Electrical service system and method for photovoltaic large base

Abstract

The invention discloses an electrical service system and method for a photovoltaic large base, and relates to the technical field of photovoltaic power systems, and the method comprises the steps: comprehensively collecting the equipment operation data and environment data of the photovoltaic large base through an equipment data collection module, and improving the data quality through a data preprocessing module; by means of the state monitoring and evaluation module, the operation state of the equipment is accurately mastered. Energy distribution is optimized through the energy management and scheduling module, the power generation efficiency is improved, and the light abandoning rate is reduced. The fault diagnosis and early warning module can timely discover potential faults of equipment and give out early warning, and the intelligent maintenance scheduling module realizes optimal configuration of maintenance resources and improves the maintenance efficiency. Omnibearing and intelligent management of photovoltaic large-base electrical services is achieved, the operation efficiency and economic benefits of a photovoltaic large base are effectively improved, and the photovoltaic large-base electrical service management system has important engineering application value.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of photovoltaic power systems, in particular to an electrical service system and method for a photovoltaic large base. BACKGROUND

[0002] In the current construction and operation of photovoltaic large bases, there are many problems to be solved. On the one hand, photovoltaic large bases usually have a wide area and the equipment is extremely dispersed, covering a large number of photovoltaic components, inverters, combiner boxes, transformers and other electrical equipment. Traditional manual inspection and monitoring methods cannot achieve real-time and accurate monitoring of all equipment, resulting in delayed discovery of equipment failures and serious impact on power generation efficiency. On the other hand, the power generation of photovoltaic large bases is greatly affected by natural environmental factors such as light and temperature, and the energy scheduling and management is difficult. The existing technology lacks an effective energy optimization distribution mechanism, resulting in energy waste. At the same time, the existing photovoltaic electrical service system has deficiencies in the stability and security of data transmission and the intelligent level of equipment maintenance, with a high risk of data leakage and low maintenance and scheduling efficiency, which cannot meet the needs of large-scale and efficient operation of photovoltaic large bases. Therefore, there is an urgent need for a photovoltaic large base electrical service system and method that can achieve all-around monitoring, intelligent management, efficient maintenance and data security. SUMMARY

[0003] The purpose of the present application is to provide a photovoltaic large base electrical service system and method that can achieve all-around monitoring, intelligent management, and improve the operation efficiency and economic benefits of photovoltaic large bases.

[0004] The present application is implemented by using the following technical solutions.

[0005] In a first aspect, the present application provides an electrical service method for a photovoltaic large base, comprising: collecting operation data and environmental data of electrical equipment of the photovoltaic large base; preprocessing the collected operation data and environmental data to obtain structured data; based on the structured data, performing quantitative evaluation based on a pre-constructed equipment state evaluation model to obtain an equipment state evaluation report; based on the state evaluation report and the operation data, generating an energy scheduling strategy through a first optimization algorithm; receiving state monitoring data, analyzing based on a pre-constructed fault diagnosis knowledge base and fault diagnosis model according to the state monitoring data and the state evaluation report, and outputting fault warning information; based on the fault warning information, outputting a maintenance plan through a second optimization algorithm.

[0006] Optionally, the preprocessing is to process the collected operation data and environmental data, first data cleaning, removing outliers and correcting; then data denoising, eliminating noise interference; then data format conversion and missing value filling; finally data standardization processing, generating structured data for subsequent analysis.

[0007] Optionally, the pre-constructed equipment state evaluation model is a data input model that combines convolutional neural network and long short-term memory network, the pre-processed structured data is input into the equipment state evaluation model, the equipment state features library and the equipment state evaluation model are combined to monitor and quantitatively evaluate the running state of the equipment in real time, calculate the health index of the equipment, and generate an equipment state evaluation report.

[0008] Optionally, the first optimization algorithm is a particle swarm optimization algorithm, according to the equipment state evaluation report, real-time power generation data and power grid load demand, an energy optimization scheduling model is called, the energy optimization scheduling model takes maximizing the power generation of the photovoltaic base, minimizing the light rejection rate and realizing stable grid connection with the power grid as the goal, considers the power generation characteristics of the photovoltaic components, the charge and discharge characteristics of the energy storage equipment and the constraint conditions of the scheduling requirements of the power grid, adopts the particle swarm optimization algorithm to solve the optimal energy scheduling strategy, generates energy scheduling instructions and sends them to the corresponding execution equipment, realizes the optimization of energy allocation and scheduling.

[0009] Optionally, the training of the pre-constructed fault diagnosis knowledge base and fault diagnosis model is based on device historical fault data and real-time operation data, the evaluation results of the state monitoring and evaluation module are received in real time, when the equipment health index is lower than the set threshold or abnormal fluctuation occurs, the related data is input into the fault diagnosis model, the fault type is diagnosed and the fault occurrence probability is predicted, and the corresponding level of warning information is issued according to the fault severity Optionally, the second optimization algorithm is a genetic algorithm, according to the fault warning information, the importance of the equipment and the distribution of the maintenance resources, a maintenance scheduling optimization model is constructed, the genetic algorithm is adopted to formulate the optimal maintenance plan, the dynamic scheduling of maintenance personnel, vehicles and tools is realized through the GIS geographic information system, and the maintenance plan is generated.

[0010] Optionally, it further comprises: establishing an equipment state feature library; at least one of the operation data, the environmental data, the structured data, the equipment state evaluation report, the energy adjustment strategy, the fault warning information and the maintenance plan is stored in an encrypted database, and the equipment state feature library and the fault diagnosis knowledge base are updated; The monitoring system runs state, real-time monitoring system module running state, communication link state and device acquisition state; data security module regularly carries out data security detection and vulnerability scanning, records and alarms abnormal situation in time to the discovered abnormal situation and security vulnerability.

[0011] In a second aspect, the present application provides an electrical service system for a large photovoltaic base, characterized in that it comprises: a device data acquisition module, a data preprocessing module, a state monitoring and evaluation module, an energy management and scheduling module, a fault diagnosis and early warning module, an intelligent maintenance scheduling module, a communication transmission module, a user interaction module, a system management module and a data security module; The device data acquisition module is used for collecting operation data and environmental data of electrical equipment in the large photovoltaic base. The data preprocessing module is used for processing the operation data and environmental data to obtain preprocessed data. The state monitoring and evaluation module performs real-time monitoring and comprehensive evaluation on the electrical equipment according to the preprocessed data. The energy management and scheduling module formulates energy scheduling strategies according to the device state evaluation report and operation data. The fault diagnosis and early warning module analyzes and outputs fault early warning information based on a fault diagnosis knowledge base and a fault diagnosis model. The intelligent maintenance scheduling module obtains a maintenance plan according to the fault early warning information. The communication transmission module is used for realizing data interaction and transmitting data to the monitoring center. The user interaction module provides a visual operation interface to facilitate system operation and information viewing by staff. The system management module is responsible for user management, parameter configuration and log recording of the system. The data security module ensures the security and integrity of system data in the process of collection, transmission, storage and use. Advantages

[0012] This invention comprehensively collects equipment operation and environmental data from large-scale photovoltaic (PV) bases through an equipment data acquisition module, improves data quality through a data preprocessing module, and achieves precise control over equipment operation status through a status monitoring and evaluation module. The energy management and scheduling module optimizes energy allocation, improving power generation efficiency and reducing curtailment rates. The fault diagnosis and early warning module promptly detects potential equipment faults and issues warnings, while the intelligent maintenance scheduling module optimizes the allocation of maintenance resources, improving maintenance efficiency. The communication transmission module ensures stable and reliable data transmission, the user interaction module provides an intuitive and convenient operating interface, and the system management and data security modules ensure stable system operation and data security. By comprehensively collecting and analyzing equipment operation and environmental data from large-scale PV bases, this invention achieves precise monitoring of equipment operation status, optimized energy scheduling, timely fault diagnosis and early warning, and intelligent scheduling of maintenance resources. This improves the power generation efficiency and operational management level of large-scale PV bases, providing strong support for their safe and stable operation and possessing significant engineering application value. Attached Figure Description

[0013] Figure 1 The diagram shows a flowchart of the electrical service method for large-scale photovoltaic bases. Figure 2 The diagram shows an electrical service system for a large-scale photovoltaic base. Figure 3 The diagram shown is a schematic diagram of the device data acquisition module. Figure 4 The diagram shown is a schematic diagram of the condition monitoring and assessment module. Detailed Implementation

[0014] The present invention will be further described below with reference to specific embodiments. Example 1

[0015] like Figure 1 As shown in the figure, this embodiment introduces an electrical service method for large-scale photovoltaic bases, including: The administrator logs into the system management module to enter information such as the model, location, and parameters of photovoltaic modules, inverters, combiner boxes, and other equipment within the photovoltaic base; performs accuracy calibration and confirms the installation location of all sensors; tests the connectivity and transmission rate of fiber optic and 5G communication links; sets the normal threshold for the health index in the status assessment model to 70 points, and sets the fault warning threshold according to the fault type; initially configures the objective function weights of the energy dispatch strategy, creates administrator, operator, and maintenance personnel accounts and assigns corresponding permissions; and updates basic parameters such as energy storage device capacity and charging / discharging efficiency, as well as grid load constraints, in the energy management and dispatch module.

[0016] In the daily operation of the photovoltaic large base, the equipment data acquisition module collects the voltage, current and power of the photovoltaic module every 5 minutes through the sensors deployed on each equipment, the output voltage, output current and conversion efficiency of the inverter, and collects the environmental data such as illumination intensity, environmental temperature, wind speed, wind direction and precipitation every 10 minutes. The collected data is transmitted in real time to the data preprocessing module through optical fiber and LoRa wireless communication.

[0017] After the data preprocessing module receives the original data, it first uses the Relyda criterion to detect outliers, and corrects the data with a deviation of more than 3 times the standard deviation as an outlier according to the historical data trend; then uses the wavelet transform denoising algorithm to denoise the voltage and current data containing noise; converts the heterogeneous data collected by the sensor in XML format, CSV format, etc. into unified JSON format structured data; for missing data caused by communication interruption, linear interpolation method is used to fill in according to the data of the previous and next time; finally, the min-max standardization method is used to map all data to the [0, 1] interval, and generate the preprocessed data set.

[0018] The state monitoring and evaluation module inputs the preprocessed data set into the equipment state evaluation model fused with CNN and LSTM. The CNN layer extracts the correlation features between different equipment parameters, and the LSTM layer analyzes the trend of the change of the equipment parameters with time. The model calculates the health index of each equipment. The health index of a certain inverter is calculated as 65 points, which is lower than the normal threshold of 70 points. Combined with the individual state feature library of the inverter and the average health index of 75 points of the same type of inverter group, it is judged that the operation state of the inverter is abnormal, and the equipment state evaluation report containing the health index, abnormal parameters and comparative analysis is generated.

[0019] The energy management and scheduling module receives the equipment state evaluation report, combines the real-time monitored total power generation of 5000kW and the power grid load demand of 4500kW, and calls the energy optimization scheduling model. The model considers the current remaining capacity of the energy storage equipment as 30%, the charging and discharging efficiency as 90% and other constraint conditions, and uses the particle swarm optimization algorithm to solve the scheduling strategy: control part of the photovoltaic module to reduce the output by 500kW, and instruct the energy storage equipment to charge with a charging power of 500kW. The scheduling instruction is sent to the photovoltaic module controller and the energy storage equipment control system.

[0020] The fault diagnosis and early warning module monitors that the inverter health index is 65 points and continuously decreases, inputs voltage fluctuation data, temperature data and the like of the inverter into the fault diagnosis model. The model compares historical data in the fault diagnosis knowledge base, diagnoses that the inverter may exist IGBT module aging failure, and predicts that the failure occurrence probability is 90%. Since the inverter has a greater impact on the power generation of the entire photovoltaic array, an emergency early warning information is sent, prompting that "the inverter IGBT module is aging, and a failure may occur within 24 hours, and maintenance needs to be arranged immediately".

[0021] The intelligent maintenance scheduling module receives the emergency early warning information, combines with a GIS geographic information system to display that the inverter is located in a photovoltaic base A area, is 3 kilometers away from a location of a nearest operation and maintenance personnel A, the operation and maintenance personnel A has inverter maintenance qualification, a maintenance vehicle is in good condition and carries corresponding spare parts. A maintenance scheduling optimization model is constructed, a genetic algorithm is adopted to plan an optimal maintenance route, and a maintenance task list is generated: the operation and maintenance personnel A is assigned to go to the A area to repair the inverter, an estimated departure time is 10 minutes later, an arrival time is 25 minutes later, and one IGBT module is needed. The maintenance task list is pushed to a user interactive interface of the operation and maintenance personnel A.

[0022] The system stores, after encryption by using an AES algorithm, original sensor data collected this time, structured data after preprocessing, a device state evaluation report, an energy scheduling strategy instruction, a fault diagnosis result, emergency early warning information, a maintenance task list and the like in a database, and updates an individual state feature library of the inverter, and adds the fault diagnosis result to a fault diagnosis knowledge base.

[0023] The system management module displays that each sensor collects normally, a communication link transmission is stable, and each module operates normally in real time; the data security module performs monthly data security detection, and no security loophole and intrusion behavior is found; an administrator views operation logs of the day through the system management module, confirms that all operation records are complete, and manually backs up system data. Embodiment Two

[0024] As shown in Figure 2 The embodiment provides an electrical service system for a photovoltaic base, which comprises a device data acquisition module, a data preprocessing module, a state monitoring and evaluation module, an energy management and scheduling module, a fault diagnosis and early warning module, an intelligent maintenance scheduling module, a communication transmission module, a user interactive module, a system management module and a data security module.

[0025] The principle of the device data acquisition module is as shown in Figure 3As shown, the instructions enter the device data collection module, pass through the sensor group and the transmission unit in turn, and finally output data to the data preprocessing module. The device data collection module is responsible for collecting the operation data of various electrical equipment in the photovoltaic base and environmental data. The electrical equipment operation data includes but is not limited to the voltage, current and power of the photovoltaic module, the output voltage, output current and conversion efficiency of the inverter, the direct current voltage and current of the busbar box, the oil temperature, winding temperature and insulation resistance of the transformer, etc. The environmental data includes the light intensity, ambient temperature, wind speed, wind direction and precipitation, etc. The collection equipment includes but is not limited to voltage sensor, current sensor, power sensor, temperature sensor, light sensor, wind speed sensor, wind direction sensor and rain sensor, etc. The collection equipment temperature and ambient temperature, the light sensor collects the light intensity, the wind speed and wind direction sensor collects the wind speed and wind direction, and the rain sensor collects the precipitation. The data transmission is realized by combining wired and wireless modes. The wired transmission adopts optical fiber communication, and the wireless transmission adopts LoRa, NB-IoT and other Internet of Things communication technologies.

[0026] The data preprocessing module processes the original data, including data cleaning, removing abnormal values caused by sensor failure or interference, using the Ralda criterion to judge and correct the abnormal values; data noise reduction, using wavelet transform noise reduction algorithm to process the collected noise data; data format conversion, converting the heterogeneous data collected by different types of sensors into a unified structured data format, such as converting different format data into JSON format structured data, missing value filling, using linear interpolation method based on time series to fill the missing values in the data transmission process; data standardization, standardizing the data of different magnitudes, mapping the data to the [0, 1] interval through the min-max standardization method, facilitating subsequent data analysis and model calculation.

[0027] The principle of the state monitoring and evaluation module is as follows Figure 4As shown, the data pre-processing data input state monitoring and evaluation module, feature extraction and time series analysis on the data, the health status of the equipment evaluation, the results output to the energy management module and fault diagnosis module. The specific steps are to build a deep learning-based equipment state evaluation model, input the pre-processed data into the model for analysis. The model uses a deep learning model that combines convolutional neural networks (CNN) and long short-term memory networks (LSTM). CNN is used to extract spatial features in the data, such as the relationship between different device parameters. LSTM is used to capture the time series characteristics of the data and analyze the dynamic trends of the device operating state. The health index of the device is calculated by the model, with a range of 0-100. The higher the health index, the better the device operating state. At the same time, an individual state feature library for each device and a comparative model for similar devices are established to achieve accurate evaluation and abnormal identification of the device operating state. When the device health score is below 70 points, it is marked as an abnormal state.

[0028] The energy management and scheduling module is based on the equipment operating state evaluation results, real-time power generation data and power grid load demand to construct an energy optimization scheduling model. The model aims to maximize the power generation of the photovoltaic base, minimize the light rejection rate and achieve stable grid connection with the power grid. It considers the power generation characteristics of photovoltaic components, the charging and discharging characteristics of energy storage devices, and the scheduling requirements of the power grid and other constraints. Particle swarm optimization algorithm is used to solve the model and develop the optimal energy scheduling strategy, including photovoltaic component output adjustment, energy storage device charging and discharging plan, and power exchange scheme with the power grid, etc. The scheduling instructions are sent to the corresponding execution devices. For example, when the light intensity is too high and the power generation exceeds the grid demand, the energy storage device charging instruction is issued.

[0029] The fault diagnosis and early warning module collects historical fault data and real-time operating data of the equipment, and constructs a fault diagnosis knowledge base and a deep learning-based fault diagnosis model. When the state monitoring and evaluation module finds that the device health index is below the set threshold or there is abnormal fluctuation, the relevant data is input into the fault diagnosis model. The model analyzes the data and identifies the possible fault types of the equipment, such as photovoltaic component aging, inverter failure, and busbar short circuit, and predicts the probability of fault occurrence and the possible impact. According to the severity and urgency of the fault, the corresponding warning information is issued, and the warning levels are divided into general warning, important warning and emergency warning. For example, when the device has abnormal voltage fluctuation and the health score continues to decline, it is diagnosed as inverter failure with a probability of 85%, and an important warning message is issued.

[0030] The intelligent maintenance scheduling module constructs a maintenance scheduling optimization model according to the early warning information issued by the fault diagnosis and early warning module, the importance of the equipment, and the distribution of maintenance resources. The model comprehensively considers factors such as the skill level of maintenance personnel, the position and state of maintenance vehicles, the inventory situation of maintenance tools and spare parts, etc., and uses a genetic algorithm to solve the optimal maintenance plan, including the assignment of maintenance personnel, the planning of maintenance routes, and the arrangement of maintenance time, etc. Through the GIS geographic information system, the positions of the equipment and the maintenance personnel are displayed in real time, realizing dynamic scheduling and optimal allocation of maintenance resources, improving maintenance efficiency, and reducing maintenance costs.

[0031] The communication transmission module adopts a layered communication architecture, including a perception layer, a network layer, and an application layer. The perception layer is responsible for the collection and preliminary transmission of equipment data, and uses short-range wireless communication technology to transmit sensor data to a local data acquisition terminal; the network layer uses a combination of optical fiber communication and 5G communication technology to realize large data transmission between the local data acquisition terminal and the remote monitoring center, ensuring high speed and low delay of data transmission; the application layer is responsible for data interaction and sharing between system modules, and uses the MQTT communication protocol to realize reliable transmission and real-time updating of data. At the same time, a communication state monitoring mechanism is set up, which automatically switches to a backup communication link when a communication failure is found, ensuring the continuity of data transmission.

[0032] The user interaction module provides a visual operation interface and data display platform for workers, supports management personnel to view the overall operation status of the photovoltaic large base, including real-time power generation, equipment health status, energy scheduling situation, fault early warning information, etc.; supports operation and maintenance personnel to view detailed operation data, fault diagnosis reports and maintenance task lists of specific equipment; allows management personnel to customize system parameters, early warning thresholds and energy scheduling strategies, etc. The interface uses graphical display methods such as line charts, column charts, pie charts, and geographic information maps, to intuitively and clearly present various types of information.

[0033] The system management module includes user account management, role permission allocation, system parameter configuration, data backup and recovery, operation log recording, etc. The administrator creates user accounts of different roles, such as administrators, operators, and operation and maintenance personnel, through this module, and allocates corresponding operation permissions for each role to ensure the safety and standardization of system operation. The system parameter configuration function allows the administrator to set parameters such as data collection frequency, state evaluation threshold, and fault early warning level according to the actual situation of the photovoltaic large base. The data backup and recovery function regularly backs up system data to prevent data loss and quickly recover data when the system fails. The operation log recording function records the user's login, operation, configuration modification, etc. in detail, facilitating system auditing and fault tracing.

[0034] The data security module adopts multiple security protection mechanisms to guarantee data security. In the data acquisition stage, the data collected by the sensor is encrypted, and the symmetric encryption algorithm AES is used to encrypt the data; in the data transmission stage, the SSL / TLS protocol is used to establish a secure transmission channel to prevent data from being stolen or tampered with during transmission; in the data storage stage, the stored data is encrypted using database encryption technology, and an access permission control mechanism is set up, only authorized users can access related data; regular data security detection and vulnerability scanning are carried out, and the intrusion detection system IDS and firewall technology are used to prevent network attacks, to ensure the security and privacy of the data.

[0035] The application realizes accurate monitoring of the equipment operation state, optimization of energy scheduling, timely diagnosis and early warning of faults and intelligent scheduling of maintenance resources by comprehensively collecting and analyzing the equipment operation data and environmental data of the photovoltaic large base, improves the power generation efficiency and operation management level of the photovoltaic large base, and provides a strong guarantee for the safe and stable operation of the photovoltaic large base.

[0036] The above is only the preferred embodiment of the application, and it should be pointed out that for ordinary skilled persons in the technical field, several improvements and modifications can be made without departing from the technical principles of the application, and these improvements and modifications should be regarded as the protection scope of the application.

Claims

1. A method for electrical service for a photovoltaic utility scale, characterized by, The method comprises the following steps: Collecting operation data and environmental data of photovoltaic large base electrical equipment; Preprocessing the collected operation data and environmental data to obtain structured data; According to the structured data, based on the pre-constructed equipment state evaluation model, quantitative evaluation is carried out to obtain the equipment state evaluation report; Based on the state evaluation report and the operation data, an energy scheduling strategy is generated through a first optimization algorithm; Receiving state monitoring data, based on the state monitoring data and the state evaluation report, analyzing based on the pre-constructed fault diagnosis knowledge base and fault diagnosis model to output fault warning information; Based on the fault warning information, a maintenance plan is output through a second optimization algorithm.

2. The method for electrical service for a photovoltaic farm of claim 1, wherein, The preprocessing is at least one of data cleaning, data denoising, data format conversion, missing value filling and data standardization processing.

3. The method for electrical service for a photovoltaic farm of claim 1, wherein, The pre-constructed equipment state evaluation model is a data input model that fuses convolutional neural network and long short-term memory network.

4. The method for electrical service for a photovoltaic farm of claim 1, wherein, The first optimization algorithm is a particle swarm optimization algorithm.

5. The method for electrical service for a photovoltaic farm of claim 1, wherein, The training of the pre-constructed fault diagnosis knowledge base and fault diagnosis model is based on equipment historical fault data and real-time operation data.

6. The method for electrical service for a photovoltaic farm of claim 1, wherein, The second optimization algorithm is a genetic algorithm.

7. The method for electrical service for a photovoltaic farm of claim 1, wherein, Further comprising: Establishing an equipment state feature library; At least one of the operation data, the environmental data, the structured data, the equipment state evaluation report, the energy adjustment strategy, the fault warning information and the maintenance plan is stored in an encrypted database, and the equipment state feature library and the fault diagnosis knowledge base are updated; Monitoring the system running state, carrying out data security detection and vulnerability scanning, recording and alarming abnormal conditions.

8. An electrical service system for a photovoltaic farm, comprising: The method comprises the following steps: Comprising a device data acquisition module, a data preprocessing module, a state monitoring and evaluation module, an energy management and scheduling module, a fault diagnosis and early warning module, an intelligent maintenance scheduling module, a communication transmission module, a user interaction module, a system management module and a data security module; The device data acquisition module is used for collecting operation data and environmental data of photovoltaic large base electrical equipment; The data preprocessing module is used for processing the operation data and environmental data to obtain preprocessed data; The state monitoring and evaluation module monitors and comprehensively evaluates the electrical equipment in real time according to the preprocessed data; The energy management and scheduling module formulates an energy scheduling strategy according to the equipment state evaluation report and the operation data; The fault diagnosis and early warning module analyzes and outputs fault warning information based on the fault diagnosis knowledge base and the fault diagnosis model; The intelligent maintenance scheduling module obtains a maintenance plan according to the fault warning information; The communication transmission module is used for realizing data interaction and transmitting data to the monitoring center; The user interaction module provides a visual operation interface to facilitate system operation and information viewing by workers; The system management module is responsible for user management, parameter configuration and log recording of the system; The data security module ensures the security and integrity of system data in the process of collection, transmission, storage and use.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program, which is executed by a processor, implements the steps of the method for electrical service of a photovoltaic large site according to any one of claims 1 to 7.

Images