Management and control system for distributed photovoltaic power generation

Abstract

The invention discloses a management and control system for distributed photovoltaic power generation. The management and control system comprises a plurality of combined early warning modules, a plurality of cleaning comparison modules, a data transmission subsystem, a composite secrecy subsystem, a plurality of sensor acquisition units, a cloud platform and a remote terminal, the cloud platform calculates a plurality of environmental parameters by using a BP neural network to obtain a cleaning threshold value and a fixed period; in any cleaning comparison module, a comparison board is used for simulating the same surface pollution condition of a corresponding photovoltaic module, a detection board is used for automatically cleaning in a fixed period, an MCU is used for obtaining a voltage comparison value and receiving the fixed period and a cleaning threshold value from the cloud platform, the detection board is automatically cleaned according to the fixed period, and when the voltage comparison value reaches the cleaning threshold value, the detection board is automatically cleaned according to the fixed period. And the MCU sends feedback information to the cloud platform. According to the invention, the BP neural network and the cleaning contrast module are combined, and the cleaning threshold and the fixed period can be adaptively adjusted based on the environmental parameters, so that when the photovoltaic module needs to be cleaned can be accurately known, and the photovoltaic module can generate power efficiently.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent energy management and control of photovoltaic power generation, and in particular to a management and control system for distributed photovoltaic power generation. Background Art

[0002] Photovoltaic power generation, as a clean and renewable energy form, has become an important development direction in the energy field. In recent years, with the advancement of science and technology and the reduction of costs, the technology of photovoltaic power generation using various photovoltaic modules has developed rapidly. In order to improve the efficiency and reliability of photovoltaic power generation, smart energy management and control technology for photovoltaic power generation has also emerged. By introducing advanced information technology and intelligent systems, smart photovoltaic power stations have achieved comprehensive monitoring, management and optimization of the photovoltaic power generation process, further improving the stability and economy of photovoltaic power stations.

[0003] As the core component of the photovoltaic power generation system, the cleanliness of the photovoltaic module surface directly affects the power generation efficiency. During the operation of the photovoltaic power station, impurities such as dust, resin, droplets and even bird droppings will inevitably adhere to the surface of the photovoltaic module. These impurities will significantly reduce the ability of the photovoltaic module surface to reflect and absorb sunlight, resulting in a decrease in power generation efficiency; in addition, long-term accumulation of dirt may also damage the photovoltaic module and shorten its service life. Therefore, the photovoltaic module needs to be cleaned in time to keep it working in the best condition to ensure power generation efficiency.

[0004] Distributed photovoltaic power generation is a widely used form of photovoltaic power generation. Multiple photovoltaic modules are often placed in different areas or different locations in the same area. The environmental parameters faced by each photovoltaic module are not necessarily the same, resulting in different current states and actual working efficiencies of the photovoltaic modules, making it difficult to manage them uniformly. It is difficult to supervise the equipment of distributed photovoltaic power stations, and it is easy to have poor supervision, resulting in low power generation efficiency and economic losses. In addition, photovoltaic modules have to be exposed to the sun for a long time in the open air, and combined with the complex environment, the possibility of safety hazards such as fire increases.

[0005] At present, one way to manage the photovoltaic modules used in distributed photovoltaic power generation is to inspect and clean them manually to ensure the luminous efficiency of the photovoltaic modules; another way is to use cleaning robots to clean them. As for the technologies for managing and controlling distributed photovoltaic power generation, some use LoRa technology to combine the information of photovoltaic modules with remote control, while others connect each photovoltaic module to the grid dispatching and control center for overall control. However, these measures currently adopted contain at least the following three problems: 1) Distributed photovoltaic power generation systems will divide a large number of photovoltaic modules into multiple areas for power generation. The environment they face is complex. Each photovoltaic module needs to be cleaned at different times, which makes it difficult to manage them uniformly. At the same time, if the photovoltaic modules are not cleaned in time, it will not only affect the power generation efficiency, but may also cause damage to the photovoltaic modules, resulting in low actual power generation efficiency. 2) When cleaning distributed photovoltaic modules, due to the large differences in the environment, there are too many and different environmental parameters that need to be paid attention to, making it difficult to implement targeted cleaning strategies taking into account the various situations in which the photovoltaic modules are covered with dust, rain, leaves and other debris; 3) When the management and control system is controlling a large number of photovoltaic panels, it often requires long-distance communication transmission and a large amount of data transmission, but lacks a sufficiently effective confidentiality protection mechanism, especially the key management of LoRa technology is weak and vulnerable to attacks. Summary of the invention

[0006] In response to the above three problems, the purpose of the present invention is to propose a management and control system for distributed photovoltaic power generation, which combines the BP neural network and the cleaning control module, and can adaptively adjust the cleaning threshold and fixed cleaning cycle based on environmental parameters, so as to accurately know when the photovoltaic components need to be cleaned, so that the photovoltaic components can generate electricity efficiently; at the same time, the early warning signal is optimized, and multiple early warning signals are used for reminders to prevent staff from ignoring the early warning signals due to environmental or site reasons; in addition, a hybrid encryption unit is added in the transmission to ensure the security of long-distance transmission communications.

[0007] This is achieved through the following technical solutions: A management and control system for distributed photovoltaic power generation is used for intelligent management and control of multiple photovoltaic components in the distributed photovoltaic power generation system. The management and control system includes multiple combined warning modules, multiple cleaning control modules, data transmission subsystems, composite confidentiality subsystems, multiple sensor acquisition units, cloud platforms and remote terminals. The data transmission subsystems are respectively connected to the composite confidentiality subsystems, the cloud platform, each sensor acquisition unit, each cleaning control module and each combined warning module. The cloud platform is connected to the remote terminal. Each photovoltaic component under intelligent management is correspondingly provided with a combined warning module, a sensor acquisition unit and a cleaning control module. Among them, any sensor acquisition unit is used to obtain multiple environmental parameters of the corresponding photovoltaic component, and transmit the multiple environmental parameters to the cloud platform via the data transmission subsystem. The cloud platform uses a BP neural network to calculate the multiple environmental parameters as an input layer to obtain the corresponding cleaning threshold and fixed cycle. Any cleaning control module includes: a control board, a detection board, an MCU, a protection circuit, a drive unit and a cleaning machine. The control board is used to simulate the same surface pollution of the corresponding photovoltaic component, and the control board converts the light signal of sunlight into the first The detection board is automatically cleaned at a fixed period by a cleaning machine, and the detection board converts the light signal of sunlight into a second signal and transmits it to the MCU via the protection circuit; the MCU is used to compare the first signal and the second signal to obtain a voltage comparison value; the MCU is also used to receive a fixed period and a cleaning threshold from the cloud platform, and use a driving unit to drive the cleaning machine to automatically clean the detection board according to a fixed period, and when the voltage comparison value reaches the cleaning threshold, the MCU sends feedback information to the cloud platform; the cloud platform controls the combined early warning module to issue an early warning based on the feedback information; the combined early warning module includes a two-color LED warning unit, a buzzer and a command early warning unit, which is used to send two-color light, sound and prompt information to the staff when an early warning is required; the composite confidentiality subsystem uses a hybrid encryption algorithm containing at least two algorithms to encrypt the data transmission subsystem; the remote terminal is used to set corresponding range intervals for the cleaning threshold and the fixed period respectively, and when the cleaning threshold or the fixed period calculated by the BP neural network in the cloud platform exceeds the range interval, the cloud platform recalculates the cleaning threshold or the fixed period until it meets the corresponding range interval.

[0008] Preferably, the multiple environmental parameters include light intensity, temperature, humidity, rainfall, dust index and wind force. Based on the multiple environmental parameters, a comprehensive calculation result can be obtained.

[0009] Preferably, the BP neural network adopts a 6-9-2 structure, including an input layer, a hidden layer and an output layer. The input layer uses 6 neurons for inputting light intensity, temperature, humidity, rainfall, dust index and wind force respectively, the hidden layer uses 9 neurons for calculation, and the output layer uses 2 neurons for outputting the cleaning threshold and fixed period. The BP neural network with a 6-9-2 structure can take into account both calculation accuracy and resource consumption.

[0010] Preferably, the protection circuit includes two groups of sub-circuits with the same structure, each of which includes a parallel resistor and two groups of parallel and identical RC filter circuits, the parallel resistor is composed of at least two identical resistors connected in parallel, and is used to convert the first signal of the control board or the second signal of the detection board into a voltage form; the two groups of parallel and identical RC filter circuits are used to suppress the voltage fluctuation of the first signal or the second signal during transmission. The use of the protection circuit can effectively ensure the transmission of the signal.

[0011] Preferably, the dual-color LED warning unit includes a LED subunit 1 and an LED subunit 2 connected in series, wherein the LED subunit 1 is composed of a plurality of blue LED tubes connected in series in parallel, and the LED subunit 2 is composed of a plurality of red LED tubes connected in series in parallel. The use of red and blue dual-color tubes for warning is eye-catching and easy for staff to find, so that they can process the photovoltaic components that need to be processed.

[0012] Preferably, the remote terminal is provided with a storage module of a double-layer architecture, including a first storage layer and a second storage layer, the capacity of the first storage layer is smaller than the capacity of the second storage layer, the read and write speed of the first storage layer is greater than the read and write speed of the second storage layer, the first storage layer is used to store current data in the cloud platform, and the second storage layer is used to store historical data in the cloud platform. Setting up a storage module of a double-layer architecture facilitates classification and processing of large amounts of data, and is convenient for use by remote personnel.

[0013] Preferably, the composite confidentiality subsystem uses an AES encryption subunit and an RSA encryption subunit. The AES encryption subunit is used to perform AES encryption on the data transmitted in the data transmission subsystem, and the RSA encryption submodule is used to perform secondary encryption and management on the key corresponding to the AES encryption. The use of hybrid encryption can effectively improve the security of transmitted data.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The technical solution of the present invention combines the BP neural network and the cleaning control module, which can adaptively adjust the cleaning threshold and the fixed cleaning cycle based on environmental parameters, so as to accurately know when the photovoltaic components need to be cleaned, thereby enabling the photovoltaic components to generate electricity efficiently; a double-layer storage module is also used in the remote terminal to facilitate remote personnel to monitor and analyze; at the same time, the early warning signal is optimized, and multiple early warning signals are used for reminders to avoid staff ignoring the early warning signal due to environmental or site reasons; in addition, a hybrid encryption unit is added in the transmission to ensure the security of long-distance transmission communications. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 The present invention is a structural schematic diagram of a remote terminal controlling a photovoltaic module in a control system for distributed photovoltaic power generation. DETAILED DESCRIPTION

[0016] The technical solutions in the embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0017] like Figure 1 As shown, it is a structural schematic diagram of a remote terminal controlling a photovoltaic component in a management and control system for distributed photovoltaic power generation. The management and control system can intelligently control each photovoltaic component in the distributed photovoltaic power generation system. The management and control system includes multiple combined early warning modules, multiple cleaning control modules, a data transmission subsystem, a composite confidentiality subsystem, multiple sensor acquisition units, a cloud platform and a remote terminal. The data transmission subsystem is respectively connected to the composite confidentiality subsystem, the cloud platform, each sensor acquisition unit, each cleaning control module and each combined early warning module. The cloud platform is connected to the remote terminal to receive monitoring from remote personnel; each photovoltaic component under intelligent control is correspondingly provided with a combined early warning module, a sensor acquisition unit and a cleaning control module.

[0018] In this embodiment, each cleaning control module has the same structure, including: a control board, a detection board, an MCU, a protection circuit, a driving unit and a cleaning machine, and the control board and the detection board are the same photovoltaic panels. The control board is placed on the surface of the corresponding photovoltaic module to simulate the same surface pollution of the corresponding photovoltaic module. The control board converts the light signal of sunlight into a first signal and transmits it to the MCU via the protection circuit. The detection board uses a cleaning machine to automatically clean it at a fixed cycle to regularly ensure its own cleanliness. The detection board converts the light signal of sunlight into a second signal and transmits it to the MCU via the protection circuit. The MCU is used to compare the first signal and the second signal to obtain a voltage comparison value, which reflects the surface pollution of the control board compared to the detection board in real time, and also reflects the surface pollution of the photovoltaic module compared to the detection board; the MCU is also used to receive a fixed cycle and a cleaning threshold from the cloud platform, and use the driving unit to drive the cleaning machine to automatically clean the detection board according to a fixed cycle to keep the detection board clean, and when the voltage comparison value reaches the cleaning threshold, the MCU sends feedback information to the cloud platform. The cloud platform controls the combined early warning module to issue an early warning based on the feedback information.

[0019] In this embodiment, the combined warning module includes a two-color LED warning unit, a buzzer and a command warning unit, which are used to send two-color light, sound and prompt information to the staff at the same time when a warning is needed. The two-color LED warning unit includes a LED subunit 1 and an LED subunit 2 connected in series, wherein the LED subunit 1 is composed of a plurality of blue LED tubes connected in series and in parallel, and the LED subunit 2 is composed of a plurality of red LED tubes connected in series and in parallel. The use of red and blue two-color tubes for warning is eye-catching and easy for the staff to find, so as to process the photovoltaic components that need to be processed. In actual scenarios, the staff at the photovoltaic power generation site are often equipped with corresponding working equipment, that is, the command warning unit, which can be a smart watch or a mobile phone and other equipment capable of receiving information, and can receive signals from a remote terminal or a cloud platform. Therefore, in order to avoid the situation that the staff is not near the photovoltaic component or is in a nearby building with blocked vision and hearing during the warning, so that the two-color light cannot be seen and the buzzer cannot be heard, the remote terminal or the cloud platform will also send the corresponding information that the photovoltaic component needs to be cleaned to the smart watch or mobile phone, thereby ensuring the effect of the warning reminder.

[0020] In this embodiment, the remote terminal is a terminal for remote staff to supervise the cloud platform, and can also manually set corresponding range intervals for the cleaning threshold and fixed period. When the cleaning threshold or fixed period calculated by the BP neural network in the cloud platform exceeds the range interval, the cloud platform recalculates the cleaning threshold or fixed period until it meets the corresponding range interval. The manually set range interval can avoid serious errors in the calculation of the cloud platform.

[0021] It should be noted that, when each photovoltaic module is actually in operation, there will be corresponding staff to perform irregular or regular manual cleaning to simply alleviate the surface pollution of each photovoltaic module as a whole. When the control board is set on the corresponding photovoltaic module, it is subject to the same manual cleaning of the photovoltaic module, thereby simulating the same surface pollution. However, the number of photovoltaic modules is large and they are easily affected by different environments, resulting in different surface pollution conditions. In particular, distributed photovoltaic modules of distributed photovoltaic power stations are more likely to have such problems. Some photovoltaic modules can barely guarantee their own operating efficiency according to the above irregular or regular manual cleaning, but the surface of other photovoltaic modules may be quickly contaminated again after irregular or regular manual cleaning. Therefore, it is necessary to clean the detection board regularly, and then obtain the real-time voltage comparison value of the detection board and the control board to judge the surface condition of the corresponding photovoltaic module. The more serious the actual surface pollution, the greater the comparison value, until the set threshold, that is, the cleaning threshold, is reached, an early warning can be issued, so that the staff can carry out targeted cleaning to ensure the operating efficiency of each photovoltaic module as much as possible.

[0022] In this embodiment, for any sensor acquisition unit, multiple sensors are set to obtain multiple environmental parameters of the corresponding photovoltaic module, and the multiple environmental parameters are transmitted to the cloud platform via the data transmission subsystem. When obtaining multiple environmental parameters, the light intensity can be obtained by the light sensor, the temperature can be obtained by the temperature sensor, the humidity can be obtained by the humidity sensor, the rainfall can be obtained by the rainfall sensor, the dust index can be obtained by the dust sensor, and the wind speed can be obtained by the anemometer. These environmental parameters are important factors affecting the working efficiency of the photovoltaic module. Based on multiple environmental parameters, comprehensive calculation results can be obtained. After obtaining multiple environmental parameters, the cloud platform uses the BP neural network to calculate the multiple environmental parameters as the input layer, and then obtains the cleaning threshold of the corresponding photovoltaic module and the fixed period of the corresponding detection board.

[0023] In this embodiment, the BP neural network in the cloud platform can adopt a 6-9-2 structure, including an input layer, a hidden layer and an output layer. The input layer uses 6 neurons to input light intensity, temperature, humidity, rainfall, dust index and wind force respectively, the hidden layer uses 9 neurons for calculation, and the output layer uses 2 neurons for outputting cleaning thresholds and fixed cycles. If the number of neurons in the hidden layer is too small, the calculation accuracy will be poor, and if the number is too large, it will lead to excessive resource consumption. Therefore, limiting the number of neurons in the hidden layer to 1.5 times the number of neurons in the input layer can take into account both calculation accuracy and resource consumption.

[0024] In this embodiment, the protection circuit includes 2 groups of sub-circuits with the same structure, each of which includes 1 parallel resistor and 2 groups of parallel and identical RC filter circuits, and the parallel resistor is composed of at least 2 identical resistors in parallel, which is used to convert the first signal of the control board or the second signal of the detection board into a voltage form; the 2 groups of parallel and identical RC filter circuits are used to suppress the voltage fluctuations that may be generated during the transmission of the first signal or the second signal, so as to maintain the stability and accuracy of the transmission. Taking the control board connected to the MCU through the corresponding sub-circuit as an example, the control board is connected to the MCU through the positive transmission line and the negative transmission line respectively, and the two RC filter circuits each include a resistor and a capacitor connected in series, and the two RC filter circuits are connected in parallel between the two transmission lines, and a parallel resistor is also connected in series on the positive transmission line, which is used to convert the first signal in the current form output by the control board into a voltage form, so that the MCU can receive and process it, and the resistor is composed of 2 identical resistors in parallel.

[0025] In this embodiment, the remote terminal is provided with a storage module with a double-layer architecture, including a first storage layer and a second storage layer, the capacity of the first storage layer is smaller than the capacity of the second storage layer, and the read and write speed of the first storage layer is greater than the read and write speed of the second storage layer; the first storage layer is used to store the current data in the cloud platform, which is small in amount and has a high probability of being viewed, so it requires a small capacity and a fast read and write speed; the second storage layer is used to store the historical data in the cloud platform, which is large in amount and has a low probability of being viewed currently, so it requires a large capacity and a slow read and write speed, which can save costs. The storage module with a double-layer architecture is convenient for classifying and processing a large amount of data, and is convenient for remote personnel to use.

[0026] In this embodiment, the composite confidentiality subsystem uses a hybrid encryption algorithm containing at least two algorithms to encrypt the data transmission subsystem, for example, an AES encryption subunit and an RSA encryption subunit are used, the AES encryption subunit is used to perform AES encryption on the data transmitted in the data transmission subsystem, and the RSA encryption submodule is used to perform secondary encryption and management on the key corresponding to the AES encryption. Hybrid encryption can effectively ensure data transmission security. When data is exchanged, the AES encryption algorithm is first used to encrypt the transmitted data, and then the RSA encryption algorithm is used to encrypt and manage the key in the AES encryption algorithm.

[0027] In summary, the present invention combines the BP neural network and the cleaning control module, and can adaptively adjust the cleaning threshold and the fixed cleaning cycle based on environmental parameters, so as to accurately know when the photovoltaic components need to be cleaned, thereby enabling the photovoltaic components to generate electricity efficiently; a double-layer storage module is also used in the remote terminal to facilitate remote personnel to monitor and analyze; at the same time, the early warning signal is optimized, and multiple early warning signals are used for reminders to avoid staff ignoring the early warning signal due to environmental or site reasons; in addition, a hybrid encryption unit is added in the transmission to ensure the security of long-distance transmission communications, which is significantly advanced.

[0028] The above embodiments are only for illustrating the technical idea of ​​the present invention, and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the present invention.

Claims

1. A management and control system for distributed photovoltaic power generation, used for intelligent management and control of multiple photovoltaic modules in a distributed photovoltaic power generation system, characterized in that: The control system includes multiple combined warning modules, multiple cleaning control modules, a data transmission subsystem, a composite confidentiality subsystem, multiple sensor acquisition units, a cloud platform and a remote terminal. The data transmission subsystem is respectively connected to the composite confidentiality subsystem, the cloud platform, each sensor acquisition unit, each cleaning control module and each combined warning module, and the cloud platform is connected to the remote terminal; each photovoltaic component under intelligent control is correspondingly provided with a combined warning module, a sensor acquisition unit and a cleaning control module; Among them, any sensor acquisition unit is used to obtain multiple environmental parameters of the corresponding photovoltaic module, and transmit the multiple environmental parameters to the cloud platform via the data transmission subsystem; the cloud platform uses the BP neural network to calculate the multiple environmental parameters as the input layer to obtain the corresponding cleaning threshold and fixed period; Any cleaning control module includes: a control board, a detection board, an MCU, a protection circuit, a driving unit and a cleaning machine; the control board is used to simulate the same surface contamination of the corresponding photovoltaic module, and the control board converts the optical signal of sunlight into a first signal and transmits it to the MCU via the protection circuit; the detection board is automatically cleaned at a fixed cycle by the cleaning machine, and the detection board converts the optical signal of sunlight into a second signal and transmits it to the MCU via the protection circuit; the MCU is used to compare the first signal and the second signal to obtain a voltage comparison value; the MCU is also used to receive a fixed cycle and a cleaning threshold from the cloud platform, and use the driving unit to drive the cleaning machine to automatically clean the detection board according to a fixed cycle, and And when the voltage comparison value reaches the cleaning threshold, the MCU sends feedback information to the cloud platform; the cloud platform controls the combined warning module to issue a warning based on the feedback information; the combined warning module includes a two-color LED warning unit, a buzzer and a command warning unit, which are used to emit two-color light, sound and prompt information to the staff when a warning is needed; the composite confidentiality subsystem uses a hybrid encryption algorithm containing at least two algorithms to encrypt the data transmission subsystem; the remote terminal is used to set corresponding range intervals for the cleaning threshold and fixed period respectively. When the cleaning threshold or fixed period calculated by the BP neural network in the cloud platform exceeds the range interval, the cloud platform recalculates the cleaning threshold or fixed period until it meets the corresponding range interval.

2. A management and control system for distributed photovoltaic power generation according to claim 1, characterized in that: Multiple environmental parameters include light intensity, temperature, humidity, rainfall, dust index and wind speed.

3. A management and control system for distributed photovoltaic power generation according to claim 2, characterized in that: The BP neural network adopts a 6-9-2 structure, including an input layer, a hidden layer and an output layer. The input layer uses 6 neurons to input light intensity, temperature, humidity, rainfall, dust index and wind speed respectively, the hidden layer uses 9 neurons for calculation, and the output layer uses 2 neurons to output the cleaning threshold and fixed period.

4. A management and control system for distributed photovoltaic power generation according to claim 1, characterized in that: The protection circuit includes two groups of sub-circuits with the same structure, each of which includes a parallel resistor and two groups of parallel and identical RC filter circuits. The parallel resistor is composed of at least two identical resistors in parallel, and is used to convert the first signal of the control board or the second signal of the detection board into a voltage form; the two groups of parallel and identical RC filter circuits are used to suppress voltage fluctuations of the first signal or the second signal during transmission.

5. A management and control system for distributed photovoltaic power generation according to claim 1, characterized in that: The dual-color LED warning unit comprises a LED sub-unit 1 and an LED sub-unit 2 connected in series. The LED sub-unit 1 is composed of a plurality of blue LED tubes connected in parallel, and the LED sub-unit 2 is composed of a plurality of red LED tubes connected in parallel.

6. A management and control system for distributed photovoltaic power generation according to claim 1, characterized in that: The remote terminal is provided with a storage module with a double-layer architecture, including a first storage layer and a second storage layer. The capacity of the first storage layer is smaller than that of the second storage layer, and the read and write speed of the first storage layer is greater than the read and write speed of the second storage layer. The first storage layer is used to store current data in the cloud platform, and the second storage layer is used to store historical data in the cloud platform.

7. A management and control system for distributed photovoltaic power generation according to claim 1, characterized in that: The composite confidentiality subsystem adopts AES encryption subunit and RSA encryption subunit. The AES encryption subunit is used to perform AES encryption on the data transmitted in the data transmission subsystem, and the RSA encryption submodule is used to perform secondary encryption and management on the key corresponding to the AES encryption.

Images