Campus solar energy charging control method and device and storage medium

Abstract

The invention discloses a solar energy charging control method for a campus, and the method comprises the following steps: S1, collecting the historical electricity consumption information of the campus, including the electricity consumption, the electricity consumption peak value and the low ebb value of different time periods, S2, carrying out the deep analysis of the historical electricity consumption data through a data analysis tool and algorithm, and S3, constructing an electricity consumption prediction model through the data analysis result of S2, and carrying out the prediction of the electricity consumption. S4, customizing an energy charging strategy and an electricity utilization plan according to the electricity utilization model constructed in S3, and S5, interacting the energy charging strategy and the electricity utilization plan with the smart power grid, and adjusting the output power and the electricity utilization demand of the campus solar power generation system according to the load condition and the electricity price change of the power grid.

Description

Technical Field

[0001] The invention relates to solar energy equipment, and in particular to a method, equipment and storage medium for charging control of solar energy for campus use. Background Art

[0002] In the campus solar charging control method, electricity demand forecasting is crucial. However, the existing technology has some obvious shortcomings in campus electricity demand forecasting. The existing technology often relies only on historical electricity consumption data for demand forecasting, while ignoring other important factors. For example, campus activity arrangements, course changes, special events, etc. have a significant impact on electricity demand, but these factors are difficult to be fully considered in traditional forecasting methods, resulting in deviations between the forecast results and actual demand. There is a lack of real-time data updates, and many existing forecasting methods cannot promptly incorporate the latest electricity consumption data and environmental change information. The campus' electricity consumption may change rapidly due to seasonal changes, sudden weather changes, equipment failures and other factors. If the data cannot be updated in real time and the forecast model cannot be adjusted, the forecast results will lose timeliness and accuracy. Summary of the invention

[0003] In order to solve the above-mentioned technical problems, the present invention provides a campus solar energy charging control method, device and storage medium.

[0004] The technical solution of the present invention is achieved in this way:

[0005] A solar energy charging control method for campus includes the following steps:

[0006] S1, collect historical electricity consumption information on campus, including electricity consumption in different time periods, peak and valley values; S2, use data analysis tools and algorithms to conduct in-depth analysis of historical electricity consumption data;

[0007] S3, constructs a prediction model for electricity consumption based on the data analysis results of S2;

[0008] S4, customizes charging strategy and power consumption plan based on the power consumption model constructed in S3;

[0009] S5, interacts the charging strategy and power consumption plan with the smart grid, and adjusts the output power and power demand of the campus solar power generation system according to the load conditions and electricity price changes of the grid.

[0010] Preferably, the S1 further comprises the following sub-steps:

[0011] S101, classify and sort the historical electricity consumption data, and subdivide them according to different areas, seasons, and weather conditions on campus, so as to better analyze the impact of the above factors on electricity demand;

[0012] S102, collects information on the course schedule, exam time, and large-scale event plans on campus; these activities will directly affect the electricity demand in specific areas. For example, the electricity consumption of lighting and air conditioning in teaching buildings may increase during exams;

[0013] S103, install environmental monitoring equipment to collect real-time weather data on campus, including temperature, humidity, and light intensity; weather conditions have a great impact on the use of electrical equipment such as air conditioners and lighting. For example, high temperatures can cause a significant increase in air conditioning power consumption;

[0014] S104, monitoring the flow of people on campus can be achieved by installing flow counting devices at the entrance of the campus, the entrance of the teaching building, or at multiple campus parks or building entrances and exits; changes in the number of people will affect the use of lighting and electrical equipment in public areas.

[0015] Preferably, the S2 further comprises the following sub-steps:

[0016] S201, through data visualization technology, the historical electricity consumption data is displayed in the form of charts, so as to observe the electricity consumption pattern more intuitively;

[0017] S202, performing correlation analysis on campus activity information and historical electricity consumption data to find out the relationship between activities and electricity demand; for example, calculating the increase in electricity consumption of teaching buildings during exams to determine the impact of large-scale activities on electricity demand in specific areas;

[0018] S203, analyzing the relationship between real-time weather data and electricity demand; for example, establishing a regression model between temperature and air-conditioning electricity consumption, and determining the changing trend of air-conditioning electricity consumption under different temperatures; in this way, the change of air-conditioning electricity demand can be predicted according to real-time weather conditions.

[0019] Preferably, the step S3 further comprises the following sub-steps:

[0020] S301, select a suitable prediction algorithm based on the characteristics of campus electricity demand; commonly used algorithms include time series analysis, regression analysis, neural network, etc.; time series analysis is suitable for predicting electricity data with obvious time trends; regression analysis can consider the impact of multiple factors on electricity demand; neural network can capture complex nonlinear relationships;

[0021] S302, comparing and evaluating different prediction algorithms, and selecting an algorithm with high prediction accuracy and good stability as the main electricity demand prediction method; the performance of the algorithm can be evaluated by cross-validation, root mean square error and other indicators;

[0022] S303, using the selected prediction algorithm and the collected data to construct a campus electricity demand prediction model; the input of the model may include historical electricity consumption data, campus activity information, real-time environmental data, etc., and the output is the predicted value of electricity demand for a period of time in the future;

[0023] S304, adjusting and optimizing the parameters of the prediction model to improve the prediction accuracy; this can be achieved by adjusting the parameters of the algorithm, adding data features, improving the model structure, etc.; for example, different time window lengths, different combinations of regression variables, etc. can be tried to find the optimal model parameter settings.

[0024] Preferably, the S4 further comprises the following sub-steps:

[0025] S401, formulate a reasonable charging control strategy based on the power demand forecast result and the solar energy resource monitoring data; for example, store the excess solar energy during the low power demand period; and give priority to using the stored energy or adjusting the output power of the solar power generation system during the peak power demand period;

[0026] S402, considering the differences in electricity demand in different areas, formulate personalized charging control strategies; for example, for teaching buildings with high electricity consumption, the number of solar panels installed can be increased or the battery storage capacity can be improved; for areas with low electricity consumption, the configuration of solar power generation systems can be appropriately reduced;

[0027] S403, as time goes by and the actual power consumption changes, the power demand forecast results and charging control strategy are adjusted in real time; if there is a large deviation between the actual power demand and the forecast value, the operating state of the solar power generation system can be adjusted in time to ensure a stable supply of electricity for the campus;

[0028] S404 works in collaboration with the campus equipment management system to arrange equipment maintenance and overhaul plans in advance based on electricity demand forecast results to avoid the impact of equipment failures on electricity demand; at the same time, it can dynamically adjust the equipment operation mode according to changes in electricity demand to improve energy efficiency.

[0029] A campus solar energy charging device includes at least one processor, at least one memory and computer program instructions stored in the memory. When the computer program instructions are executed by the processor, the campus solar energy charging control method is implemented.

[0030] A storage medium implements the above-mentioned campus solar energy charging control method when computer program instructions are executed by a processor.

[0031] The present invention solves the shortcoming of the prior art that it often only relies on historical electricity consumption data for demand forecasting while ignoring other important factors. It also solves the impact of weather season and climate on predicted electricity demand, and achieves the purpose of accurate prediction and reducing waste. Through accurate electricity demand forecasting, it can reasonably arrange the storage and use of solar power generation according to actual needs, avoid excessive storage or insufficient supply of energy, greatly improve the utilization efficiency of solar energy, and achieve the effect of dynamic adjustment and optimization configuration. As the actual electricity consumption and solar energy resources change, the charging control strategy is adjusted in real time to ensure that the solar power generation system always operates in the optimal state. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a schematic diagram of a campus solar energy charging control method of the present invention. DETAILED DESCRIPTION

[0033] In order to further explain the technical means and effects taken by the present invention to achieve the predetermined invention purpose, exemplary embodiments will be described in detail here, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are only examples of methods and systems consistent with some aspects of the present application as detailed in the attached claims.

[0034] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "the", and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0035] Example 1

[0036] The present invention provides a campus solar energy charging control method, comprising the following steps:

[0037] S1, collect historical electricity consumption information of the campus, including electricity consumption in different time periods, peak values ​​and valley values; S101, classify and sort the historical electricity consumption data, and subdivide them according to different areas, seasons and weather conditions of the campus, so as to better analyze the impact of the above factors on electricity demand;

[0038] S102, collects information on the course schedule, exam time, and large-scale event plans on campus; these activities will directly affect the electricity demand in specific areas. For example, the electricity consumption of lighting and air conditioning in teaching buildings may increase during exams;

[0039] S103, install environmental monitoring equipment to collect real-time weather data on campus, including temperature, humidity, and light intensity; weather conditions have a great impact on the use of electrical equipment such as air conditioners and lighting. For example, high temperatures can cause a significant increase in air conditioning power consumption;

[0040] S104, monitoring the flow of people on campus can be achieved by installing flow counting devices at the entrance of the campus, the entrance of the teaching building, or at multiple campus parks or building entrances and exits; changes in the number of people will affect the use of lighting and electrical equipment in public areas.

[0041] S105, associates the information collected by S101, S102, S103 and S104, that is, associates the historical electricity consumption information of the day with the information of campus courses, campus weather, and personnel, and calculates the electricity consumption of a single person in each different building on a single day.

[0042] S2, using data analysis tools and algorithms to conduct in-depth analysis of historical electricity consumption data;

[0043] S201, using data visualization technology, the calculation result of step S105 is displayed in the form of a chart, so as to observe the power consumption pattern more intuitively;

[0044] S202, analyzing campus activity information and historical electricity consumption data to find out the relationship between activities and electricity demand; for example, counting the increase in electricity consumption in teaching buildings during exams to determine the impact of large-scale activities on electricity demand in specific areas;

[0045] S203, analyzing the relationship between real-time weather data and electricity demand; for example, establishing a regression model between temperature and air-conditioning electricity consumption, and determining the changing trend of air-conditioning electricity consumption under different temperatures; in this way, the change of air-conditioning electricity demand can be predicted according to real-time weather conditions.

[0046] S3, constructs a prediction model for electricity consumption based on the data analysis results of S2;

[0047] S301, selecting a prediction algorithm according to the characteristics of campus electricity demand;

[0048] S302, comparing and evaluating different prediction algorithms, and selecting an algorithm with high prediction accuracy and good stability as the main electricity demand prediction method; the performance of the algorithm can be evaluated by cross-validation, root mean square error and other indicators;

[0049] S303, using the selected prediction algorithm and the collected data to construct a campus electricity demand prediction model; the input of the model may include historical electricity consumption data, campus activity information, real-time environmental data, etc., and the output is the predicted value of electricity demand for a period of time in the future;

[0050] S304, adjust and optimize the parameters of the prediction model to improve the prediction accuracy; this can be achieved by adjusting the parameters of the algorithm, adding data features, improving the model structure, etc.; for example, different time window lengths, different combinations of regression variables, etc. can be tried to find the optimal model parameter settings

[0051] S4, customizes charging strategy and power consumption plan based on the power consumption model constructed in S3;

[0052] S401, formulate a reasonable charging control strategy based on the power demand forecast result and the solar energy resource monitoring data; for example, store the excess solar energy during the low power demand period; and give priority to using the stored energy or adjusting the output power of the solar power generation system during the peak power demand period;

[0053] S402, considering the differences in electricity demand in different areas, formulate personalized charging control strategies; for example, for teaching buildings with high electricity consumption, the number of solar panels installed can be increased or the battery storage capacity can be improved; for areas with low electricity consumption, the configuration of solar power generation systems can be appropriately reduced;

[0054] S403, as time goes by and the actual power consumption changes, the power demand forecast results and charging control strategy are adjusted in real time; if there is a large deviation between the actual power demand and the forecast value, the operating state of the solar power generation system can be adjusted in time to ensure a stable supply of electricity for the campus;

[0055] S404 works in collaboration with the campus equipment management system to arrange equipment maintenance and overhaul plans in advance based on electricity demand forecast results to avoid the impact of equipment failures on electricity demand; at the same time, it can dynamically adjust the equipment operation mode according to changes in electricity demand to improve energy efficiency.

[0056] S5, interacts the charging strategy and power consumption plan with the smart grid, and adjusts the output power and power demand of the campus solar power generation system according to the load conditions and electricity price changes of the grid.

[0057] Preferably, the prediction algorithm is an ARIMA algorithm. By analyzing the historical electricity consumption data, the general formula of (ARIMA (p, d, q)) is:

[0058]

[0059] Y tAmong them, is the observed value at time t (electricity consumption data).

[0060] c is a constant term.

[0061] p is the order of the autoregressive term, which means considering the impact of observations at p time points in the past on the current value.

[0062] is the autoregressive coefficient, i=1, 2, ....p.

[0063] Y t-i is the observed value at time ti.

[0064] d is the number of differences, which is used to make the time series stationary.

[0065] q is the order of the moving average term, which means that the impact of the random error term at the past q time points on the current value is considered.

[0066] θ i is the moving average coefficient, i=1, 2, ....q.

[0067] ∈ t-j is the random error term at time tj.

[0068] ∈ t is a random error term at time t, usually assumed to be white noise.

[0069] In the present invention, based on the known past power consumption data, Yt, Yt-1, Yt-2, ..., and other collected information parameters, the estimated value of the random error term is calculated. Initially, for ∈ t+i (i=1,2,....h), initialized to 0. According to the formula Calculate the forecast values ​​step by step.

[0070] Assume that p=2,q=1 has been determined, and the parameter c=10 has been estimated. θ1=0.2, given the past electricity consumption data Y t-1 =50, Y t-2 =45, and the previously calculated ∈t -1 =2, now we need to predict Y t+1 , first calculate Here ∈t is assumed to be O, and then calculated

[0071] If at this time

[0072] Knowing that Yt=48, we can calculate The value of .

[0073] Preferably, in step S1, the data is preprocessed, including checking the completeness and accuracy of the data, and processing missing values ​​and outliers.

[0074] Preferably, in step S2, the ACF and PACF of the historical electricity consumption data are calculated, and the order ranges of the autoregressive term and the moving average term are preliminarily determined based on their graphical features.

[0075] In the ARIMA model, the random error term ∈t is set as white noise, which generally does not contain specific known parameters. The random error term itself usually does not contain directly determinable parameters, but its statistical characteristics are considered in the estimation and analysis of the overall model.

[0076] Furthermore, for large-scale data sets and complex model calculations, we can consider using parallel computing technology, such as distributed computing frameworks or GPU acceleration. This can greatly shorten the time for model training and prediction and improve efficiency.

[0077] Furthermore, the model can be compressed and optimized to reduce the number of model parameters and computational complexity. For example, pruning techniques can be used to remove redundant parameters in the model, or quantization techniques can be used to reduce the storage and computational requirements of the model. This can not only improve the operating efficiency of the model, but also improve the generalization ability of the model to a certain extent.

[0078] A campus solar energy charging device comprises a solar energy charging device for realizing the campus solar energy charging control method.

[0079] A storage medium includes computer program instructions that, when executed by a processor, implement the above-mentioned campus solar energy charging control method.

[0080] Example 2

[0081] This embodiment is a campus solar energy charging control method, which has the following characteristics different from the first embodiment: data collection and preprocessing optimization

[0082] Electricity consumption data collection in different areas: Different areas on campus have different functions and usage characteristics, which leads to different electricity consumption patterns. For example, the teaching building consumes more electricity during class time, mainly for lighting, multimedia equipment, etc.; the dormitory consumes more electricity at night and during rest time, which may involve lighting, computers, electrical equipment, etc. By collecting electricity consumption data in different areas, we can more accurately capture the electricity consumption patterns of these specific areas, provide more detailed information for the model, and thus improve the prediction accuracy of the overall campus electricity demand.

[0083] Analyzing electricity consumption data in different areas can also identify potential energy-saving opportunities. For example, if it is found that the electricity consumption in a certain area is abnormally high, further investigation can be conducted to find out the cause, which may be aging equipment, improper use or waste, and then corresponding energy-saving measures can be taken.

[0084] Special equipment or activity power consumption data collection: Special equipment on campus, such as laboratory equipment and large computers, usually has high power demand and irregular usage time. Collecting power consumption data of these devices can help the model better predict power demand fluctuations under special circumstances, so as to prepare for solar charging in advance.

[0085] Large-scale events, such as sports games and cultural performances, will increase temporary electricity demand during a specific period of time. By understanding the time and electricity demand of these events, you can adjust the output power of the solar power generation system or start the backup power supply in advance to ensure power supply during the event.

[0086] Advanced Data Preprocessing:

[0087] Combining multiple stationary test methods, the ADF test is a commonly used stationary test method, but it may misjudge in some cases. The KPSS test tests the stationarity of the data from another perspective, and combined with the ADF test can improve the accuracy of the judgment. For example, if the ADF test shows that the data is not stationary, but the KPSS test shows that the data is stationary, then it is necessary to further analyze the characteristics of the data, and different difference methods or models may be needed to process it.

[0088] Choosing the appropriate difference method according to the specific situation of the data can better make the data stable. Seasonal difference can handle data with seasonal characteristics, and first-order difference can handle data with trends. Combining these two difference methods can be more flexible to adapt to different types of data.

[0089] Traditional interpolation and correction methods may not be accurate enough for missing values ​​and outliers, especially when the data is complex or there are a large number of missing values. Using machine learning methods such as random forest or K nearest neighbor algorithms can predict missing values ​​based on other features of the data and improve data integrity.

[0090] For outliers, machine learning methods can determine the rationality of outliers by analyzing the distribution and pattern of data. If the outliers are caused by data entry errors or sensor failures, they can be corrected; if the outliers are reasonable exceptions caused by special circumstances, they can be marked separately and their impact can be considered in the model.

[0091] Data normalization: Data of different variables may have different scales and units, which will affect model training and comparison. Normalization can convert data to the same scale, making the weights of different variables in the model more reasonable. For example, Min-Max normalization can map data to the [0, 1] interval, and Z-SCore normalization can make data have zero mean and unit variance.

[0092] Model building and optimization

[0093] Dynamic Model Selection:

[0094] Switch models based on data characteristics. Different time series data may have different characteristics, and a single ARIMA model may not be suitable for all situations. When the data shows obvious seasonality and trend, the SARIMA model or the Holt-Winters exponential smoothing model may be more suitable. For example, campus electricity demand may vary significantly in different seasons, and the SARIMA model can better capture this seasonal variation.

[0095] Combining machine learning algorithms for model selection can fully utilize the advantages of different models. For example, using stacking or blending methods to combine the prediction results of multiple different time series models can improve the accuracy and stability of the prediction. By weighted averaging or voting on the prediction results of different models, the error and uncertainty of a single model can be reduced.

[0096] Establish a model library for automatic selection. Establishing a model library can easily manage and compare different time series models. Regularly evaluate the performance of different models on historical data, and automatically select the best model for prediction based on the evaluation results. For example, you can use cross-validation or other evaluation indicators to compare the performance of different models, and select the model that performs best on historical data for current predictions.

[0097] As data is constantly updated and changed, the performance of the model may also change. Therefore, it is necessary to regularly re-evaluate and select the model to ensure that the model always maintains good predictive ability.

[0098] Advanced parameter optimization algorithms. Traditional grid search and random search methods are less efficient when the parameter space is large. Advanced parameter optimization algorithms such as particle swarm optimization (PSO) and simulated annealing can search in a larger parameter space and find the optimal model parameter combination more quickly. These algorithms simulate physical processes or biological behaviors in nature and continuously adjust parameter values ​​to find the optimal solution.

[0099] The Bayesian optimization method uses prior knowledge and historical data to optimize model parameters more efficiently. By building a probability distribution model for the parameters, the prior distribution is continuously updated based on historical data, gradually approaching the optimal parameters. This method can find a better combination of parameters within a smaller number of iterations, improving the accuracy and efficiency of parameter estimation.

[0100] Considering the time variability of parameters, model parameters may change as time goes by and data is updated. For example, the electricity consumption behavior of a campus may change with changes in teaching arrangements, student numbers, equipment updates, and other factors. Therefore, considering the time variability of parameters and using online learning algorithms to update model parameters in real time can better adapt to the dynamic changes of data.

[0101] Online learning algorithms can continuously adjust model parameters based on new data, so that the model can reflect the latest changes in data in a timely manner. This method can improve the real-time and adaptability of the model and better respond to emergencies and changes.

[0102] In-depth analysis of external factors shows that in addition to common factors such as temperature and season, factors such as changes in population density around the campus and urban development planning may also have potential impacts on campus electricity demand. For example, an increase in surrounding population density may lead to an increase in commercial activities around the campus, thereby affecting campus electricity demand; infrastructure construction in urban development planning may affect the stability of campus power supply.

[0103] Considering these external factors can enable the model to more comprehensively understand the reasons for changes in electricity demand and improve the accuracy of prediction. By analyzing the relationship between external factors and electricity demand, possible changes in electricity demand can be predicted in advance, providing a more accurate basis for solar charging control.

[0104] Deep learning methods extract relationships, deep learning methods, such as convolutional neural networks (CNN) or long short-term memory networks (LSTM), can automatically extract the complex relationship between external factors and electricity demand. These methods have powerful feature extraction capabilities and can handle high-dimensional data and complex nonlinear relationships.

[0105] By using external factors as input features and inputting them into the deep learning model together with historical electricity consumption data, the model can learn the impact of external factors on electricity demand. For example, using LSTM can capture long-term dependencies in time series data and better handle the impact of external factors on electricity demand over time.

[0106] By building a dynamic model of external factors, predicting future temperature changes based on weather forecast data and incorporating it into the electricity demand forecast model, the impact of temperature on electricity demand can be considered in advance. For example, before the arrival of high temperatures in summer, the output power of the solar power generation system can be adjusted or the capacity of the energy storage equipment can be increased in advance to meet the electricity demand of equipment such as air conditioners.

[0107] Establishing a dynamic model of external factors can improve the foresight and accuracy of the forecast. By continuously updating the forecast values ​​of external factors, the power demand forecast and solar charging control strategy can be adjusted in a timely manner to better adapt to changes in the external environment.

[0108] Model evaluation and validation optimization

[0109] Introducing more evaluation indicators, traditional indicators such as mean absolute error (MAE) and root mean square error (RMSE) mainly measure the absolute error and variance between the predicted value and the actual value. Introducing indicators such as mean absolute percentage error (MAPE), symmetric mean absolute percentage error (SMAPE), and Theil inequality coefficient can evaluate the performance of the model from different angles.

[0110] MAPE can measure the relative error between the predicted value and the actual value, and has better comparability for data of different magnitudes; SMAPE solves the calculation problem of MAPE when the actual value is zero to a certain extent; Theil inequality coefficient can measure the proportional difference between the predicted value and the actual value, reflecting the predictive stability of the model.

[0111] Visual evaluation, drawing a comparison chart of predicted values ​​and actual values, error distribution chart and other visual charts can intuitively observe the prediction effect and error distribution of the model. The comparison chart can show the prediction accuracy of the model in different time periods and whether there is a systematic deviation.

[0112] The error distribution plot can show the distribution of forecast errors and help determine whether the model's errors meet assumptions such as normal distribution. If the error distribution is skewed or kurtotic, further adjustments to the model or improved data processing methods may be required.

[0113] Cross-validation and robustness test:

[0114] Strict cross-validation methods, leave-one-out cross-validation (LOOCV) and time series cross-validation can more strictly evaluate the generalization ability of the model. Leave-one-out cross-validation only leaves one sample as the test set each time, and the remaining samples are used as the training set for multiple validations, which can make full use of limited data for model evaluation. Time series cross-validation takes into account the sequential nature of time series data, divides the data into multiple subsets in chronological order, and performs cross-validation. This method can better simulate the situation in actual applications and avoid data leakage problems.

[0115] Robustness test and hypothesis test, the robustness test of the model can test the performance of the model under different data subsets, different parameter settings and different external factors. For example, randomly extract different data subsets for multiple verifications to observe whether the prediction results of the model are stable.

[0116] Introducing hypothesis testing methods, such as t-test or F-test, can test whether the difference between the model's predicted results and the actual values ​​is statistically significant. If the difference is statistically significant, it is necessary to further optimize the model or adjust the data processing method to improve the accuracy of the model.

[0117] Application and decision support optimization

[0118] Establishing a real-time data monitoring system and collecting campus electricity consumption data and solar energy resource data in real time can allow the model to understand the current electricity demand and solar power generation in a timely manner. Through the sensor network and data acquisition system, real-time monitoring of electricity consumption and solar power generation systems in various areas of the campus can be achieved.

[0119] Based on real-time data, the model prediction is continuously updated to achieve real-time electricity demand prediction and solar charging control adjustment. For example, when electricity demand suddenly increases or solar power generation is insufficient, the model can immediately adjust the output power of the solar power generation system or start the backup power supply to ensure the stability of power supply.

[0120] Automated decision support system, the development of an automated decision support system can automatically generate solar charging control strategy recommendations based on the model's prediction results and real-time data. For example, when it is predicted that the peak of electricity demand is approaching, the system can automatically adjust the output power of the solar power generation system, optimize the charging and discharging strategy of the energy storage device, or start the backup power supply.

[0121] Automated decision support systems can improve the efficiency and response speed of energy management, reduce the need for manual intervention, and reduce management costs. At the same time, the system can provide a variety of strategy options according to different situations, allowing decision makers to adjust according to actual conditions.

[0122] Establish an early warning mechanism. When the model predicts that the power demand may exceed the supply capacity of the solar power generation system or other abnormal situations occur, timely warnings can be issued to allow relevant personnel to take corresponding measures. For example, adjust the power consumption plan in advance, start the emergency power supply, and notify users to save electricity.

[0123] The early warning mechanism can improve the safety and reliability of energy management and avoid problems such as equipment damage and interruption of teaching and scientific research activities due to insufficient power supply.

[0124] The long-term prediction capability of the model can provide decision support for campus energy planning and sustainable development. According to the future electricity demand trend and solar energy resource changes predicted by the model, long-term energy development strategies and investment plans can be formulated.

[0125] For example, if it is predicted that electricity demand will continue to grow in the future, you can consider increasing the capacity of the solar power generation system, building energy storage facilities, or working more closely with the power grid. At the same time, according to the changes in solar energy resources, the layout and installation angle of the solar power generation system should be reasonably planned to improve the efficiency of solar energy utilization. Considering the synergy of energy systems, the campus energy system usually includes multiple components such as solar power generation, grid power supply, and energy storage equipment. Considering the synergy of these energy systems can formulate a more optimized energy supply strategy.

[0126] For example, through interaction with the power grid, excess electricity can be sold to the grid when there is excess solar power generation, and electricity can be purchased from the grid when there is insufficient solar power generation, thus achieving optimal energy allocation. At the same time, reasonable configuration of energy storage equipment can improve energy utilization efficiency and stability and reduce dependence on the power grid.

[0127] Conducting user behavior analysis and formulating personalized energy-saving suggestions and incentives based on electricity demand forecast results and user behavior data can encourage teachers and students to actively participate in energy conservation and sustainable development actions. For example, by analyzing users' electricity usage habits, providing users with targeted energy-saving suggestions, such as reasonably arranging electricity usage time and shutting down unnecessary electrical equipment.

[0128] At the same time, an incentive mechanism can be established to reward users who save electricity and improve their energy-saving awareness and enthusiasm. Conducting user behavior analysis can manage energy from the demand side and achieve sustainable use of energy.

Claims

1. A solar energy charging control method for campus, characterized in that: The steps include: S1: Collect historical electricity consumption information on campus, including electricity consumption in different time periods, peak and valley values; S2: Use data analysis tools and algorithms to conduct in-depth analysis of historical electricity consumption data; S3, builds the electricity consumption prediction model 2 based on the data analysis results of S2 S4, customizes charging strategy and power consumption plan based on the power consumption model constructed in S3 S5, interacts the charging strategy and power consumption plan with the smart grid, and adjusts the output power and power demand of the campus solar power generation system according to the load conditions and electricity price changes of the grid.

2. A campus solar energy charging control method according to claim 1, characterized in that: The step S1 also includes the following sub-steps: S101, classify and sort out the historical electricity consumption data, and subdivide it according to different areas, seasons, and weather conditions on campus. S102, collect information on campus course schedules, exam times, and large-scale event plans. S103, install environmental monitoring equipment to collect real-time weather data on campus, including temperature, humidity, and light intensity. S104, monitoring the flow of people on campus can be achieved by installing flow counting devices at the entrance of the campus or at the entrance of the teaching building at multiple campus parks or building entrances and exits.

3. A campus solar energy charging control method according to claim 1, characterized in that: The step S2 further comprises the following sub-steps: S201, through data visualization technology, the historical electricity consumption data is displayed in the form of charts. S202, conduct correlation analysis between campus activity information and historical electricity consumption data to find out the relationship between activities and electricity demand. S203, analyzing the relationship between real-time weather data and electricity demand.

4. A campus solar energy charging control method according to claim 1, characterized in that: The step S3 further comprises the following sub-steps: S301, select a suitable prediction algorithm based on the characteristics of campus electricity demand S302, compare and evaluate the prediction algorithms, and evaluate the performance of the algorithms through indicators such as cross validation and root mean square error. S303, using the selected prediction algorithm and the collected data, construct a campus electricity demand prediction model 2. The input of the model includes historical electricity consumption data, campus activity information, and real-time environmental data, and the output is the predicted value of electricity demand for a period of time in the future. S304, adjusting and optimizing the parameters of the prediction model to improve the prediction accuracy by adjusting the parameters of the algorithm, adding data features, and improving the model structure.

5. A campus solar energy charging control method according to claim 1, characterized in that: The step S4 also includes the following sub-steps: S401, based on the electricity demand forecast results and combined with the solar energy resource monitoring data, formulate a reasonable charging control strategy II S402, considering the differences in electricity demand in different areas, formulate personalized charging control strategies. For teaching buildings with high electricity consumption, increase the number of solar panels installed or increase the battery storage capacity. For areas with low electricity consumption, reduce the configuration of solar power generation systems. S403, as time goes by and the actual power consumption changes, the power demand forecast results and charging control strategy are adjusted in real time. If there is a large deviation between the actual power demand and the predicted value, the operating status of the solar power generation system can be adjusted in time to ensure a stable supply of campus electricity. S404 works in collaboration with the campus equipment management system to arrange equipment maintenance and overhaul plans in advance based on electricity demand forecast results to avoid the impact of equipment failures on electricity demand; at the same time, it can dynamically adjust the equipment operation mode according to changes in electricity demand to improve energy efficiency.

6. A campus solar energy charging control method according to claim 4, characterized in that: The prediction algorithm of step 302 is the ARIMA algorithm, which is performed by analyzing historical electricity consumption data.

7. A solar charging device for campus, characterized in that: The system comprises at least one processor, at least one memory and computer program instructions stored in the memory, and when the computer program instructions are executed by the processor, the methods described in 1 to 6 above are implemented.

8. A storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the methods 1 to 6 described above are implemented.

Images