Localized display and energy management method of solar controller

Abstract

The invention provides a localized display and energy management method of a solar controller, and relates to the technical field of solar energy management. The method comprises the steps of collecting output parameters and environmental parameters of a solar cell panel in real time; constructing an adjustment model based on a recurrent neural network, optimizing model hyper-parameters by using a genetic algorithm, inputting output parameters and environmental parameters of the solar cell panel, and outputting an adjustment strategy; and generating a control instruction according to the adjustment strategy, and adjusting the working parameters of the solar controller. According to the invention, by constructing the adjustment model and collecting historical data for training, the optimal control instruction of the solar controller can be obtained according to the current output parameters of the solar cell panel and the environmental parameters. The energy utilization efficiency is improved, the user monitoring capability is enhanced, and a new way is opened up for wide application and future development of solar energy.

Description

Technical Field

[0001] The present invention relates to the technical field of solar energy management, and particularly to a method for local display and energy management of a solar controller. Background Art

[0002] In the context of the growing global energy demand and the increasing emphasis on environmental protection, solar energy, as a clean and renewable energy source, has been expanding its application scope. While solar power generation systems provide us with green energy, they also face a series of technical challenges. Traditional energy supply models mainly rely on fossil fuels, which not only face the crisis of resource depletion but also cause great pollution and damage to the environment. With the progress of technology and the enhancement of environmental awareness, people have started to actively explore and utilize renewable energy sources. Solar energy, due to its rich resources and pollution-free characteristics, has become one of the most promising energy sources. A solar power generation system usually consists of components such as solar panels, inverters, storage batteries, and controllers. Among them, the solar controller plays a crucial role. It is responsible for managing the power generation of solar panels, the charging and discharging of storage batteries, to ensure the stable operation and efficient energy utilization of the entire system.

[0003] However, in early solar controllers, users could not directly obtain the key operating parameters and status information of the system at the equipment installation site, which brought great inconvenience to the monitoring and maintenance of the system. Users were difficult to understand the working conditions of the system in real time and could not timely discover and solve potential problems, thus affecting the reliability and energy utilization efficiency of the system. At the same time, the solar controller could not be optimized and adjusted according to the real-time energy supply and demand situation and environmental conditions. For example, it could not make full use of solar energy for efficient charging when the light was sufficient, or could not reasonably adjust the discharge strategy of the storage battery when the load demand was low, resulting in energy waste and a decline in system performance. There are significant differences in climate conditions, light intensity, and load demands in different regions, but traditional solar controllers often adopt unified and fixed control parameters and strategies, and cannot flexibly adapt to various actual application scenarios, further restricting the wide application and energy utilization effect of solar energy systems. Summary of the Invention

[0004] The present invention provides a method for local display and energy management of a solar controller to solve the defects existing in the prior art.

[0005] The present invention provides a method for local display and energy management of a solar controller, including:

[0006] Collecting the output parameters and environmental parameters of the solar panel in real time, and preprocessing the output parameters and environmental parameters to obtain preprocessed data.

[0007] Build a solar controller intelligent adjustment model based on a recurrent neural network, and use a genetic algorithm to optimize the hyperparameters of the intelligent adjustment model. Take the preprocessed data as input and output an adjustment strategy, where the adjustment strategy includes the adjustment amount of the charging current and the adjustment amount of the charging voltage.

[0008] Generate control instructions according to the adjustment strategy, adjust the working parameters of the solar controller through the control instructions, and perform local display.

[0009] Continuously evaluate the adjustment effect of the adjustment strategy, and optimize the intelligent adjustment model according to the adjustment effect.

[0010] According to a local display and energy management method for a solar controller provided by the present invention, the output parameters include output voltage, output current, and output power, and the environmental parameters include light intensity, temperature, and humidity.

[0011] According to a local display and energy management method for a solar controller provided by the present invention, the preprocessing process includes cleaning, screening, and conversion of the working parameters and environmental parameters, removing outliers and noise, and obtaining preprocessed data.

[0012] According to a local display and energy management method for a solar controller provided by the present invention, the process of building a solar controller intelligent adjustment model based on a recurrent neural network includes:

[0013] Collect historical data of solar panels, where the historical data includes historical working parameters, historical environmental parameters, and corresponding historical adjustment strategies. Preprocess the historical data to obtain historical preprocessed data, and divide it into a training set and a test set.

[0014] Extract input features from the historical preprocessed data, where the input features include historical output voltage features, historical output current features, historical output power features, historical light intensity features, historical temperature features, and historical humidity features.

[0015] Design a basic recurrent neural network model, where the basic model includes an input layer, a recurrent layer, and an output layer. The input layer is used to receive the input features and set the dimension of the input layer according to the quantity of the input features. The recurrent layer includes a preset number of neurons, which are used to capture the temporal features in the input features and generate control information. The output layer is used to convert the control information into a continuous adjustment strategy using a linear activation function.

[0016] Take the input features as input and the historical adjustment strategy as output, train the basic model, and retain the model parameters that meet the test accuracy to obtain the intelligent adjustment model.

[0017] A method for local display and energy management of a solar controller provided by the present invention. The process of using a genetic algorithm to optimize the hyperparameters of an intelligent adjustment model includes:

[0018] Taking the learning rate, the number of layers of the recurrent layer, and the number of neurons in the recurrent layer as hyperparameters to be optimized.

[0019] Setting a search space for each hyperparameter, where the search space represents the value range of the hyperparameter.

[0020] Randomly generating hyperparameter combinations from the search space as gene encodings, and constructing an initial population by generating multiple gene encodings. Using each gene encoding in the initial population to train the intelligent adjustment model, evaluating the performance index of the intelligent adjustment model on the validation set, where the performance index is the mean squared error. Judging whether the performance index reaches a preset value. If so, selecting the hyperparameter combination corresponding to the gene encoding with the smallest performance index as the hyperparameter combination of the intelligent adjustment model. Otherwise, updating the gene encodings in the initial population. The process includes:

[0021] Sorting the hyperparameter combinations in the initial population according to the performance index.

[0022] Based on the evolutionary mechanism of the genetic algorithm, performing crossover and mutation operations on the gene encodings in the initial population to generate new gene encodings.

[0023] Calculating the performance index of the intelligent adjustment model under the new gene encodings and adding them to the initial population.

[0024] A method for local display and energy management of a solar controller provided by the present invention. The process of performing crossover and mutation operations on the hyperparameter combinations in the initial population includes: According to the sorting of the performance index, selecting the two sets of hyperparameter combinations with the smallest mean squared error as the parents. Setting a crossover probability and selecting the multi-point crossover method, exchanging the values of the hyperparameters in the parent hyperparameter combinations, and mutating the values of the hyperparameters in the parent hyperparameter combinations within the search space according to a preset mutation probability to obtain new hyperparameter combinations.

[0025] A method for local display and energy management of a solar controller provided by the present invention. The process of generating a control instruction according to an adjustment strategy includes:

[0026] Collecting the working parameters of the solar controller, where the working parameters include charging current data and charging voltage data, and calculating the target charging current value and the target charging voltage value according to the adjustment strategy.

[0027] Generating a digital pulse signal according to the target charging current value and the target charging voltage value.

[0028] Amplify the digital pulse signal using a signal amplification circuit to obtain a control instruction, where the control instruction includes the adjustment amount of the current regulator and the adjustment amount of the voltage regulator.

[0029] According to a method for local display and energy management of a solar controller provided by the present invention, the process of local display includes:

[0030] Collect relevant data in real time, where the relevant data includes the output parameters of the solar panel, environmental parameters, adjustment strategies, and control instructions of the solar controller.

[0031] Sort, convert, and format the relevant data.

[0032] Connect the display device to the solar controller through a parallel interface.

[0033] Write a driver program and a display program for the display device according to the performance parameters of the display device and the format of the relevant data, so that the relevant data is displayed according to the preset layout, format, font, and color.

[0034] According to a method for local display and energy management of a solar controller provided by the present invention, the process of evaluating the effect of the adjustment strategy includes:

[0035] Collect the output parameters of the solar panel and the working parameters of the solar controller in real time.

[0036] Calculate the energy productivity according to the output parameters of the solar panel and the working parameters of the solar controller.

[0037] Evaluate the effect of the adjustment strategy by comparing the energy productivity before implementing the adjustment strategy and the energy productivity after implementing the adjustment strategy.

[0038] According to a method for local display and energy management of a solar controller provided by the present invention, the ways to optimize the intelligent adjustment model according to the adjustment effect include optimizing the hyperparameters of the intelligent adjustment model and improving the structure of the intelligent adjustment model.

[0039] The localization display and energy management method of the solar controller provided by the present invention can provide a solid foundation for subsequent intelligent control and energy management decisions by obtaining the most accurate and timely working parameters and environmental parameters of the solar panel and performing preprocessing, avoiding control errors and energy waste caused by inaccurate or lagged data. By constructing an intelligent control model based on a recurrent neural network and using advanced algorithms to optimize hyperparameters, the accuracy and adaptability of the control strategy are improved. It can extract key input features from the preprocessed data, thereby outputting a more reasonable control strategy, including precisely adjusting the charging current and voltage. This not only effectively improves the conversion efficiency of solar energy, extends the service life of the solar panel and the storage battery, but also reduces the maintenance cost of the system. The realization of the localization display function enables users to intuitively and conveniently obtain the real-time working status and key parameters of the system on-site at the device. This greatly improves the user's monitoring ability of the solar energy system, enables users to promptly discover potential problems and take corresponding measures, reduces the possibility of failures and the repair time, and ensures the stable operation of the system. The control instructions generated according to the control strategy can precisely control the working status of the solar controller, realizing the refined management of energy. By intelligently adjusting the charging current and voltage, the solar energy resources can be fully utilized, storing energy to the maximum extent when the sunlight is sufficient and reasonably releasing the stored energy when the sunlight is insufficient or the load demand is high, effectively reducing energy waste and improving the energy self-sufficiency ability of the entire system. Continuously evaluating the effect of the control strategy and optimizing and improving the intelligent control model forms a virtuous feedback loop, enabling the system to continuously adapt to environmental changes and user needs and always maintain the optimal working state. This helps to reduce the dependence on traditional energy sources, reduce carbon emissions, and is of great significance to environmental protection and sustainable development. At the same time, improving the performance and reliability of the solar energy system will also promote the application and popularization of solar energy technology in a wider range of fields, making a positive contribution to the global energy transformation. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0041] Figure 1 It is a schematic flowchart of the localization display and energy management method of the solar controller provided by the embodiment of the present invention;

[0042] Figure 2 It is a schematic flowchart of optimizing the hyperparameters of the intelligent adjustment model in the embodiment of the present invention;

[0043] Figure 3 It is a schematic flowchart of generating control instructions according to an adjustment strategy in an embodiment of the present invention. Specific embodiments

[0044] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts shall fall within the protection scope of the present invention.

[0045] The following will be combined with Figures 1 - 3 Describe the localization display and energy management method of the solar controller of the present invention.

[0046] Figure 1 It is a schematic flowchart of the localization display and energy management method of the solar controller provided by an embodiment of the present invention.

[0047] As Figure 1 shown, the localization display and energy management method of the solar controller provided by an embodiment of the present invention includes:

[0048] Collect the output parameters and environmental parameters of the solar panel in real time, and preprocess the output parameters and environmental parameters to obtain preprocessed data.

[0049] The output parameters include output voltage, output current and output power, and the environmental parameters include light intensity, temperature and humidity.

[0050] The process of preprocessing includes cleaning, screening and converting the working parameters and environmental parameters, removing outliers and noise, and obtaining preprocessed data.

[0051] In this embodiment, a Hall sensor is selected to monitor the voltage output by the solar panel, a Hall effect sensor is used to monitor the current output, and the output power is obtained by multiplying the current and voltage. A light intensity sensor such as a photoelectric sensor or a solar radiation sensor is selected to monitor the light intensity in real time. A temperature sensor is used to monitor the environmental temperature. A humidity sensor is used to measure the relative humidity.

[0052] According to the specifications and historical data of the device, set a reasonable physical range. For example, the output voltage should be between 0 - 24V, and the temperature range may be between -40°C and 85°C. Use the standard deviation method to set a reasonable threshold to identify and mark outliers. Apply the moving average method to smooth the data and remove noise. Only retain the valid data that meets the threshold range and discard the outliers in the records. Use data normalization to process the data and normalize the influence of different parameters on model training to the same magnitude.

[0053] Construct an intelligent adjustment model of a solar controller based on a recurrent neural network, and use a genetic algorithm to optimize the hyperparameters of the intelligent adjustment model. Take the preprocessed data as input and output an adjustment strategy, where the adjustment strategy includes the adjustment amount of the charging current and the adjustment amount of the charging voltage.

[0054] The process of constructing an intelligent adjustment model of a solar controller based on a recurrent neural network includes:

[0055] Collect historical data of the solar panel, where the historical data includes historical working parameters, historical environmental parameters, and corresponding historical adjustment strategies. Preprocess the historical data to obtain historical preprocessed data, and divide it into a training set and a test set.

[0056] Extract input features from the historical preprocessed data, where the input features include historical output voltage features, historical output current features, historical output power features, historical light intensity features, historical temperature features, and historical humidity features.

[0057] Design a basic recurrent neural network model, where the basic model includes an input layer, a recurrent layer, and an output layer. The input layer is used to receive the input features and set the dimension of the input layer according to the number of input features. The recurrent layer includes a preset number of neurons, which are used to capture the temporal features in the input features and generate control information. The output layer is used to convert the control information into a continuous adjustment strategy using a linear activation function.

[0058] Take the input features as input and the historical adjustment strategy as output, train the basic model, and retain the model parameters that meet the test accuracy to obtain an intelligent adjustment model.

[0059] Figure 2 It is a schematic flow chart of optimizing the hyperparameters of the intelligent adjustment model in the embodiments of the present invention.

[0060] As Figure 2 shown, the process of using a genetic algorithm to optimize the hyperparameters of the intelligent adjustment model includes:

[0061] Take the learning rate, the number of layers of the recurrent layer, and the number of neurons in the recurrent layer as hyperparameters to be optimized.

[0062] Set a search space for each hyperparameter, where the search space represents the value range of the hyperparameter.

[0063] Randomly generate hyperparameter combinations from the search space as gene encodings, and construct an initial population by generating multiple gene encodings. Use each gene encoding in the initial population to train the intelligent adjustment model, evaluate the performance metrics of the intelligent adjustment model on the validation set. The performance metric is the mean squared error. Determine whether the performance metric reaches the preset value. If so, select the hyperparameter combination corresponding to the gene encoding with the smallest performance metric as the hyperparameter combination of the intelligent adjustment model. Otherwise, update the gene encodings in the initial population. The process includes:

[0064] Sort the hyperparameter combinations in the initial population according to the performance metric.

[0065] Based on the evolutionary mechanism of the genetic algorithm, perform crossover and mutation operations on the gene encodings in the initial population to generate new gene encodings. The process includes: According to the sorting of the performance metric, select the two sets of hyperparameter combinations with the smallest mean squared error as the parents. Set the crossover probability and select the multi-point crossover method to exchange the values of the hyperparameters in the parent hyperparameter combinations. Mutate the values of the hyperparameters in the parent hyperparameter combinations according to the preset mutation probability within the search space to obtain new hyperparameter combinations.

[0066] Calculate the performance metric of the intelligent adjustment model under the new gene encoding and add it to the initial population.

[0067] Generate a control instruction according to the adjustment strategy, adjust the working parameters of the solar controller through the control instruction, and perform local display.

[0068] Figure 3 It is a schematic flowchart of generating a control instruction according to the adjustment strategy in an embodiment of the present invention.

[0069] As Figure 3 shown, the process of generating a control instruction includes:

[0070] Collect the working parameters of the solar controller. The working parameters include charging current data and charging voltage data, and calculate the target charging current value and the target charging voltage value according to the adjustment strategy.

[0071] Generate a digital pulse signal according to the target charging current value and the target charging voltage value.

[0072] Use a signal amplification circuit to amplify the digital pulse signal to obtain a control instruction. The control instruction includes the adjustment amount of the current regulator and the adjustment amount of the voltage regulator.

[0073] In this embodiment, a current sensor and a voltage sensor are installed to monitor the charging current and charging voltage of the solar controller in real time. Data on the charging current and charging voltage are collected from the sensors regularly. Based on maximizing the charging efficiency and extending the battery life, an adjustment strategy is formulated to set the desired values for the charging current and voltage. According to the current charging current and voltage values and the ambient light intensity, the adjustment strategy is used to calculate the target charging current value and the target charging voltage value. If the charging current is lower than the set value, the target charging current is increased.

[0074] The calculated target charging current and voltage values are converted into digital signals suitable for use by the controller and these values are normalized to the range of 0 to 255 to adapt to the pulse-width modulation signal output. The output current and voltage are adjusted by changing the duty cycle of the signal, and a pulse-width modulation signal is generated using the digital signal.

[0075] An amplifier circuit is designed using an operational amplifier to amplify the pulse-width modulation signal to the required control amplitude. The output of the pulse-width modulation signal is connected to the input terminal of the amplifier, and the output terminal of the amplifier is connected to the control input terminals of the current regulator and the voltage regulator. The amplified pulse-width modulation signal serves as a control instruction indicating the adjustment amounts of the current regulator and the voltage regulator. By adjusting the outputs of these devices, the actual charging current and voltage will be adjusted to the target values, thereby achieving effective charging control.

[0076] The process of performing local display includes:

[0077] Relevant data is collected in real time. The relevant data includes the output parameters of the solar panel and the environmental parameters, as well as the adjustment strategy and control instructions of the solar controller.

[0078] The relevant data is sorted, converted, and formatted.

[0079] The display device is connected to the solar controller through a parallel interface.

[0080] According to the performance parameters of the display device and the format of the relevant data, a driver program and a display program are written for the display device so that the relevant data is displayed in a preset layout, format, font, and color.

[0081] In this embodiment, the sensors should be able to monitor the output parameters of the solar panel and the environmental parameters in real time to ensure that all sensors are correctly installed and operating properly. At the same time, the adjustment strategy and control instructions of the solar controller are collected. A fixed time interval is set, and various parameter data are regularly read through the interface of the controller for data collection to ensure the timeliness and accuracy of the data.

[0082] An object containing all the parameters is created, with each relevant value nested within.

[0083] Check the reasonable ranges of values such as voltage and current, remove data points that do not meet the conditions, and ensure that the data has no null values or anomalies.

[0084] Convert the values into specific formats according to the requirements of the display device, such as floating-point type, string, etc. The processed data should meet the input requirements of the display device, including type and precision. Ensure that the units of all parameters are unified.

[0085] Determine the appropriate interface type for connection according to the display device used. Connect the display device to the controller through a suitable parallel interface. Ensure that the power supply, ground, and data lines are correctly connected during connection to avoid short circuits or signal interference problems.

[0086] Write a driver program to control the basic functions of the display device according to the technical documentation and interface protocol of the display device. This includes initializing the display, setting the display mode, and processing data reading of input signals.

[0087] Develop a display program that is responsible for displaying the sorted data on the display device according to the preset layout, format, font, and color. Ensure that the design meets the visual layout, including:

[0088] Layout: Determine the positions of different data items on the display screen and create a clear information display structure.

[0089] Format: Use appropriate formats for display according to the data type.

[0090] Font and color: Select appropriate font sizes and colors to ensure that the information is clear and easy to read. Different colors and styles can also be used to highlight key data.

[0091] Continuously evaluate the adjustment effect of the adjustment strategy and optimize the intelligent adjustment model according to the adjustment effect.

[0092] The process of evaluating the effect of the adjustment strategy includes:

[0093] Collect the output parameters of the solar panel and the working parameters of the solar controller in real time.

[0094] Calculate the energy productivity based on the output parameters of the solar panel and the working parameters of the solar controller.

[0095] Evaluate the effect of the adjustment strategy by comparing the energy productivity before implementing the adjustment strategy and the energy productivity after implementing the adjustment strategy.

[0096] The ways to optimize the intelligent adjustment model according to the adjustment effect include optimizing the hyperparameters of the intelligent adjustment model and improving the structure of the intelligent adjustment model.

[0097] Example 1: Output parameters of the solar panel: Output voltage: 36V, Output current: 8A, Output power: 288W. Environmental parameters: Light intensity: 800lx, Temperature: 25°C, Humidity: 60%.

[0098] Removing outliers: If the detected voltage is below 20V or above 40V, it is marked as noise. All the collected data is screened, and the preprocessed output data is: Organized output voltage: 36V, Organized output current: 8A, Organized light intensity: 800lx, Organized temperature: 25°C, Organized humidity: 60%.

[0099] The collected historical data includes data records of 5 cycles, summarized as follows: Output voltage: [34V, 35V, 36V, 37V, 36V], Output current: [7A, 7.5A, 8A, 7.8A, 8A], Output power: [238W, 262.5W, 288W, 294.6W, 288W], Light intensity: [700lx, 750lx, 800lx, 850lx, 820lx], Temperature: [24°C, 25°C, 26°C, 25°C, 24.5°C], Humidity: [58%, 59%, 60%, 61%, 60%]. These data are preprocessed to obtain historical preprocessed data.

[0100] Extract input features from the historical data as the input, and the historical adjustment strategy as the output. By building a recurrent neural network, including: Input layer: containing 6 neurons, Recurrent layer: set with 50 neurons, Output layer: output the adjustment strategy.

[0101] Train the recurrent neural network. After training, the test accuracy of the model reaches over 90%.

[0102] Optimized hyperparameters include: Learning rate, range: 0.001 to 0.1. Number of recurrent layers, range: 1 to 3 layers. Number of neurons in each layer, range: 20 to 100.

[0103] The initial population includes a set of gene encodings: [0.01, 2, 50], [0.005, 1, 70], [0.05, 3, 30]. By evaluating the mean squared error, select the best hyperparameters, and it is expected that the MSE of the final model is reduced to 0.015.

[0104] Current charging current adjustment amount: 2A, Current charging voltage adjustment amount: 1V.

[0105] Calculate the target value according to the adjustment strategy:

[0106] Target charging current value: Current current 8A + adjustment amount 2A = 10A.

[0107] Target charging voltage value: Current voltage 36V + Adjustment amount 1V = 37V.

[0108] Collect the working parameters of the solar energy controller: Charging current data: 10A, Charging voltage data: 37V.

[0109] Calculate the digital pulse signals for the target charging current value and the target charging voltage value:

[0110] The digital pulse signal generated according to the target charging current value is: Signal value = 200 (PWM signal).

[0111] The digital pulse signal generated according to the target charging voltage value is: Signal value = 210 (PWM signal).

[0112] Use a signal amplification circuit to amplify the digital pulse signal and generate a control instruction:

[0113] The control instruction includes:

[0114] Adjustment amount of the current regulator: 200 (corresponding control signal).

[0115] Adjustment amount of the voltage regulator: 210 (corresponding control signal).

[0116] After the adjustment strategy is implemented, real-time monitor the output parameters of the solar panel and the working parameters of the solar energy controller: Output voltage: 37V. Output current: 10A. Output power: 370W.

[0117] Energy productivity before implementing the adjustment strategy: 288W. Energy productivity after implementing the adjustment strategy: 370W.

[0118] The comparison result is the energy productivity before and after implementing the adjustment strategy: 370W (after) > 288W (before), indicating that the adjustment strategy effectively improves the energy output.

[0119] By comparing the energy productivity before and after implementation, judge that the adjustment strategy is effective, and confirm the positive impact of the charging current and voltage adjustment on energy generation.

[0120] According to the evaluation results, if it is found that the adjustment effect of the target charging current is not ideal, it is possible to:

[0121] Re-optimize the hyperparameters of the intelligent adjustment model, for example, optimize the learning rate from 0.01 to 0.005 to further improve the model prediction accuracy.

[0122] Improve the model structure, for example, increase the number of neurons in the recurrent layer from 50 to 70 to more effectively capture the temporal relationship of the input features.

[0123] In summary, this embodiment provides a method for local display and energy management of a solar controller. By obtaining the most accurate and timely working parameters and environmental parameters of the solar panel and performing preprocessing, it can provide a solid foundation for subsequent intelligent control and energy management decisions, avoiding control errors and energy waste caused by inaccurate or lagging data. By constructing an intelligent control model based on a recurrent neural network and using advanced algorithms to optimize hyperparameters, the accuracy and adaptability of the control strategy are improved. It can extract key input features from the preprocessed data, thereby outputting a more reasonable control strategy, including precisely adjusting the charging current and voltage. This not only effectively improves the conversion efficiency of solar energy, extends the service life of the solar panel and the storage battery, but also reduces the maintenance cost of the system. The implementation of the local display function enables users to intuitively and conveniently obtain the real-time working status and key parameters of the system on-site at the device. This greatly improves the user's monitoring ability of the solar system, enables users to detect potential problems in a timely manner and take corresponding measures, reduces the likelihood of failures and the repair time, and ensures the stable operation of the system. The control instructions generated according to the control strategy can accurately control the working status of the solar controller, realizing refined management of energy. By intelligently adjusting the charging current and voltage, it can make full use of solar energy resources, store energy to the greatest extent when the light is sufficient, and reasonably release the stored energy when the light is insufficient or the load demand is high, effectively reducing energy waste and improving the energy self-sufficiency of the entire system. Continuously evaluating the effect of the control strategy and optimizing and improving the intelligent control model forms a virtuous feedback loop, enabling the system to continuously adapt to environmental changes and user needs and always maintain the optimal working state. This helps to reduce the dependence on traditional energy, lower carbon emissions, and is of great significance for environmental protection and sustainable development. At the same time, improving the performance and reliability of the solar system will also promote the application and popularization of solar energy technology in a wider range of fields, making a positive contribution to the global energy transition.

[0124] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. One can select some or all of the modules according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

[0125] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0126] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A localization display and energy management method for a solar controller, characterized in that, Including: Collecting the output parameters and environmental parameters of the solar panel in real time, and preprocessing the output parameters and the environmental parameters to obtain preprocessed data; Constructing an intelligent adjustment model of the solar controller based on a recurrent neural network, optimizing the hyperparameters of the intelligent adjustment model using a genetic algorithm, taking the preprocessed data as input, and outputting an adjustment strategy, where the adjustment strategy includes the adjustment amount of the charging current and the adjustment amount of the charging voltage; Generating a control instruction according to the adjustment strategy, adjusting the working parameters of the solar controller through the control instruction, and performing local display; Continuously evaluating the adjustment effect of the adjustment strategy, and optimizing the intelligent adjustment model according to the adjustment effect.

2. The localization display and energy management method of the solar energy controller according to claim 1, characterized in that The output parameters include output voltage, output current, and output power, and the environmental parameters include light intensity, temperature, and humidity.

3. The localization display and energy management method of the solar controller according to claim 1, characterized in that, The process of the preprocessing includes cleaning, screening, and converting the working parameters and the environmental parameters, removing outliers and noise, to obtain preprocessed data.

4. The localization display and energy management method of the solar controller according to claim 1, wherein The process of constructing an intelligent adjustment model of the solar controller based on a recurrent neural network includes: Collecting historical data of the solar panel, where the historical data includes historical working parameters, historical environmental parameters, and corresponding historical adjustment strategies, preprocessing the historical data to obtain historical preprocessed data, and dividing it into a training set and a test set; Extracting input features from the historical preprocessed data, where the input features include historical output voltage features, historical output current features, historical output power features, historical light intensity features, historical temperature features, and historical humidity features; Designing a basic recurrent neural network model, where the basic model includes an input layer, a recurrent layer, and an output layer; the input layer is used to receive the input features and set the dimension of the input layer according to the quantity of the input features; the recurrent layer includes a preset number of neurons, which are used to capture the temporal features in the input features and generate control information; the output layer is used to convert the control information into a continuous adjustment strategy using a linear activation function; Taking the input features as input and the historical adjustment strategy as output, training the basic model, and retaining the model parameters that meet the test accuracy to obtain an intelligent adjustment model.

5. The localization display and energy management method of the solar energy controller according to claim 1, characterized in that, The process of optimizing the hyperparameters of the intelligent adjustment model using a genetic algorithm includes: Taking the learning rate, the number of layers of the recurrent layer, and the number of neurons in the recurrent layer as hyperparameters to be optimized; Setting a search space for each hyperparameter, where the search space represents the value range of the hyperparameter; Randomly generating hyperparameter combinations from the search space as gene encodings, constructing an initial population by generating multiple gene encodings, training the intelligent adjustment model using each gene encoding in the initial population, evaluating the performance index of the intelligent adjustment model on the validation set, where the performance index is the mean square error, determining whether the performance index reaches a preset value, if so, selecting the hyperparameter combination corresponding to the gene encoding with the minimum performance index as the hyperparameter combination of the intelligent adjustment model, otherwise updating the gene encodings in the initial population, and the process includes: Sort the hyperparameter combinations in the initial population according to the performance metrics; Based on the evolutionary mechanism of the genetic algorithm, perform crossover and mutation operations on the gene encoding in the initial population to generate new gene encoding; Calculate the performance metrics of the intelligent adjustment model under the new gene encoding and add them to the initial population.

6. The localization display and energy management method of the solar energy controller according to claim 5, characterized in that The process of performing crossover and mutation operations on the hyperparameter combinations in the initial population includes: According to the sorting of the performance metrics, select the two sets of hyperparameter combinations with the smallest mean squared error as the parents; Set the crossover probability and select the multi-point crossover method to exchange the values of the hyperparameters in the hyperparameter combinations of the parents, and mutate the values of the hyperparameters in the hyperparameter combinations of the parents within the search space according to the preset mutation probability to obtain new hyperparameter combinations.

7. The localization display and energy management method of the solar energy controller according to claim 1, characterized in that, The process of generating control instructions according to the adjustment strategy includes: Collect the working parameters of the solar controller, where the working parameters include charging current data and charging voltage data, and calculate the target charging current value and the target charging voltage value according to the adjustment strategy; Generate a digital pulse signal according to the target charging current value and the target charging voltage value; Amplify the digital pulse signal using a signal amplification circuit to obtain a control instruction, where the control instruction includes the adjustment amount of the current regulator and the adjustment amount of the voltage regulator.

8. The localization display and energy management method of the solar energy controller according to claim 1, characterized in that, The process of performing local display includes: Collect relevant data in real time, where the relevant data includes the output parameters and environmental parameters of the solar panel, the adjustment strategy and control instructions of the solar controller; Sort, transform, and format the relevant data; Connect the display device to the solar controller through a parallel interface; Write a driver program and a display program for the display device according to the performance parameters of the display device and the format of the relevant data, so that the relevant data is displayed according to the preset layout, format, font, and color.

9. The method for local display and energy management of the solar controller according to claim 1, characterized in that, The process of evaluating the effect of the adjustment strategy includes: Collect the output parameters of the solar panel and the working parameters of the solar controller in real time; Calculate the energy productivity according to the output parameters of the solar panel and the working parameters of the solar controller; Evaluate the effect of the adjustment strategy by comparing the energy productivity before implementing the adjustment strategy and the energy productivity after implementing the adjustment strategy.

10. The method for local display and energy management of the solar controller according to claim 1, characterized in that, The ways to optimize the intelligent adjustment model according to the adjustment effect include optimizing the hyperparameters of the intelligent adjustment model and improving the structure of the intelligent adjustment model.

Images