An unattended photoelectric monitoring and intelligent operation and maintenance device and method

Abstract

This invention discloses an unattended intelligent operation and maintenance device for photovoltaic monitoring, comprising a solar photovoltaic panel, a wind turbine, a wind-solar conversion controller, a power supply box, a battery pack, equipment to be powered, an analysis and processing unit, a load, and multi-source sensor equipment. The wind-solar conversion controller connects to the solar photovoltaic panel, the wind turbine, the multi-source sensor equipment, and the power supply box, which in turn connects to the battery pack and the equipment to be powered. This invention solves the problems of manual operation and maintenance and difficulties in troubleshooting in traditional equipment operation and maintenance management through real-time monitoring, intelligent power supply switching, power usage duration prediction, and intelligent fault prediction. It is suitable for temporary or mobile deployment scenarios requiring autonomous power supply and communication, thus eliminating the need for infrastructure.

Description

Technical Field

[0001] This invention belongs to the field of optoelectronic monitoring and operation and maintenance management technology, and relates to the construction of unattended optoelectronic monitoring intelligent operation and maintenance device and intelligent operation and maintenance method based on unattended optoelectronic monitoring application, specifically an unattended optoelectronic monitoring intelligent operation and maintenance device and method. Background Technology

[0002] The main background for intelligent operation and maintenance management devices based on wind and solar power technology lies in the rapid development of renewable energy and the urgent need for efficient energy utilization. With global emphasis on reducing carbon emissions and achieving sustainable development, the application of renewable energy sources such as wind and solar power is becoming increasingly widespread. However, these energy sources are intermittent and unstable, greatly affected by natural conditions. This means that power generation may exceed demand at certain times, while it may be insufficient at other times. Therefore, the use of intelligent operation and maintenance management devices to monitor and regulate energy supply and demand in real time is proposed to ensure efficient energy utilization. However, the capacity of the batteries configured on-site can only sustain the load for a few hours. If backup batteries are not replaced in time, the load often depletes the battery's charge, causing over-discharge and irreversible damage. To avoid this, current solutions often involve assigning dedicated personnel for monitoring or rotating shifts, which undoubtedly increases personnel input and wastes human and financial resources. To address the aforementioned challenges, this invention proposes an unattended, intelligent photovoltaic monitoring and maintenance device and method. This invention efficiently utilizes wind and solar energy combined with batteries to provide power to the load equipment. Through software, it calculates photovoltaic efficiency and battery capacity in real time, predicting battery life under current power consumption and sending early warnings to users when the battery is low. Simultaneously, the device is equipped with a wireless communication interface and sensors such as anemometers and GPS / BeiDou navigation systems, enabling data analysis of recent local weather, light intensity, wind speed, and temperature, allowing for remote dynamic control of the power supply mode.

[0003] The background technology used in this invention is as follows: 1. Energy storage technology Wind and solar power generation typically requires energy storage devices to address the instability of power output, such as lithium batteries, lead-acid batteries, or new energy storage technologies (e.g., supercapacitors, flywheel energy storage). When wind and solar resources are insufficient, energy storage devices can provide power support; when wind and solar resources are abundant, energy storage devices store excess electricity. Therefore, the coordinated operation and maintenance management of energy storage and storage is particularly important.

[0004] 2. Power Electronics Technology The development of power electronics technology has enabled wind and solar power generation to convert and manage electrical energy more efficiently. Key equipment such as inverters and converters use power electronic devices to convert the direct current (DC) or alternating current (AC) generated by wind turbines and photovoltaic power generation to meet the electricity needs of the grid or users. At the same time, power electronics technology also makes energy management more intelligent, enabling rapid response to load changes and fluctuations in energy input.

[0005] 3. Internet of Things and Big Data Technology Through sensors and communication modules, multi-dimensional data such as wind speed, light intensity, temperature, and battery power can be collected in real time. This data can then be processed through a cloud platform and big data analytics to perform predictive maintenance, fault diagnosis, and power generation forecasting, thereby improving reliability and economy.

[0006] 4. Energy management algorithms and artificial intelligence technologies Intelligent operation and maintenance management often relies on advanced energy management algorithms, which can optimize various objectives such as cost minimization, energy efficiency maximization, and equipment lifespan extension. With the development of artificial intelligence and machine learning technologies, self-learning and adaptive mechanisms can be used to more accurately control and optimize the operation of wind and solar power generation units. By analyzing historical and real-time data, AI can predict future load demand and power generation capacity, thereby making more optimized scheduling decisions. Summary of the Invention

[0007] (a) Purpose of the invention The purpose of this invention is to provide an unattended photoelectric monitoring and intelligent operation and maintenance device and method, solving the following technical problems: 1. The intermittency and instability of wind and solar power generation Wind and solar energy are greatly affected by natural conditions, exhibiting randomness and intermittency. Photovoltaic power generation depends on sunlight intensity, cannot generate electricity at night, and its efficiency decreases on cloudy or rainy days; wind power generation is affected by wind speed, and excessively low or high wind speeds are unsuitable for power generation. Unattended photovoltaic monitoring and intelligent operation and maintenance management devices, by integrating multiple energy sources and optimizing scheduling, can maximize the utilization of wind and solar resources under different times and conditions, reducing the adverse effects of power generation fluctuations.

[0008] 2. Problem of insufficient energy utilization efficiency Wind and solar power generation is greatly affected by the external environment. Currently, many existing wind and solar power generation devices have insufficient energy utilization efficiency. The root cause is the lack of accurate environmental data prediction for the location. Energy supply is often estimated using empirical or ideal values, which may lead to power shortages or waste.

[0009] 3. Insufficient Automated Management Currently, some devices lack automated control methods, requiring personnel to shut down the load or switch the power supply when the battery level approaches a critical value, mechanically regulating charging and discharging, which greatly increases the maintenance burden and labor costs. On the other hand, existing devices lack real-time dynamic early warning mechanisms and load derating contingency plans, resulting in a high risk of battery over-discharge.

[0010] (II) Technical Solution To address the aforementioned technical problems, this invention provides an unattended intelligent operation and maintenance device for photoelectric monitoring, comprising a solar photovoltaic panel, a wind turbine, a wind-solar conversion controller, a power supply box, a battery pack, equipment to be powered, an analysis and processing unit, a load, and multi-source sensor equipment. The wind-solar conversion controller connects the solar photovoltaic panel, the wind turbine, the multi-source sensor equipment, and the power supply box, which in turn connects the battery pack and the equipment to be powered. The solar photovoltaic panel and the wind turbine convert solar and wind energy into electrical energy, respectively, and the wind-solar conversion controller outputs a stable voltage to the power supply box, which then provides power to the battery pack and the equipment to be powered. The wind-solar conversion controller collects data on local wind speed, humidity, and sunlight information by connecting to the multi-source sensor equipment. The analysis and processing unit then performs intelligent analysis and control based on the acquired data and actual needs, enabling intelligent switching of power supply modes.

[0011] Furthermore, the power supply box is equipped with a 4G / 5G wireless communication interface, which allows for real-time interaction with external devices and user terminals, receiving weather and environmental data, and enabling remote control of the power supply mode.

[0012] Furthermore, the multi-source sensor device includes a light sensor, a humidity sensor, a temperature sensor, and an anemometer arranged around the photovoltaic and wind power generation equipment to collect environmental data in real time, including solar radiation intensity, air humidity, equipment temperature, and wind speed.

[0013] This invention also provides an unattended intelligent operation and maintenance method for photoelectric monitoring, which includes the following processes: S1: Intelligent switching of wind-solar hybrid power supply mode; S2: Real-time power prediction and dynamic energy efficiency control; S3: Intelligent fault prediction based on big data models.

[0014] Furthermore, in S1, the collected data, including local solar radiation intensity, wind speed, wind direction, temperature, location, and time, are sent to the analysis and processing unit via a light sensor, anemometer, temperature sensor, and GPS / BeiDou. The analysis and processing unit performs intelligent analysis and controls the wind-solar conversion controller to switch power supply modes.

[0015] Furthermore, in S1, the analysis and processing unit compares local solar radiation intensity, wind speed, wind direction, temperature, location, and time data with the performance parameters of photovoltaic and wind power generation to determine in real time whether the current environmental conditions are suitable for power generation; predicts the power generation capacity for a future period based on wind speed and solar intensity data; and predicts current and future load demand by combining historical load data to ensure sufficient power supply after switching. Based on power generation capacity and load demand forecasts, wind power or photovoltaic power is dynamically selected as the main power source; when wind speeds are expected to be high and sunlight is insufficient, wind power is given priority; otherwise, photovoltaic power is switched to when sunlight is sufficient.

[0016] Furthermore, S2 includes the following process: 1) Environmental data acquisition and initial input a. Location setting and meteorological data import Set the geographical location information of the photovoltaic device and import historical meteorological data of that location as initial reference data; b. Real-time weather data connection Connect to weather forecast data sources to obtain real-time and forecast weather information, including solar intensity, cloud cover, rainfall, and wind speed parameters for the next few days, in order to more accurately predict the power generation and equipment usage time of the operation and maintenance equipment; 2) Power generation forecast Using location and real-time weather data, the daily effective sunshine duration is calculated based on the sun's position and solar intensity parameters. The solar trajectory, geographical location, and historical meteorological data are then used to simulate the daily radiation for the current day and the next few days, predicting the actual power generation of the photovoltaic modules at various time periods. The power generation calculation formula is as follows: in, Average solar irradiance: W / m², provided by real-time weather data; A represents the total area of ​​the photovoltaic modules; The overall efficiency of the photovoltaic device includes conversion efficiency, temperature loss, and reflection loss; H is the predicted sunshine duration. 3) Adaptive adjustment of self-learning algorithm a. Establish an environmental self-learning model Accumulate historical data, including the impact of sunlight, wind speed, and solar radiation intensity on power generation, and combine them with machine learning algorithms, regression models, or LSTM to update the power generation prediction model. b. Adjustment of Weather Impact Weighting Weights for wind and light conditions are set using weather forecast data and a self-learning model. 4) Remaining power prediction and usage duration assessment Combining weather forecast data and current device battery level, the system predicts battery usage for the next few days based on a self-learning model. Based on the forecast data, and considering the current battery level and power consumption requirements, it calculates the remaining battery life. When the battery status approaches a set threshold, it automatically generates an early warning message, including the current remaining battery level, available time, and load power, and sends a real-time warning to the user. 5) User recommendations and intelligent control After accessing weather data, depending on different weather conditions, we recommend that users maintain normal operation, or recommend that users switch to wind power mode, or suggest that users reduce unnecessary loads or add external power. Based on weather forecast data and current battery status, the power supply mode is intelligently adjusted, and the priority of photovoltaic or wind power supply is dynamically switched.

[0017] Furthermore, S3 includes the following processes: 1) Data Acquisition and Initial Modeling a. Multidimensional data acquisition During normal operation of the maintenance equipment, power supply status, power generation, communication status, and sensor data are collected as "health" baseline data for the equipment's operation. b. Big Data Model Establishment A health operation model of the device is established using the collected data. Through statistical analysis of the data, the characteristics of the device's communication frequency, data packet transmission interval, and voltage and current fluctuation range during normal operation are obtained. Each data point is recorded and correlated for use in training subsequent fault prediction models. 2) Real-time monitoring and anomaly detection a. Real-time monitoring of communication and power supply status During operation, the system continuously monitors the status of communication and power supply, including the stability of data transmission, signal strength, and battery voltage level; if communication data transmission is interrupted or abnormal power output occurs, it will automatically enter the abnormal detection state. b. Threshold comparison and anomaly detection The current monitoring data is compared and analyzed with the health operation model. The feature values ​​in the big data model are compared with the current values. If the data exceeds the set threshold range, it is initially judged that the device may be abnormal and enters the further fault prediction stage. 3) Intelligent prediction of fault causes a. Power supply fault detection By analyzing battery voltage and charging / discharging current data, it is determined whether the device has an abnormal power supply. If abnormal voltage fluctuations or insufficient power supply are detected, the device's historical power supply data and current power generation will be compared. If the power consumption significantly exceeds expectations, it will be determined as a power supply failure. b. Equipment self-inspection and step-by-step troubleshooting If the device's power supply is unstable, it will be tested step by step to check whether the fault originates from the battery, photovoltaic panel, or wind turbine. This process is based on the normal operating characteristic values ​​in the health model, and gradually analyzes the deviation between the current data and the historical average to identify possible faulty components. 4) Fault propagation and intelligent early warning a. Multi-fault combination prediction In the event of insufficient power supply and poor communication in the device, analyze whether it is a complex fault using big data model; based on the fault performance of similar situations in historical data, determine the possible combined faults of the device, and provide the warning level and suggested troubleshooting sequence; b. Fault propagation of self-learning models Regularly update the big data model, input device fault data and prediction results into the model for self-learning, automatically analyze the relationship between device load changes and communication delays, and form an automated prediction capability for such new faults. c. Early warning information push and user notification After completing the fault cause prediction, a detailed fault report is sent to the user terminal, explaining the specific faulty component, possible causes, and warning level; if the fault prediction is a simple fault, the device operation mode will be automatically adjusted to ensure the continuity of device operation.

[0018] (III) Beneficial Effects The unattended photoelectric monitoring intelligent operation and maintenance device and method provided by the above technical solution have the following beneficial effects: 1. Intelligent fault prediction and diagnosis The fault intelligent prediction module based on big data models achieves accurate prediction and intelligent diagnosis of equipment faults through historical data modeling and real-time data analysis, and combines big data models to perform combined analysis of multiple faults. Compared with the single fault detection methods in existing technologies, the device's self-learning algorithm can dynamically adapt to changes in equipment status and intelligently predict new fault modes. By improving the breadth and accuracy of fault identification, it reduces the time spent on manual troubleshooting and fault handling.

[0019] 2. Dynamic power supply mode switching and energy efficiency optimization Based on intelligent power supply mode switching of wind and solar complementarity, and combined with environmental data to optimize the power supply mode, the system ensures maximum power generation efficiency. Photovoltaic power generation is prioritized when there is sufficient sunshine, while wind power generation is prioritized under high wind speed conditions. Compared to traditional fixed power supply modes, this dynamic power supply regulation can effectively adapt to the power demand under different environmental conditions, significantly improving the overall energy efficiency of the system.

[0020] 3. Intelligent power prediction and adaptive regulation It can intelligently predict future power generation and usage by combining weather forecast data and self-learning algorithms, and adaptively adjust power supply priorities according to weather changes. This intelligent forecasting function provides users with accurate power consumption suggestions. When severe weather is predicted, it recommends reducing load or adding power sources to extend battery life and ensure long-term stable operation without supervision. Compared with existing technologies, it is closer to user needs and achieves highly personalized energy efficiency management.

[0021] 4. Intelligent remote connection and multi-level intelligent early warning Through a wireless communication module, key information such as real-time data, fault warnings, and remaining battery power is transmitted to the user terminal, allowing users to remotely view and control the device status in real time. The push of multi-dimensional warning information, including real-time power consumption, remaining battery life, and environmental conditions, facilitates immediate action by users in case of operational anomalies, preventing device failures from impacting system continuity. Compared to traditional passive monitoring solutions, this remote monitoring offers higher response speed and management flexibility. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the workflow of each functional unit in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the overall structure of the device according to an embodiment of the present invention.

[0024] Figure 3 This is a flowchart illustrating the intelligent switching of power supply modes in an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram illustrating the principle of calculating photovoltaic power generation in an embodiment of the present invention.

[0026] Figure 5 This is a flowchart illustrating the early warning and reminder process in an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, contents, and advantages of the present invention clearer, the specific embodiments of the present invention will be described in further detail below with reference to the accompanying drawings and examples.

[0028] This embodiment of the unattended photoelectric monitoring and intelligent operation and maintenance device is used for energy management of battery packs, referring to... Figure 1 and Figure 2As shown, the intelligent operation and maintenance device includes solar photovoltaic panels, a wind turbine, a wind-solar conversion controller, a power supply box, a battery pack, equipment to be powered, an analysis and processing unit, a load, and multi-source sensor equipment. The wind-solar conversion controller connects the solar photovoltaic panels, the wind turbine, the multi-source sensor equipment, and the power supply box, which in turn connects the battery pack and the equipment to be powered. The solar photovoltaic panels and the wind turbine convert solar and wind energy into electrical energy, respectively, and the wind-solar conversion controller outputs a stable voltage to the power supply box, which then provides power to the battery pack and the equipment to be powered. The wind-solar conversion controller collects data on local wind speed, humidity, and sunlight information by connecting to the multi-source sensor equipment. The analysis and processing unit then performs intelligent analysis and control based on the acquired data and actual needs, enabling intelligent switching of power supply modes.

[0029] The power supply box is equipped with a 4G / 5G wireless communication interface, which allows it to interact with external devices and user terminals in real time, receive weather and environmental data, and remotely control the power supply mode.

[0030] The functions of each unit in this embodiment are described as follows: S1: Intelligent switching of wind-solar hybrid power supply mode This device connects to multiple sensor devices, including anemometers, temperature sensors, and GPS / BeiDou navigation systems, and sends the collected data, such as local solar radiation intensity, wind speed, wind direction, and temperature, to an analysis and processing unit. The unit then performs intelligent analysis and controls the power supply mode switching. The specific implementation steps are as follows: (1) Multi-source sensor data acquisition and environmental monitoring Multi-source sensor equipment deployment: Light sensors, humidity sensors, temperature sensors, anemometers, etc. are deployed near photovoltaic and wind power generation equipment to collect environmental data such as solar radiation intensity, air humidity, equipment temperature and wind speed in real time.

[0031] GPS / BeiDou positioning system support: The system integrates a GPS / BeiDou positioning module to provide precise location and time information for the device, providing geographic reference data for subsequent remote management.

[0032] Data integration and transmission: Sensor data is aggregated by the wind-solar conversion controller and transmitted to the analysis and processing unit as the basis for power supply mode switching decisions.

[0033] (2) Energy management algorithm and prediction switching Environmental data, including current solar irradiance, wind speed, and temperature, is monitored in real time using multi-source sensor devices. The analysis and processing unit compares this data with the performance parameters of photovoltaic and wind power generation to determine in real time whether the current environmental conditions are suitable for power generation. Using machine learning models or time series forecasting algorithms, the power generation capacity for a future period is predicted based on data such as wind speed and solar irradiance. Combined with historical load data, current and future load demands can also be predicted to ensure sufficient power supply after the switchover.

[0034] Based on power generation capacity and load demand forecasts, wind power or solar power is dynamically selected as the primary power source. When high wind speeds and insufficient sunlight are predicted, wind power is prioritized; conversely, when sunlight is abundant, the system switches to solar power to improve energy efficiency. Based on the predicted power generation capacity of wind and solar power, the system with the stronger capacity is selected as the primary energy source. See the attached diagram for a detailed process. Figure 3 As shown.

[0035] S2: Real-time power prediction and dynamic energy efficiency control The software integrates weather forecast data, self-learning algorithms, and intelligent recommendations to achieve intelligent prediction and dynamic control of the power supply for the entire device. It can accurately predict power generation and remaining power, providing users with appropriate power usage suggestions. The specific implementation steps are as follows: (1) Environmental data acquisition and initial input a. Location setting and meteorological data import Set the geographical location information of the photovoltaic device (including latitude, longitude, altitude, etc.), and import the historical meteorological data of that location (such as average annual sunshine duration, temperature, humidity, etc.) as initial reference data.

[0036] b. Real-time weather data connection Connect to weather forecast data sources to obtain real-time and forecast weather information, including parameters such as sunshine intensity, cloud cover, rainfall, and wind speed for the next few days, in order to more accurately predict the power generation of the device and the usage time of the equipment.

[0037] (2) Power generation forecast Using location and real-time weather data, the daily effective sunshine duration is calculated based on parameters such as solar position and solar radiation intensity. Then, using solar trajectory, geographical location, and historical meteorological data, the daily radiation for the current day and the next few days is simulated to predict the actual power generation of the photovoltaic modules at various time periods. See the attached diagram for a detailed explanation of the principle. Figure 4 As shown. The formula for calculating power generation is as follows: in, The average solar irradiance (W / m²) is provided by real-time weather data; A represents the total area of ​​the photovoltaic modules. H represents the overall efficiency of the photovoltaic device, including conversion efficiency, temperature loss, reflection loss, etc.; H represents the predicted sunshine duration.

[0038] The daily power generation can be accurately predicted based on the sunshine conditions in the next few days, and the remaining power generation time will be assessed based on this prediction.

[0039] (3) Adaptive adjustment of self-learning algorithm a. Environmental self-learning model Historical data is accumulated during long-term operation, including the impact of sunlight, wind speed, and solar radiation intensity on power generation. Based on this data, the software combines machine learning algorithms, such as regression models or LSTM (Long Short-Term Memory), to gradually update its power generation prediction model, making its predictions of environmental changes more accurate.

[0040] b. Adjustment of Weather Impact Weighting Weather forecast data and a self-learning model are used to weight wind and solar power conditions. During periods of continuous cloudy or rainy weather, the weight of solar power generation is reduced, while the priority of wind power generation is increased. If solar power conditions improve in the coming days, the system adaptively switches to prioritize solar power to extend battery life.

[0041] (4) Remaining power prediction and usage duration assessment Combining weather forecast data and current device power levels, a self-learning model predicts power usage for the next few days. If high winds and weak sunlight are expected in the coming days, battery status and load power monitoring will be used to estimate whether wind power can meet the equipment's needs. If so, wind power will be prioritized. If the weather is expected to remain sunny in the near future, the sufficiency of photovoltaic power generation will be calculated, and the remaining power's availability time will be predicted, reminding users that they do not need to pay excessive attention to the equipment.

[0042] Based on forecast data, combined with current battery level and power consumption requirements, the remaining battery life is accurately calculated. If adverse weather conditions or high load demand are predicted, a real-time alert will be sent to the user to ensure sufficient time for action. The alert mechanism is illustrated in the attached diagram. Figure 5 As shown, when the battery status approaches a set threshold, an early warning message is automatically generated, including the current remaining battery power, estimated usable time, and load power. An integrated wireless communication module sends real-time warnings to the user via SMS, email, or mobile phone. The warning content includes the current remaining battery power, estimated usable time, real-time power consumption under the current load, and suggested measures (such as reducing the load or replenishing power).

[0043] (5) User usage recommendations and intelligent control After integrating weather data, if ample sunshine is expected in the coming days, the battery life is predicted to be sufficient, and users are advised to maintain normal operation, even under heavy load conditions, without worrying about insufficient power. If wind speeds increase and remain high in the coming days, users are advised to switch to wind power mode to reduce the operational burden on the photovoltaic modules. In severe weather (such as rain or prolonged overcast skies), users will be advised to reduce unnecessary loads or add external power to extend battery life.

[0044] Based on weather forecast data and current battery status, the system intelligently adjusts the power supply mode, dynamically switching between photovoltaic (PV) and wind power priorities. Under conditions of strong sunlight and low wind speed, the system prioritizes PV power generation, while switching to wind power generation under conditions of high wind speed and insufficient sunlight, ensuring power supply stability.

[0045] S3: Intelligent Fault Prediction Based on Big Data Models The intelligent fault prediction function based on big data models intelligently predicts the causes of device failures by statistically analyzing historical operating data and current status, ensuring stable operation even under unattended conditions. This module can intelligently identify faults based on the device's power supply and communication status, including whether they are communication failures, power failures, or failures in other modules, facilitating quick problem troubleshooting for users. The specific implementation steps are as follows: (1) Data acquisition and initial modeling a. Multidimensional data acquisition During normal operation, the equipment continuously collects various operational data through multiple sensors and communication modules, including power supply status, power generation, communication status, and sensor data, as "health" benchmark data for equipment operation.

[0046] b. Big Data Model Establishment A health operation model for the equipment is established using the collected data. Statistical analysis of the data reveals characteristics such as communication frequency, data packet transmission interval, and voltage and current fluctuation range during normal operation. Each data point is recorded and correlated for use in training subsequent fault prediction models.

[0047] (2) Real-time monitoring and anomaly detection a. Real-time monitoring of communication and power supply status During operation, the system continuously monitors the status of the communication module and power supply, including data transmission stability, signal strength, and battery voltage level. If communication data transmission is interrupted or an abnormal power output occurs, the system will automatically enter an anomaly detection state.

[0048] b. Threshold comparison and anomaly detection The current monitoring data is compared and analyzed with the health operation model. By comparing the feature values ​​(such as communication frequency, voltage and current characteristics, etc.) in the big data model with the current values, if the data exceeds the set threshold range, it is preliminarily judged that the equipment may be abnormal, and the process enters the further fault prediction stage.

[0049] (3) Intelligent prediction of fault causes a. Power supply fault detection By analyzing power data such as battery voltage and charging / discharging current, it is determined whether the equipment is experiencing power supply abnormalities. If abnormal voltage fluctuations or insufficient power supply are detected, the historical power supply data of the equipment will be compared with the current power generation. If the power consumption significantly exceeds expectations, it will be determined as a power supply failure.

[0050] b. Equipment self-inspection and step-by-step troubleshooting If the equipment's power supply is unstable, the system will perform a step-by-step inspection from the power supply module to check whether the fault originates from the battery, photovoltaic panel, or wind turbine, in order to accurately identify the cause of the fault. This process is based on the normal operating characteristic values ​​in the health model, and gradually analyzes the deviation between the current data and the historical average to identify the possible faulty components.

[0051] (4) Fault propagation and intelligent early warning a. Multi-fault combination prediction In cases of insufficient power supply and communication problems, further analysis using big data models will be conducted to determine if the fault is complex (e.g., simultaneous failure of both power supply and communication modules). Based on historical data of similar fault occurrences, the system will identify potential combined faults and provide a warning level and suggested troubleshooting sequence.

[0052] b. Fault propagation of self-learning models Regularly update the big data model, inputting equipment failure data and prediction results into the model for self-learning, thereby identifying more failure modes in future inspections. It can automatically analyze the relationship between equipment load changes and communication latency, forming an automated prediction capability for such new failures, enhancing the breadth and accuracy of failure prediction.

[0053] c. Early warning information push and user notification After predicting the cause of the fault, a detailed fault report is sent to the user terminal via the communication module, specifying the faulty component, possible causes, and warning level. If the fault prediction indicates a simple fault, the equipment operating mode will be automatically adjusted to ensure continuous equipment operation.

[0054] Example This invention provides an unattended photoelectric monitoring and intelligent operation and maintenance device and method for temporary or mobile deployment scenarios requiring autonomous power supply and communication ("de-infrastructure-based"), targeting such applications. The device utilizes a combination of solar photovoltaic panels, wind turbines, and battery packs for power supply, and integrates multi-sensor monitoring, intelligent power switching, power prediction, and intelligent fault prediction to achieve comprehensive control over power management.

[0055] 1. Equipment installation and initial configuration Solar photovoltaic panels are installed in areas with ample sunlight, and wind turbines are installed in open areas with high wind speeds. Both are connected to a wind-solar conversion controller to ensure stable power output. Light sensors, anemometers, humidity sensors, and GPS / BeiDou positioning modules are deployed around the equipment to monitor environmental data and equipment operating status, ensuring 24 / 7 monitoring of photovoltaic efficiency, wind speed, and other parameters. Finally, a remote monitoring platform is connected via 4G / 5G wireless communication modules to achieve real-time transmission of equipment status and early warning information, enabling users to remotely view and control the equipment.

[0056] 2. Dynamic switching of intelligent power supply mode Sensors monitor environmental data in real time, including light intensity, wind speed, and temperature, and transmit the data to the analysis and processing unit. The wind-solar conversion controller automatically switches the power supply mode based on the current data to ensure stable power supply under different environmental conditions.

[0057] Future power supply conditions are predicted based on current wind speed and sunlight intensity. If sufficient sunlight is expected, photovoltaic power generation will be prioritized; if wind speed is high, wind power generation will be prioritized. During the switchover, short-term power demand will be supplemented by batteries to smooth power supply and avoid power outages caused by the switchover.

[0058] 3. Intelligent power consumption prediction and dynamic energy efficiency control The power generation calculation is based on the location of the input device and historical meteorological data. Real-time weather data (such as solar radiation intensity and wind speed) is obtained through a connection with a weather forecasting platform and serves as the basis for power generation prediction. Power generation is predicted based on weather forecast data, solar radiation intensity, wind speed, etc., and the remaining usage time is calculated in conjunction with the current battery level. When insufficient power generation is detected in the next few days, energy-saving suggestions (such as reducing load or increasing wind power generation) are automatically pushed to ensure a continuous power supply. The power generation prediction model is continuously updated through historical data and environmental changes, and the priority of photovoltaic and wind power generation is automatically adjusted according to weather conditions to improve the accuracy of power generation prediction and the device's adaptability.

[0059] 4. Intelligent Fault Prediction A health operation model is established based on collected voltage, current, and communication status data. During operation, real-time data is compared with the health model. When multiple faults are detected, the model is used to analyze whether the fault is complex and historical data is used to identify the fault location. Early warning information is automatically generated and pushed to the user terminal through the communication module to ensure that the user can take timely measures.

[0060] 5. User recommendations and remote control Based on weather forecast data, dynamic power consumption recommendations are provided to users. In the event of several days of sustained strong sunlight, users are advised to prioritize photovoltaic power generation; in the event of severe weather, users are advised to reduce load or increase external power supply. Users can monitor equipment status and remaining power in real time through the monitoring platform, and switch power supply modes or adjust loads according to the recommendations, achieving intelligent remote operation and maintenance management.

[0061] As can be seen from the above technical solution, the present invention has the following significant features: (1) This invention constructs an intelligent power assessment method by integrating real-time weather data and a self-learning algorithm. It can dynamically predict the remaining usage time based on light intensity, wind speed, and current battery power, and provide users with energy efficiency management suggestions. At the same time, it performs fault analysis and prediction based on big data models, identifies and judges the causes of equipment failures, and promptly pushes early warning information to users, thereby realizing intelligent and unmanned management of equipment.

[0062] (2) The unattended photoelectric monitoring intelligent operation and maintenance device of the present invention solves the problems of manual duty and difficulty in fault diagnosis in traditional equipment operation and maintenance management by real-time monitoring, intelligent power supply switching, power usage time prediction and fault intelligent prediction. It is suitable for temporary or mobile deployment scenarios that require autonomous power supply and autonomous communication to "de-infrastructure".

[0063] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An unattended photoelectric monitoring and intelligent operation and maintenance device, characterized in that, The system includes solar photovoltaic panels, wind turbines, a wind-solar conversion controller, a power supply box, a battery bank, equipment to be powered, an analysis and processing unit, a load, and multi-source sensor equipment. The wind-solar conversion controller connects the solar photovoltaic panels, wind turbines, multi-source sensor equipment, and the power supply box, which in turn connects the battery bank and the equipment to be powered. The solar photovoltaic panels and wind turbines convert solar and wind energy into electrical energy, respectively. The wind-solar conversion controller provides a stable voltage output to the power supply box, which in turn provides power to the battery bank and the equipment to be powered. The wind-solar conversion controller collects data on local wind speed, humidity, and sunlight information by connecting to multi-source sensor equipment. The analysis and processing unit then uses this data to perform intelligent analysis and control based on actual needs, enabling intelligent switching of power supply modes.

2. The unattended photoelectric monitoring and intelligent operation and maintenance device as described in claim 1, characterized in that, The power supply box is equipped with a 4G / 5G wireless communication interface, which allows it to interact with external devices and user terminals in real time, receive weather and environmental data, and remotely control the power supply mode.

3. The unattended photoelectric monitoring and intelligent operation and maintenance device as described in claim 2, characterized in that, The multi-source sensor device includes light sensors, humidity sensors, temperature sensors, and anemometers arranged around photovoltaic and wind power generation equipment to collect environmental data in real time, including solar radiation intensity, air humidity, equipment temperature, and wind speed.

4. A method for unattended photoelectric monitoring and intelligent operation and maintenance based on the unattended photoelectric monitoring and intelligent operation and maintenance device according to claim 3, characterized in that, Includes the following processes: S1: Intelligent switching of wind-solar hybrid power supply mode; S2: Real-time power prediction and dynamic energy efficiency control; S3: Intelligent fault prediction based on big data models.

5. The unattended photoelectric monitoring intelligent operation and maintenance method as described in claim 4, characterized in that, In S1, data collected by light sensors, anemometers, temperature sensors, and GPS / BeiDou, including local solar radiation intensity, wind speed, wind direction, temperature, location, and time, are sent to the analysis and processing unit. The analysis and processing unit performs intelligent analysis and controls the wind-solar conversion controller to switch power supply modes.

6. The unattended photoelectric monitoring intelligent operation and maintenance method as described in claim 5, characterized in that, In S1, the analysis and processing unit compares local solar radiation intensity, wind speed, wind direction, temperature, location and time data with the performance parameters of photovoltaic and wind power generation to determine in real time whether the current environmental conditions are suitable for power generation; predicts the power generation capacity for a period of time in the future based on wind speed and solar intensity data; and predicts the current and future load demand by combining historical load data to ensure sufficient power supply after the switch. Based on power generation capacity and load demand forecasts, wind power or photovoltaic power is dynamically selected as the main power source; when wind speeds are expected to be high and sunlight is insufficient, wind power is given priority; otherwise, photovoltaic power is switched to when sunlight is sufficient.

7. The unattended photoelectric monitoring intelligent operation and maintenance method as described in claim 6, characterized in that, S2 includes the following process: 1) Environmental data acquisition and initial input a. Location setting and meteorological data import Set the geographical location information of the photovoltaic device and import historical meteorological data of that location as initial reference data; b. Real-time weather data connection Connect to weather forecast data sources to obtain real-time and forecast weather information, including solar intensity, cloud cover, rainfall, and wind speed parameters for the next few days, in order to more accurately predict the power generation and equipment usage time of the operation and maintenance equipment; 2) Power generation forecast Using location and real-time weather data, the daily effective sunshine duration is calculated based on the sun's position and solar intensity parameters. The solar trajectory, geographical location, and historical meteorological data are then used to simulate the daily radiation for the current day and the next few days, predicting the actual power generation of the photovoltaic modules at various time periods. The power generation calculation formula is as follows: in, Average solar irradiance: W / m², provided by real-time weather data; A represents the total area of ​​the photovoltaic modules; The overall efficiency of the photovoltaic device includes conversion efficiency, temperature loss, and reflection loss; H is the predicted sunshine duration. 3) Adaptive adjustment of self-learning algorithm a. Establish an environmental self-learning model Accumulate historical data, including the impact of sunlight, wind speed, and solar radiation intensity on power generation, and combine them with machine learning algorithms, regression models, or LSTM to update the power generation prediction model. b. Adjustment of Weather Impact Weighting Weights for wind and light conditions are set using weather forecast data and a self-learning model. 4) Remaining power prediction and usage duration assessment Combining weather forecast data and current device battery level, the system predicts battery usage for the next few days based on a self-learning model. Based on the forecast data, and considering the current battery level and power consumption requirements, it calculates the remaining battery life. When the battery status approaches a set threshold, it automatically generates an early warning message, including the current remaining battery level, available time, and load power, and sends a real-time warning to the user. 5) User recommendations and intelligent control After accessing weather data, depending on different weather conditions, we recommend that users maintain normal operation, or recommend that users switch to wind power mode, or suggest that users reduce unnecessary loads or add external power. Based on weather forecast data and current battery status, the power supply mode is intelligently adjusted, and the priority of photovoltaic or wind power supply is dynamically switched.

8. The unattended photoelectric monitoring intelligent operation and maintenance method as described in claim 7, characterized in that, S3 includes the following processes: 1) Data Acquisition and Initial Modeling a. Multidimensional data acquisition During normal operation of the maintenance equipment, power supply status, power generation, communication status, and sensor data are collected as "health" baseline data for the equipment's operation. b. Big Data Model Establishment A health operation model of the device is established using the collected data. Through statistical analysis of the data, the characteristics of the device's communication frequency, data packet transmission interval, and voltage and current fluctuation range during normal operation are obtained. Each data point is recorded and correlated for use in training subsequent fault prediction models. 2) Real-time monitoring and anomaly detection a. Real-time monitoring of communication and power supply status During operation, the system continuously monitors the status of communication and power supply, including the stability of data transmission, signal strength, and battery voltage level; if communication data transmission is interrupted or abnormal power output occurs, it will automatically enter the abnormal detection state. b. Threshold comparison and anomaly detection The current monitoring data is compared and analyzed with the health operation model. The feature values ​​in the big data model are compared with the current values. If the data exceeds the set threshold range, it is initially judged that the device may be abnormal and enters the further fault prediction stage. 3) Intelligent prediction of fault causes a. Power supply fault detection By analyzing battery voltage and charging / discharging current data, it can be determined whether the device has an abnormal power supply. If abnormal voltage fluctuations or insufficient power supply are detected, the historical power supply data of the device will be compared with the current power generation. If the power consumption significantly exceeds expectations, it will be judged as a power supply failure. b. Equipment self-inspection and step-by-step troubleshooting If the device's power supply is unstable, it will be tested step by step to check whether the fault originates from the battery, photovoltaic panel, or wind turbine. This process is based on the normal operating characteristic values ​​in the health model, and gradually analyzes the deviation between the current data and the historical average to identify possible faulty components.

9. The unattended photoelectric monitoring intelligent operation and maintenance method as described in claim 8, characterized in that, S3 also Includes the following processes: 4) Fault propagation and intelligent early warning a. Multi-fault combination prediction In the event of insufficient power supply and poor communication in the device, analyze whether it is a complex fault using big data model; based on the fault performance of similar situations in historical data, determine the possible combined faults of the device, and provide the warning level and suggested troubleshooting sequence; b. Fault propagation of self-learning models Regularly update the big data model, input device fault data and prediction results into the model for self-learning, automatically analyze the relationship between device load changes and communication delays, and form an automated prediction capability for such new faults. c. Early warning information push and user notification After completing the fault cause prediction, a detailed fault report is sent to the user terminal, explaining the specific faulty component, possible causes, and warning level; if the fault prediction is a simple fault, the device operation mode will be automatically adjusted to ensure the continuity of device operation.

10. An application of an unattended photoelectric monitoring intelligent operation and maintenance method based on any one of claims 4-9 in photoelectric monitoring and operation and maintenance management technology.

Images