A remote monitoring method and system for distributed photovoltaic power stations based on the Internet of Things

Abstract

The present invention discloses a remote monitoring method and system for distributed photovoltaic power stations based on the Internet of Things. This method relates to the technical field of photovoltaic power stations and aims to address the technical issues of existing distributed photovoltaic power stations, such as the difficulty in accurately monitoring dust coverage and timely detecting its impact on power generation efficiency. The method comprises the following steps: S1, collecting real-time voltage and current data from the photovoltaic panels of each photovoltaic module and solar radiation intensity data collected by each photovoltaic module, and calculating the photovoltaic conversion power of each photovoltaic module, the total amount of solar radiation collected by each photovoltaic module, and the power conversion efficiency of each photovoltaic module; S2, sorting out the connection modes of each component in the distributed photovoltaic power station, including series and parallel connections, and simultaneously obtaining power conversion efficiency data for each component to construct an overall power conversion efficiency analysis model for the power station; S3, based on historical data. The present invention has the advantage of accurately monitoring the dust coverage rate of distributed photovoltaic power stations.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic power plants, and more specifically, to a remote monitoring method and system for distributed photovoltaic power plants based on the Internet of Things. Background Art

[0002] In the field of distributed photovoltaic power plants, with the development of Internet of Things (IoT) technology, some monitoring methods have been applied. Existing technologies use simple sensors to collect basic data from photovoltaic modules, such as voltage and current, and transmit this data to a monitoring center via wired or wireless transmission, enabling preliminary monitoring of the plant's operating status. Some systems also perform simple calculations of the module's energy conversion efficiency to assess the plant's overall performance.

[0003] However, existing technologies for remotely monitoring dust coverage in distributed photovoltaic power plants have significant shortcomings. Dust accumulation on the surface of photovoltaic modules can severely impact their ability to collect solar radiation and convert it into electricity. However, existing monitoring methods struggle to accurately predict the dust coverage of PV modules. Traditional methods rely on regular manual inspections, which are inefficient and costly. Furthermore, they lack real-time dust coverage data, hindering timely detection of power generation efficiency declines caused by dust accumulation, impacting the overall power generation efficiency and operational stability of distributed photovoltaic power plants.

[0004] In view of this, we propose a remote monitoring method and system for distributed photovoltaic power stations based on the Internet of Things. Summary of the Invention

[0005] The purpose of the present invention is to provide a remote monitoring method and system for distributed photovoltaic power stations based on the Internet of Things, so as to solve the technical problems that the dust coverage rate of existing distributed photovoltaic power stations is difficult to accurately monitor and its impact on power generation efficiency cannot be discovered in a timely manner.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a distributed photovoltaic power station remote monitoring method and system based on the Internet of Things, comprising the following steps:

[0007] S1. Collect the real-time voltage and current data of the photovoltaic panels of each photovoltaic module and the solar radiation intensity data collected by each photovoltaic module, and calculate the photovoltaic conversion power of each photovoltaic module, the total amount of solar radiation collected by each photovoltaic module, and the power conversion efficiency of each photovoltaic module;

[0008] S2. Sort out the connection methods of each component in the distributed photovoltaic power station, including series and parallel connections. At the same time, obtain the power conversion efficiency data of each component and build an overall power conversion efficiency analysis model for the power station;

[0009] S3. Based on historical data, a deep learning algorithm is used to build a dust coverage analysis model. The calculated total solar radiation data, the power conversion efficiency data of each component, and the power conversion efficiency data of the entire power station obtained in step S2 are input into the trained and optimized dust coverage analysis model to obtain the dust coverage prediction value. ;

[0010] S4. Set dust coverage threshold , the predicted value of dust coverage on the surface of the photovoltaic module determined in step S3 With the set threshold Compare and determine, if the dust coverage rate prediction value exceeds the dust coverage rate threshold, an alarm signal is generated, if it does not exceed the dust coverage rate threshold, the current state of operation is maintained;

[0011] S5. Through model analysis, clarify the comprehensive impact of dust on the collection of solar radiation on the surface of photovoltaic modules and the electrical energy conversion efficiency of photovoltaic modules.

[0012] Preferably, in S1, the calculation formula for the photovoltaic conversion power of each photovoltaic module at a certain moment is:

[0013] ;

[0014] in, is the current at a certain moment, is the voltage value at a certain moment;

[0015] The calculation formula for the total amount of solar radiation collected by each photovoltaic module at a certain moment is:

[0016] ;

[0017] in, is the total number of hours in a day, is the sunlight intensity under standard test conditions, is the voltage value under standard test conditions, is the conversion efficiency, The value at a certain moment;

[0018] The calculation formula for the power conversion efficiency of each photovoltaic module is:

[0019] ;

[0020] in, is the proportionality coefficient, is the total amount of collected solar radiation.

[0021] Preferably, in said S2, a distributed photovoltaic power station is provided. Components, The power conversion efficiency of each component is ;

[0022] When the components are connected in series, the power conversion efficiency of the entire power station ;

[0023] When the components are connected in parallel, the power conversion efficiency of the entire power station ,in, For the The weight of each component is determined by judging the connection mode of the components and calculating with the above formula to obtain the power conversion efficiency of the entire power station.

[0024] Preferably, in S3, the historical data includes historical total solar radiation data, power conversion efficiency data of each component, power conversion efficiency data of the entire power station, and corresponding actual measured values ​​of dust coverage. The deep learning algorithm uses a long short-term memory network to construct a dust coverage analysis model. The specific construction process and algorithm formula are as follows:

[0025] The input layer receives historical total solar radiation data , power conversion efficiency data of each component , the power conversion efficiency data of the entire power station And the corresponding actual measurement value of dust coverage , these data are normalized and used as input vectors .

[0026] Preferably, in said S3, the calculated total solar radiation data after normalization is , power conversion efficiency data of each component The power conversion efficiency data of the entire power station obtained in step S2 and the normalized historical data together form a new input vector Input to the long short-term memory network;

[0027] The hidden layer state update formula of the long short-term memory network is:

[0028] ;

[0029] in, is the new input vector, is the hidden layer output, is the cell state, 、 、 are the outputs of the input gate, forget gate, and output gate respectively. is the Sigmoid function, is the hyperbolic tangent function, is the weight matrix, is the bias vector;

[0030] The output layer outputs the hidden layer through the fully connected layer Mapping to dust coverage predictions , the formula is:

[0031] ;

[0032] in, is the output layer weight matrix, is the output layer bias vector.

[0033] Preferably, in said S5, the influence coefficient of dust on the solar radiation collected by the photovoltaic module surface is obtained by model analysis. and the coefficient of influence on the power conversion efficiency of photovoltaic modules , the specific calculation method is as follows:

[0034] Assume that the amount of solar radiation collected by the photovoltaic module without dust is , when there is dust ,but ;

[0035] Assume that the power conversion efficiency of the photovoltaic module without dust is , when there is dust ,but ;

[0036] Through the analysis of a large amount of historical data and model calculation, we can determine and Predicted dust coverage The functional relationship between them is:

[0037] ;

[0038] in, 、 、 、 、 、 are the coefficients obtained by fitting the data.

[0039] A distributed photovoltaic power station remote monitoring system based on the Internet of Things, comprising:

[0040] The data acquisition module is used to collect the real-time voltage and current data of the photovoltaic panels of each photovoltaic module and the solar radiation intensity data collected by each photovoltaic module;

[0041] A data calculation module is connected to the data acquisition module and is used to calculate the photovoltaic conversion power of each photovoltaic module, the total amount of solar radiation collected by each photovoltaic module, and the power conversion efficiency of each photovoltaic module based on the collected data;

[0042] The connection mode and efficiency analysis module is used to sort out the connection modes of each component in the distributed photovoltaic power station, including series and parallel connections, obtain the power conversion efficiency data of each component, and build an overall power conversion efficiency analysis model for the power station;

[0043] The dust coverage rate prediction module is used to build a dust coverage rate analysis model based on historical data using a deep learning algorithm. It receives the data output by the data calculation module and the connection mode and efficiency analysis module and inputs it into the trained and optimized dust coverage rate analysis model to obtain the dust coverage rate prediction value.

[0044] The alarm module sets a dust coverage rate threshold, compares the dust coverage rate prediction value obtained by the dust coverage rate prediction module with the set threshold, and generates an alarm signal if the dust coverage rate prediction value exceeds the dust coverage rate threshold; otherwise, the system maintains the current state of operation;

[0045] The comprehensive impact analysis module is used to clarify the comprehensive impact of dust on the collection of solar radiation on the surface of photovoltaic modules and the power conversion efficiency of photovoltaic modules through model analysis.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. The present invention constructs a dust coverage analysis model and uses a deep learning algorithm to analyze historical data and real-time collected data to accurately predict the dust coverage on the surface of photovoltaic modules. This solves the problem of the inability to accurately obtain dust coverage data in the existing technology, and realizes real-time remote monitoring of dust coverage. It enables staff to timely understand the dust coverage situation of the power station and provide a strong basis for subsequent maintenance decisions.

[0048] 2. The present invention is based on accurate dust coverage prediction results. When the predicted value exceeds the set threshold, an alarm signal is issued in time. This allows staff to arrange cleaning and other maintenance work in a targeted manner, avoiding unnecessary regular inspections and excessive maintenance, thereby reducing operation and maintenance costs and improving the efficiency and accuracy of maintenance work.

[0049] 3. The present invention also clarifies the comprehensive impact of dust on the collection of solar radiation by the surface of photovoltaic modules and the efficiency of power conversion through model analysis. This helps staff to gain a deeper understanding of the impact mechanism of dust on the power generation performance of power stations, so as to take more effective measures, such as adjusting the module layout and optimizing the cleaning plan, to minimize the negative impact of dust on power generation efficiency and improve the overall power generation efficiency of distributed photovoltaic power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 Schematic diagram of the method flow of the present invention;

[0051] Figure 2 Schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION

[0052] To facilitate those skilled in the art to understand the technical solution of the present invention, the technical solution of the present invention is further described with reference to the accompanying drawings.

[0053] Example 1: The present invention provides a remote monitoring method for a distributed photovoltaic power station based on the Internet of Things, comprising the following steps:

[0054] S1. Collect the real-time voltage and current data of the photovoltaic panels of each photovoltaic module and the solar radiation intensity data collected by each photovoltaic module, and calculate the photovoltaic conversion power of each photovoltaic module, the total amount of solar radiation collected by each photovoltaic module, and the power conversion efficiency of each photovoltaic module;

[0055] S2. Sort out the connection methods of each component in the distributed photovoltaic power station, including series and parallel connections. At the same time, obtain the power conversion efficiency data of each component and build an overall power conversion efficiency analysis model for the power station;

[0056] S3. Based on historical data, a deep learning algorithm is used to build a dust coverage analysis model. The calculated total solar radiation data, the power conversion efficiency data of each component, and the power conversion efficiency data of the entire power station obtained in step S2 are input into the trained and optimized dust coverage analysis model to obtain the dust coverage prediction value. ;

[0057] S4. Set dust coverage threshold , the predicted value of dust coverage on the surface of the photovoltaic module determined in step S3 With the set threshold Compare and determine, if the dust coverage rate prediction value exceeds the dust coverage rate threshold, an alarm signal is generated, if it does not exceed the dust coverage rate threshold, the current state of operation is maintained;

[0058] S5. Through model analysis, clarify the comprehensive impact of dust on the collection of solar radiation on the surface of photovoltaic modules and the electrical energy conversion efficiency of photovoltaic modules.

[0059] In the embodiment of the present invention, in S1, the calculation formula for the photovoltaic conversion power of each photovoltaic module at a certain moment is:

[0060] ;

[0061] in, is the current at a certain moment, is the voltage value at a certain moment;

[0062] The calculation formula for the total amount of solar radiation collected by each photovoltaic module at a certain moment is:

[0063] ;

[0064] in, is the total number of hours in a day, is the sunlight intensity under standard test conditions, is the voltage value under standard test conditions, is the conversion efficiency, The value at a certain moment;

[0065] The calculation formula for the power conversion efficiency of each photovoltaic module is:

[0066] ;

[0067] in, is the proportionality coefficient, is the total amount of collected solar radiation.

[0068] In the embodiment of the present invention, in S2, it is assumed that the distributed photovoltaic power station has Components, The power conversion efficiency of each component is ;

[0069] When the components are connected in series, the power conversion efficiency of the entire power station ;

[0070] When the components are connected in parallel, the power conversion efficiency of the entire power station ,in, For the The weight of each component is determined by judging the connection mode of the components and calculating with the above formula to obtain the power conversion efficiency of the entire power station.

[0071] In an embodiment of the present invention, in S3, the historical data includes historical total solar radiation data, power conversion efficiency data of each component, power conversion efficiency data of the entire power station, and corresponding actual measured values ​​of dust coverage. The deep learning algorithm uses a long short-term memory network to construct a dust coverage analysis model. The specific construction process and algorithm formula are as follows:

[0072] The input layer receives historical total solar radiation data , power conversion efficiency data of each component , the power conversion efficiency data of the entire power station And the corresponding actual measurement value of dust coverage , these data are normalized and used as input vectors ;

[0073] The calculated total solar radiation data , power conversion efficiency data of each component , and the power conversion efficiency data of the entire power station (when connected in series , in parallel ) is normalized using the minimum-maximum normalization method, and the formula is:

[0074] ;

[0075] ;

[0076] ;

[0077] ;

[0078] in, and are the minimum and maximum values ​​of the total solar radiation in history, and They are the minimum and maximum values ​​of the historical power conversion efficiency data of each component, and are the minimum and maximum values ​​of the historical power conversion efficiency data when power stations are connected in series. and They are respectively the minimum and maximum values ​​in the historical data of power conversion efficiency when power stations are connected in parallel.

[0079] In the embodiment of the present invention, in S3, the calculated total solar radiation data after normalization is , power conversion efficiency data of each component The power conversion efficiency data of the entire power station obtained in step S2 and the normalized historical data together form a new input vector Input to the long short-term memory network;

[0080] The hidden layer state update formula of the long short-term memory network is:

[0081] ;

[0082] in, is the new input vector, is the hidden layer output, is the cell state, 、 、 are the outputs of the input gate, forget gate, and output gate respectively. is the Sigmoid function, is the hyperbolic tangent function, is the weight matrix, is the bias vector;

[0083] The output layer outputs the hidden layer through the fully connected layer Mapping to dust coverage predictions , the formula is:

[0084] ;

[0085] in, is the output layer weight matrix, is the output layer bias vector.

[0086] In the embodiment of the present invention, in S5, the influence coefficient of dust on the solar radiation collected by the photovoltaic module surface is obtained through model analysis. and the coefficient of influence on the power conversion efficiency of photovoltaic modules , the specific calculation method is as follows:

[0087] Assume that the amount of solar radiation collected by the photovoltaic module without dust is , when there is dust ,but ;

[0088] Assume that the power conversion efficiency of the photovoltaic module without dust is , when there is dust ,but ;

[0089] Through the analysis of a large amount of historical data and model calculation, we can determine and Predicted dust coverage The functional relationship between them is:

[0090] ;

[0091] in, 、 、 、 、 、 are the coefficients obtained by fitting the data.

[0092] Example 2: A distributed photovoltaic power station remote monitoring system based on the Internet of Things, comprising:

[0093] The data acquisition module is used to collect the real-time voltage and current data of the photovoltaic panels of each photovoltaic module and the solar radiation intensity data collected by each photovoltaic module;

[0094] A data calculation module is connected to the data acquisition module and is used to calculate the photovoltaic conversion power of each photovoltaic module, the total amount of solar radiation collected by each photovoltaic module, and the power conversion efficiency of each photovoltaic module based on the collected data;

[0095] The connection mode and efficiency analysis module is used to sort out the connection modes of each component in the distributed photovoltaic power station, including series and parallel connections, obtain the power conversion efficiency data of each component, and build an overall power conversion efficiency analysis model for the power station;

[0096] The dust coverage rate prediction module is used to build a dust coverage rate analysis model based on historical data using a deep learning algorithm. It receives the data output by the data calculation module and the connection mode and efficiency analysis module and inputs it into the trained and optimized dust coverage rate analysis model to obtain the dust coverage rate prediction value.

[0097] The alarm module sets a dust coverage rate threshold, compares the dust coverage rate prediction value obtained by the dust coverage rate prediction module with the set threshold, and generates an alarm signal if the dust coverage rate prediction value exceeds the dust coverage rate threshold; otherwise, the system maintains the current state of operation;

[0098] The comprehensive impact analysis module is used to clarify the comprehensive impact of dust on the collection of solar radiation on the surface of photovoltaic modules and the power conversion efficiency of photovoltaic modules through model analysis.

[0099] The embodiments disclosed in the present invention are preferred embodiments, but are not limited to them. Ordinary technicians in this field can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. As long as they do not deviate from the spirit of the present invention, they are all within the scope of protection of the present invention.

Claims

1. A remote monitoring method for distributed photovoltaic power stations based on the Internet of Things, characterized in that: The following steps are involved: S1. Collect the real-time voltage and current data of the photovoltaic panels of each photovoltaic module and the solar radiation intensity data collected by each photovoltaic module, and calculate the photovoltaic conversion power of each photovoltaic module, the total amount of solar radiation collected by each photovoltaic module, and the power conversion efficiency of each photovoltaic module; S2. Sort out the connection methods of each component in the distributed photovoltaic power station, including series and parallel connections. At the same time, obtain the power conversion efficiency data of each component and build an overall power conversion efficiency analysis model for the power station; S3. Based on historical data, a deep learning algorithm is used to build a dust coverage analysis model. The calculated total solar radiation data, the power conversion efficiency data of each component, and the power conversion efficiency data of the entire power station obtained in step S2 are input into the trained and optimized dust coverage analysis model to obtain the dust coverage prediction value. ; S4. Set dust coverage threshold , the predicted value of dust coverage on the surface of the photovoltaic module determined in step S3 With the set threshold Compare and determine, if the dust coverage rate prediction value exceeds the dust coverage rate threshold, an alarm signal is generated, if it does not exceed the dust coverage rate threshold, the current state of operation is maintained; S5. Through model analysis, clarify the comprehensive impact of dust on the collection of solar radiation on the surface of photovoltaic modules and on the power conversion efficiency of photovoltaic modules; In S1, the calculation formula for the photovoltaic conversion power of each photovoltaic module at a certain moment is: ； in, is the current at a certain moment, is the voltage value at a certain moment; The calculation formula for the total amount of solar radiation collected by each photovoltaic module at a certain moment is: ； in, is the total number of hours in a day, is the sunlight intensity under standard test conditions, is the voltage value under standard test conditions, is the conversion efficiency, The value at a certain moment; The calculation formula for the power conversion efficiency of each photovoltaic module is: ； in, is the proportionality coefficient, is the total amount of collected solar radiation.

2. A distributed photovoltaic power station remote monitoring method based on the Internet of Things according to claim 1, characterized in that: In S2, it is assumed that the distributed photovoltaic power station has Components, The power conversion efficiency of each component is ; When the components are connected in series, the power conversion efficiency of the entire power station ; When the components are connected in parallel, the power conversion efficiency of the entire power station ,in, For the The weight of each component is determined by judging the connection mode of the components and calculating with the above formula to obtain the power conversion efficiency of the entire power station.

3. The method for remote monitoring of a distributed photovoltaic power station based on the Internet of Things according to claim 2, characterized in that: In S3, the historical data includes historical total solar radiation data, power conversion efficiency data of each component, power conversion efficiency data of the entire power station, and the corresponding actual measured values ​​of dust coverage. The deep learning algorithm uses a long short-term memory network to construct a dust coverage analysis model. The specific construction process and algorithm formula are as follows: The input layer receives historical total solar radiation data , power conversion efficiency data of each component , the power conversion efficiency data of the entire power station And the corresponding actual measurement value of dust coverage , these data are normalized and used as input vectors .

4. A distributed photovoltaic power station remote monitoring method based on the Internet of Things according to claim 3, characterized in that: In S3, the calculated total solar radiation data after normalization is , power conversion efficiency data of each component The power conversion efficiency data of the entire power station obtained in step S2 and the normalized historical data together form a new input vector Input to the long short-term memory network; The hidden layer state update formula of the long short-term memory network is: ； in, is the new input vector, is the hidden layer output, is the cell state, 、 、 are the outputs of the input gate, forget gate, and output gate respectively. is the Sigmoid function, is the hyperbolic tangent function, is the weight matrix, is the bias vector; The output layer outputs the hidden layer through the fully connected layer Mapping to dust coverage predictions , the formula is: ； in, is the output layer weight matrix, is the output layer bias vector.

5. The method for remote monitoring of distributed photovoltaic power stations based on the Internet of Things according to claim 4, characterized in that: In said S5, the influence coefficient of dust on the solar radiation collected by the photovoltaic module surface is obtained through model analysis. and the coefficient of influence on the power conversion efficiency of photovoltaic modules , the specific calculation method is as follows: Assume that the amount of solar radiation collected by the photovoltaic module without dust is , when there is dust ,but ; Assume that the power conversion efficiency of the photovoltaic module without dust is , when there is dust ,but ; Through the analysis of a large amount of historical data and model calculation, we can determine and Predicted dust coverage The functional relationship between them is: ； in, 、 、 、 、 、 are the coefficients obtained by fitting the data.

6. A monitoring system applied to a remote monitoring method for a distributed photovoltaic power station based on the Internet of Things as claimed in any one of claims 1 to 5, characterized in that: include: The data acquisition module is used to collect the real-time voltage and current data of the photovoltaic panels of each photovoltaic module and the solar radiation intensity data collected by each photovoltaic module; A data calculation module is connected to the data acquisition module and is used to calculate the photovoltaic conversion power of each photovoltaic module, the total amount of solar radiation collected by each photovoltaic module, and the power conversion efficiency of each photovoltaic module based on the collected data; The connection mode and efficiency analysis module is used to sort out the connection modes of each component in the distributed photovoltaic power station, including series and parallel connections, obtain the power conversion efficiency data of each component, and build an overall power conversion efficiency analysis model for the power station; The dust coverage rate prediction module is used to build a dust coverage rate analysis model based on historical data using a deep learning algorithm. It receives the data output by the data calculation module and the connection mode and efficiency analysis module and inputs it into the trained and optimized dust coverage rate analysis model to obtain the dust coverage rate prediction value. The alarm module sets a dust coverage rate threshold, compares the dust coverage rate prediction value obtained by the dust coverage rate prediction module with the set threshold, and generates an alarm signal if the dust coverage rate prediction value exceeds the dust coverage rate threshold; otherwise, the system maintains the current state of operation; The comprehensive impact analysis module is used to clarify the comprehensive impact of dust on the collection of solar radiation on the surface of photovoltaic modules and the power conversion efficiency of photovoltaic modules through model analysis.

Images