviation.

[0151] S22. Solve the prior temperature of the photovoltaic module: Use the output characteristic curve and irradiance of the photovoltaic module to calculate the temperature of the photovoltaic module, which is used as the prior temperature input of the temperature prediction model. Other environmental factors are added to the temperature prediction model to correct the prior temperature. Figure 4 As shown, the following steps are included:

[0152] S221, constructing an input vector: constructing a characteristic vector representing the characteristic curve of the photovoltaic module through the collected electric energy data: Z=[(0, I sc ), (V m , I m ), (V oc ,0)];

[0153] Among them, (0, I sc ) short-circuit current point, (V m , I m ) is the maximum power point, (V oc , 0) is the open circuit voltage point.

[0154] S222. Calculate the Fréchet distance d F : For two characteristic curves L1 and L2 with lengths N1 and N2 respectively, let i ∈(0, N1], j ∈(0, N2] are two points of the two curves, P ( i )=( x i , y i )and q ( j )=( x j , y j ) is the coordinate of the point on the curve, and the Fréchet distance d is solved. F :

[0155] i ≠1, j ≠1;

[0156] In the formula, d[ P ( i ), q ( j )]for i , j The direct distance between two points is calculated as:

[0157] .

[0158] S223, modified Fréchet distance d F : Relative overshoot through the specified curve Modified Fréchet distance, calculated as follows:

[0159] ;

[0160] In the formula, is the modified Fréchet distance, is the penalty factor, is the relative overshoot, and the calculation formula is:

[0161] ;

[0162] In the formula, U P1 , U P2 They are the maximum peak values ​​of characteristic curve L1 and characteristic curve L2 respectively.

[0163] It should be noted that the corrected distance represents the distance measurement result that takes the relative overshoot into account. The penalty factor is a parameter used to adjust the influence of the relative overshoot on the distance calculation. When the penalty factor is greater than 1, the influence of the relative overshoot will increase; when the penalty factor is less than 1, the influence of the relative overshoot will decrease. The appropriate penalty factor can be selected according to specific needs and trade-offs.

[0164] The Frechet distance takes into account the position and order between the data points of two sets of data, but takes the distance between some matching points as part of the similarity result. It is a local matching measurement method. Therefore, in order to improve the accuracy and globality of the Frechet distance, in this embodiment, a correlation constraint is introduced before calculating the Frechet distance, that is, the global correlation of the characteristic curve of the photovoltaic module is ensured by stipulating the relative overshoot of the curve.

[0165] S224, solve the prior temperature: select a set of data under standard test conditions as a benchmark and solve the prior temperature T x :

[0166] ;

[0167] Where G is the irradiance, a , b is a constant, and the calculation formula is as follows:

[0168] ;

[0169] ;

[0170] Where, d F1 With d F2 , T1 and T2 are the Fréchet distance and temperature at different temperatures under the same irradiance; d F3 With d F4 , G3 and G4 are the Fréchet distance and irradiance at the same temperature but different irradiances.

[0171] The standard test conditions are: component temperature is 25°C, irradiance is 1000W / m 2, as a reference object for calculating the Fréchet distance.

[0172] It should be noted that the temperature of the PV module solved by the above method takes into account the operating status and irradiance of the PV module. In the subsequent temperature prediction, the operating status of the PV module is used as prior knowledge to ensure the validity of the model prediction results. At the same time, the prediction through the model is equivalent to correcting the solved prior temperature based on other environmental influencing factors. The mutual constraints between the two improve the accuracy of the reference temperature of the PV module.

[0173] S23, establishing training samples for temperature prediction: collecting sample data according to the screened prediction influencing factors and solving the prior temperature of the sample data to construct training samples for the temperature prediction model.

[0174] S3, PV module temperature prediction: Based on the established temperature prediction training samples, a temperature prediction model is established for training, and the collected data is input into the trained prediction model to obtain the temperature output, which specifically includes the following steps:

[0175] S31. Establish a temperature prediction model: use the selected prediction factor type data and the corresponding prior temperature as input variables, the PV module temperature as output, and the number of hidden layer nodes ,in I is the number of nodes in the input layer; O is the number of nodes in the output layer; c is a constant ranging from 1 to 10, and a neural network model is established.

[0176] S32, training temperature prediction model: normalize the input variables and set the error target to 10 -8 Train the neural network of photovoltaic module temperature prediction model.

[0177] S33. Input data to predict temperature: input the collected data into the trained prediction model according to the data type of the model input to obtain the output temperature.

[0178] In this embodiment, the photovoltaic module cleaning operation is completed by predicting the temperature of the photovoltaic module as a reference temperature and comparing it with the actual measured temperature measured at multiple points by the FBG temperature sensor array, thereby assisting subsequent temperature adjustment decisions.

[0179] S4, executing the cleaning strategy: setting the temperature difference threshold of the measuring point, obtaining the temperature data of the multi-point measurement of the FBG temperature sensor array, establishing the temperature field distribution, and determining that a hot spot fault occurs when a measuring point with a temperature difference greater than the temperature difference threshold appears in the temperature field, otherwise the temperature distribution is normal;

[0180] Set the PV module temperature judgment threshold, use the predicted temperature of the PV module as the reference data, and the measured temperature of the PV module as the measured data. Calculate the absolute difference between the measured data and the reference data and compare it with the temperature judgment threshold. When the absolute difference exceeds the temperature judgment threshold, it is determined that the module state is abnormal, otherwise it is normal.

[0181] When the judgment result is that the temperature distribution is normal and the component state is abnormal, a jet cleaning command is generated;

[0182] When the judgment result is a hot spot fault and the component status is normal, a fault warning is generated;

[0183] When the judgment result is a hot spot failure and the component status is abnormal, a spray cleaning command is generated.

[0184] S5. Execute the cooling strategy: set the temperature adjustment threshold. When the measured temperature data of the photovoltaic module is greater than the temperature adjustment threshold, set the injection pressure and start the liquid injection module according to the wind direction and wind speed.

[0185] It should be noted that photovoltaic modules are very sensitive to temperature changes. Dust accumulation on the surface of photovoltaic panels increases the thermal resistance of the panels, affecting the heat dissipation of the panels. As a result, under long-term sunlight exposure, the temperature of the part covered by dust accumulation rises much higher than other parts, and the power generation unit is transformed into a load resistor that does not generate electricity, forming a hot spot. Therefore, in this embodiment, the dust accumulation problem on the surface of the photovoltaic modules is taken into consideration during the temperature adjustment of the photovoltaic modules.

[0186] Before making a temperature adjustment decision, the PV modules are cleaned based on the temperature field distribution and temperature prediction of the PV modules, eliminating temperature changes caused by dust accumulation or failures. At the same time, the temperature is adjusted based on cleaning, and the initial cooling has been performed. By executing the cleaning strategy, blind cooling operations based only on temperature decisions are avoided, and the waste of water resources is avoided. At the same time, the corrosion of water and dust on the surface of the PV modules is reduced through the decision-making cleaning and cooling operations.

[0187] The present invention first eliminates the abnormal state of the photovoltaic module by comparing the predicted temperature of the photovoltaic module with the actual temperature, and then sprays and cools the photovoltaic module according to the actual temperature, making the temperature adjustment strategy more accurate and effective. At the same time, through the cleaning decision of the photovoltaic module, the problem of abnormal temperature change of the photovoltaic module is solved, and the power generation efficiency and safety of the photovoltaic module are improved.

[0188] In the temperature prediction process of photovoltaic modules, the present invention ensures the validity of model prediction results by adding a priori temperature and improves the accuracy of temperature prediction; at the same time, the difference in output characteristics of photovoltaic modules under two environmental conditions is evaluated by similarity, and a relationship equation with temperature and irradiance is established to calculate the priori temperature, taking into account the output characteristic curve of the photovoltaic module, so that the priori temperature is more in line with the target object, and then the Fréchet distance is corrected for correlation, so as to improve the accuracy of the Fréchet distance and the global correlation of the characteristic curve of the photovoltaic module.

[0189] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. An intelligent temperature regulating device for a photovoltaic cell assembly, characterized in that: It includes an injection control module, a liquid injection module and a data processing module; The injection control module is used to control the liquid injection module to complete the cleaning and cooling operations by executing the cleaning strategy and the cooling strategy; The cleaning strategy includes: Set the temperature difference threshold of the measuring point, obtain the temperature data of multi-point measurement of the temperature sensor array, and establish the temperature field distribution. When a measuring point with a temperature difference greater than the temperature difference threshold appears in the temperature field, it is judged as a hot spot failure, otherwise the temperature distribution is normal; Set the PV module temperature judgment threshold, use the predicted temperature of the PV module as the reference data, and the measured temperature of the PV module as the measured data. Calculate the absolute difference between the measured data and the reference data and compare it with the temperature judgment threshold. When the absolute difference exceeds the temperature judgment threshold, it is determined that the module state is abnormal, otherwise it is normal. When the judgment result is that the temperature distribution is normal and the component state is abnormal, a jet cleaning command is generated; When the judgment result is a hot spot fault and the component status is normal, a fault warning is generated; When the judgment result is a hot spot failure and an abnormal component state, a jet cleaning command is generated; The cooling strategies include: Set a temperature adjustment threshold. When the actual temperature data of the photovoltaic module is greater than the temperature adjustment threshold, set the injection pressure and start the liquid injection module according to the wind direction and wind speed. When the cleaning operation is completed before the cooling operation is performed, the spray water collection operation is performed for recycling; The data processing module is used to save the collected power data and environmental monitoring data and establish a database, and use the data in the database to construct training samples for temperature prediction; The processing flow of the data processing module includes the following steps: Select prediction influencing factors: classify the data in the database, including independent variables and dependent variables, conduct correlation analysis on the independent variables and dependent variables, and select independent variables with non-zero correlation as prediction factors; the independent variables include the ambient temperature and humidity, irradiance and wind speed and direction of the working environment of the photovoltaic modules, and the dependent variables include the temperature of the photovoltaic modules; Solve the prior temperature of the photovoltaic module: Use the output characteristic curve and irradiance of the photovoltaic module to calculate the temperature of the photovoltaic module as the prior temperature input of the temperature prediction model; Calculate the prior temperature The formula is: ； In the formula, is the modified Fréchet distance, G is the irradiance, a , b is a constant, and the calculation formula is as follows: ； ； In the formula, and , and is the Fréchet distance and temperature at different temperatures under the same irradiance; and , and is the Fréchet distance and irradiance at the same temperature and different irradiances; Modified Fréchet distance The calculation formula is as follows: ； In the formula, is the Fréchet distance, is the penalty factor, is the relative overshoot, The calculation formula is: ； In the formula, U P1 , U P2 are the maximum peak values ​​of characteristic curve L1 and characteristic curve L2 respectively; Establish training samples for temperature prediction: collect sample data based on the selected prediction influencing factors and solve the prior temperature of the sample data to construct training samples for the temperature prediction model; The characteristic curve L1 and the characteristic curve L2 are obtained by collecting real-time electric energy data of the photovoltaic power generation system.

2. The intelligent temperature regulating device for a photovoltaic cell assembly according to claim 1, characterized in that: The injection control module includes: A microcontroller to which is connected at least one module having the following functions: A data acquisition module for collecting monitoring data; A data processing module for storing monitoring data; A temperature prediction module for predicting the temperature of photovoltaic modules; Wireless transmission module for data transmission; Liquid jetting module for performing cleaning and cooling operations.

3. The intelligent temperature regulating device for a photovoltaic cell assembly according to claim 1, characterized in that: It also includes a data acquisition module for collecting real-time power data and environmental monitoring data of the photovoltaic power generation system, and transmitting the data to the data processing module through communication equipment. The power data includes current, voltage and power, and the environmental monitoring data includes the ambient temperature and humidity, irradiance, cloud observation data, wind speed and direction, and temperature of the photovoltaic components.

4. The intelligent temperature regulating device for a photovoltaic cell assembly according to claim 1, characterized in that: It also includes a temperature prediction module, which is used to predict the temperature of the photovoltaic module by constructing a temperature prediction model using the screened sample data of the prediction influencing factors and the prior temperature of the sample data; The temperature prediction process comprises the following steps: Establish a temperature prediction model: use the selected prediction factor type data and the corresponding prior temperature as input variables, the PV module temperature as output, and set the number of hidden layer nodes to establish a neural network model; Training the temperature prediction model: Normalize the input variables and set the error target to 10 -8 Train the neural network of photovoltaic module temperature prediction model; Input data to predict temperature: Input the collected data into the trained prediction model according to the data type of the model input to obtain the output temperature.

5. The intelligent temperature regulating device for a photovoltaic cell assembly according to claim 1, characterized in that: The liquid spraying module is used to perform cleaning and cooling operations by spraying liquid, and the liquid spraying module includes at least one module having the following functions: Relay module for controlling injection action; A water pump module for completing the jetting action; Control module for controlling injection pressure.

6. An intelligent temperature adjustment method for a photovoltaic cell assembly, applied to an intelligent temperature adjustment device for a photovoltaic cell assembly as claimed in any one of claims 1 to 5, characterized in that: The following steps are involved: Collect monitoring data: collect real-time power data and environmental monitoring data of the photovoltaic power generation system, and transmit the data to the data processing module through communication equipment; Data preprocessing: By saving the collected power data and environmental monitoring data and establishing a database, the data in the database is used to construct training samples for temperature prediction; PV module temperature prediction: Based on the established temperature prediction training samples, a temperature prediction model is established for training, and the collected data is input into the trained prediction model to obtain the temperature output; Execute cleaning strategy: control the liquid injection module to complete the cleaning operation by executing the cleaning strategy; Execute the cooling strategy: set the temperature adjustment threshold. When the measured temperature data of the photovoltaic module is greater than the temperature adjustment threshold, set the injection pressure and start the liquid injection module according to the wind direction and wind speed.

7. The intelligent temperature regulation method of a photovoltaic cell assembly according to claim 6, characterized in that: The data preprocessing comprises the following steps: Select prediction influencing factors: classify the data in the database, including independent variables X and dependent variables Y, conduct correlation analysis on independent variables X and dependent variables Y, and select independent variables with non-zero correlation as prediction factors; The specific selection process includes data conversion and correlation coefficient calculation: k =( u - w ) / δ ； In the formula, k represents the data after mean normalization. u represents the average value of the original data, w represents the normalized data, δ Represents the variance of the original data set; r=cov (X,Y) / σ x σ y ; In the formula, r represents the Person correlation coefficient, cov represents the covariance, σ represents standard deviation; Solve the prior temperature of the photovoltaic module: Use the output characteristic curve and irradiance of the photovoltaic module to calculate the temperature of the photovoltaic module, which is used as the prior temperature input of the temperature prediction model. Other environmental factors are added to the temperature prediction model to correct the prior temperature. Establish training samples for temperature prediction: collect sample data based on the screened prediction influencing factors and solve the prior temperature of the sample data to construct training samples for the temperature prediction model.

8. The intelligent temperature regulation method of a photovoltaic cell assembly according to claim 7, characterized in that: The method for solving the prior temperature of the photovoltaic module comprises the following steps: Construct input vector: Construct a feature vector representing the characteristic curve of the photovoltaic module through the collected power data: Z=[(0, I sc ), (V m , I m ), (V oc ,0)]; Among them, (0, I sc ) short-circuit current point, (V m , I m ) is the maximum power point, (V oc , 0) is the open circuit voltage point; Calculate the Fréchet distance d F : For two characteristic curves L1 and L2 with lengths N1 and N2 respectively, let i ∈(0,N1], j ∈(0, N2] are two points of the two curves, P ( i )=( x i , y i )and q ( j )=( x j , y j ) is the coordinate of the point on the curve, and the Fréchet distance d is solved. F : ； In the formula, d[ P ( i ), q ( j )]for i , j The direct distance between two points is calculated as: ； Modified Fréchet distance : Relative overshoot through the specified curve Modified Fréchet distance, get the modified Fréchet distance ; Solving the prior temperature: Select a set of data under standard test conditions as a benchmark and solve the prior temperature T x .

Images