Offshore buoy energy consumption optimization method, device and equipment based on photovoltaic power generation

Abstract

The invention discloses an offshore buoy energy consumption optimization method, device and equipment based on photovoltaic power generation. The method comprises the following steps: carrying out importance grade grading on each power consumption element of the buoy to obtain an importance grade sequence; historical data of the buoy photovoltaic generating capacity are input into the SARIMA model, the linear trend and seasonal characteristics of the buoy photovoltaic generating capacity are captured, and a power generation prediction model is obtained; taking a residual error sequence between the daily predicted value of the buoy generating capacity and the actual value of the daily generating capacity and historical weather data as input, and taking the residual error of the generating capacity predicted value as output to train an LSTM model to obtain a residual error prediction model; predicting the power generation capacity of the buoy in a future preset period through the power generation prediction model and the residual prediction model; and distributing power consumption to the buoy in a future preset period according to a prediction result, and setting a working mode of the power consumption element according to the importance level sequence and the power consumption distributed by the buoy. According to the scheme, the operation continuity and stability of the offshore buoy can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of electric energy distribution, and in particular to a method, device and equipment for optimizing energy consumption of offshore buoys based on photovoltaic power generation. Background Art

[0002] During the operation of offshore buoys, their photovoltaic power generation capacity is often affected by environmental factors such as weather changes, seasonal variations, and complex sea conditions. This leads to high instability in photovoltaic power generation and consequently power shortages. These instabilities can manifest as short-term fluctuations or long-term power shortages, seriously impacting the proper operation of the buoy's power-consuming components and the continuity of data collection. Consequently, the buoy's navigation, navigation assistance, early warning, and alert services for ships at sea cannot be reliably provided, creating safety risks. Therefore, it is necessary to rationally distribute the power supply to offshore buoys to ensure stable operation of the offshore buoy system.

[0003] Existing methods for distributing power to offshore buoys primarily include: 1. Fixed power allocation. 2. Threshold-based control (monitoring the battery charge level of the buoy and applying simple threshold control based on the monitoring results). These solutions can improve the operational stability of offshore buoys to a certain extent. However, the fixed power allocation method cannot flexibly adjust the operating status of the equipment based on the actual power generation of the buoy, resulting in reduced operational sustainability. The threshold-based control method is less adaptable to unexpected situations and can easily lead to equipment downtime or data loss.

[0004] Therefore, when using the existing technology to distribute the electrical energy of the offshore buoy, there is a problem that the operation continuity and stability of the offshore buoy are not high. Summary of the Invention

[0005] Based on this, it is necessary to provide an offshore buoy energy consumption optimization method, device and equipment based on photovoltaic power generation to address the technical problem of low continuity and stability of offshore buoy operation when the existing technology distributes electrical energy to offshore buoys.

[0006] The present invention adopts the following technical solutions: In a first aspect, the present invention provides a method for optimizing energy consumption of an offshore buoy based on photovoltaic power generation, the method comprising: Acquiring historical data on photovoltaic power generation of a buoy in a target sea area and historical weather data of the target sea area; grading the importance of each power-consuming component of the buoy to obtain an importance level sequence of each power-consuming component; Inputting the historical data of the buoy's photovoltaic power generation into a SARIMA model, using the SARIMA model to capture the linear trend and seasonal characteristics of the buoy's photovoltaic power generation, and determining the parameters of the SARIMA model based on the linear trend and seasonal characteristics to obtain a power generation prediction model; calculating the residual between the daily predicted value of the buoy's photovoltaic power generation output by the power generation prediction model and the actual value of the power generation on that day to obtain a residual sequence, using the residual sequence and the historical weather data as input, and using the residual of the photovoltaic power generation prediction value as output to train an LSTM model, and using the LSTM model to learn the nonlinear relationship and long-term dependency between the buoy's photovoltaic power generation and historical weather during the training process to obtain a residual prediction model; The photovoltaic power generation of the buoy in a future preset period is predicted by the power generation prediction model and the residual prediction model to obtain a prediction result; The power consumption of the buoy is allocated on each day in a future preset period according to the prediction result, and the working mode of each power-consuming component is set according to the importance level sequence and the power consumption allocated to the buoy on each day in the future preset period.

[0007] Furthermore, the photovoltaic power generation of the buoy in a future preset period is predicted by the power generation prediction model and the residual prediction model to obtain a prediction result, which specifically includes: Obtaining weather data for the future preset period; Inputting historical data of the photovoltaic power generation of the buoy into the power generation prediction model according to the time length of the future preset period to obtain a predicted value of the photovoltaic power generation of the buoy in the future preset period; inputting weather data of the future preset period into the residual prediction model to obtain a residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period; The prediction result is obtained based on the predicted value of the photovoltaic power generation of the buoy in the future preset period and the residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period.

[0008] Furthermore, the power consumption of the buoy is allocated every day in a future preset period according to the prediction result, specifically including: Determining, based on the prediction results, whether there is a day in a future preset period when the battery of the buoy can be fully charged; If there is a k-th day when the battery of the buoy can be fully charged, calculate the total power generation of the buoy from the first day to the k-th day in the future preset period, where k is a positive integer; Calculating the power consumption allocated to the buoy on each day within a future preset period based on the battery power and the total power generation; For each day within a future preset period, determine whether the sum of the buoy's daily power generation and the battery power is less than the power consumption allocated for the day. If the sum of the buoy's daily power generation and the battery power is less than the power consumption allocated for the day, the power consumption allocated for the day will be determined as the sum of the buoy's daily power generation and the battery power.

[0009] Furthermore, the operating mode of each power-consuming component is set according to the importance level sequence and the power consumption allocated to the buoy every day in a future preset period, specifically including: Dividing the working modes of the power-consuming components of the buoy to obtain multiple working modes of the power-consuming components, each working mode corresponding to a different working frequency and working duration; assigning points to each of the power-consuming components in each operating mode according to the importance level sequence; Determine all combination schemes of the power-consuming components under different operating modes by an exhaustive method, and determine as a target scheme a scheme in which the sum of the power consumption of the power-consuming components among all the combination schemes is less than or equal to the power consumption allocated for each day; According to the scores of each power-consuming component in each working mode, the total scores corresponding to each solution in the target solution are obtained, the solution with the largest total score in the target solution is determined as the optimal solution, and the working mode of each power-consuming component is set according to the optimal solution.

[0010] Furthermore, the prediction result is obtained based on the predicted value of the photovoltaic power generation of the buoy in the future preset period and the residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period, which specifically includes: The photovoltaic power generation prediction value and the residual of the photovoltaic power generation prediction value are summed in a weighted average manner to obtain the prediction result.

[0011] Furthermore, before inputting the historical data set into the SARIMA model, the method further includes performing missing value processing, outlier processing and normalization processing on the historical data set.

[0012] Furthermore, the weather data includes temperature, radiometry and cloud cover.

[0013] In a second aspect, the present invention provides an offshore buoy energy consumption optimization device based on photovoltaic power generation, comprising: an acquisition module for acquiring historical data on photovoltaic power generation of the buoy in the target sea area and historical weather data of the target sea area; grading the importance of each power-consuming component of the buoy to obtain an importance level sequence of each power-consuming component; A construction module is provided for inputting the historical data of the buoy's photovoltaic power generation into a SARIMA model, using the SARIMA model to capture the linear trend and seasonal characteristics of the buoy's photovoltaic power generation, and determining the parameters of the SARIMA model based on the linear trend and seasonal characteristics to obtain a power generation prediction model; calculating the residual between the daily predicted value of the buoy's photovoltaic power generation output by the power generation prediction model and the actual value of the power generation on that day to obtain a residual sequence, using the residual sequence and the historical weather data as input, and using the residual of the photovoltaic power generation prediction value as output to train an LSTM model, and using the LSTM model to learn the nonlinear relationship and long-term dependence between the buoy's photovoltaic power generation and historical weather during the training process to obtain a residual prediction model; A prediction module, configured to predict the photovoltaic power generation of the buoy in a future preset period by using the power generation prediction model and the residual prediction model to obtain a prediction result; The energy consumption optimization module is used to allocate the power consumption of the buoy on each day in the future preset period according to the prediction results, and to set the working mode of each power-consuming component according to the importance level sequence and the power consumption allocated to the buoy on each day in the future preset period.

[0014] The present invention provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for optimizing energy consumption of offshore buoys based on photovoltaic power generation is implemented.

[0015] At least one technical solution adopted by the present invention can achieve the following beneficial effects: the present invention obtains historical data on photovoltaic power generation of buoys in the target sea area and historical weather data of the target sea area; and classifies the importance of each power-consuming component of the buoy to obtain an importance level sequence of each power-consuming component; The historical data of the buoy's photovoltaic power generation is input into a SARIMA model, and the SARIMA model is used to capture the linear trend and seasonal characteristics of the buoy's photovoltaic power generation. The parameters of the SARIMA model are determined based on the linear trend and seasonal characteristics to obtain a power generation prediction model; the residual between the daily predicted value of the buoy's photovoltaic power generation output by the power generation prediction model and the actual value of the power generation on that day is calculated to obtain a residual sequence, and the residual sequence and historical weather data are used as input, and the residual of the photovoltaic power generation prediction value is used as output to train an LSTM model. During the training process, the LSTM model is used to learn the nonlinear relationship and long-term dependence between the buoy's photovoltaic power generation and historical weather to obtain a residual prediction model; the photovoltaic power generation of the buoy in a future preset period is predicted using the power generation prediction model and the residual prediction model to obtain a prediction result; the power consumption of the buoy is allocated to each day in the future preset period based on the prediction result, and the working mode of each power-consuming component is set according to the importance level sequence and the power consumption allocated to the buoy on each day in the future preset period. This approach uses the SARIMA model to capture the linear trends and seasonal characteristics of a buoy's photovoltaic power generation, and the LSTM model to learn the nonlinear relationship and long-term dependency between the buoy's photovoltaic power generation and historical weather conditions, improving the accuracy of photovoltaic power generation forecasts for offshore buoys. Based on the buoy's photovoltaic power generation over a preset future period, power consumption is allocated to each power-consuming component within the buoy. The operating mode of each power-consuming component is then set based on its importance level and power consumption, improving the continuity and stability of offshore buoy operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0017] Figure 1 Flowchart of the method for optimizing energy consumption of offshore buoys based on photovoltaic power generation provided by the present invention; Figure 2 A flow chart of another method for optimizing energy consumption of offshore buoys based on photovoltaic power generation provided by the present invention; Figure 3 A comparison curve of the predicted power generation and actual power generation of the offshore buoy provided by the present invention in the next two weeks; Figure 4 Schematic diagram of the offshore buoy energy consumption optimization device based on photovoltaic power generation provided by the present invention; Figure 5 A schematic diagram of a computer device for implementing a method for optimizing energy consumption of an offshore buoy based on photovoltaic power generation provided by the present invention. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present invention and corresponding drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] The server mentioned in the present invention can be a server installed on a business platform, or a device such as a desktop computer or laptop computer capable of executing the solution of the present invention. For ease of explanation, the following description will only use the server as the execution entity. The following, combined with the accompanying drawings, details the technical solutions provided by various embodiments of the present invention.

[0020] refer to Figure 1 The method for optimizing energy consumption of offshore buoys based on photovoltaic power generation in the present invention specifically includes the following steps: S10: Obtain historical data on photovoltaic power generation of the buoy in the target sea area, as well as historical weather data of the target sea area; classify the importance of each power-consuming component of the buoy to obtain an importance level sequence of each power-consuming component.

[0021] In this embodiment, the target sea area refers to the sea area where the buoy is located, and the weather data includes but is not limited to temperature, radiation, and cloud cover. The importance level sequence of each power-consuming component is the working priority sequence of each power-consuming component determined according to the actual application scenario.

[0022] S20: Input the historical data of buoy photovoltaic power generation into the Seasonal Autoregressive Integrated Moving Average Model (SARIMA), use the SARIMA model to capture the linear trend and seasonal characteristics of the buoy photovoltaic power generation, and determine the parameters of the SARIMA model based on the linear trend and seasonal characteristics to obtain a power generation forecast model; calculate the residual between the daily predicted value of the buoy photovoltaic power generation output by the power generation forecast model and the actual value of the power generation on that day to obtain a residual sequence, use the residual sequence and historical weather data as input, and use the residual of the photovoltaic power generation forecast value as output to train a long short-term memory network (LSTM) model. During the training process, the LSTM model is used to learn the nonlinear relationship and long-term dependence between the buoy photovoltaic power generation and historical weather to obtain a residual forecast model.

[0023] In this embodiment, before the historical data is input into the model, missing value processing, outlier processing and normalization processing are also performed on the historical data set.

[0024] Specifically, because meteorological data and power generation data are temporally correlated, when a small amount of historical data is missing, the mean method is used to fill in the missing data. If a large amount of data is missing within a day, the data for that day is directly deleted. In this way, data loss and outlier problems caused by interference factors such as equipment failure, extreme weather at sea, or network interruption during the buoy historical data collection process are avoided, thereby significantly improving the accuracy of the model prediction. Among them, the expression for filling in the missing value with the mean of the data at the time point of the day before and the day after is:

[0025] ; in, x’ Refers to the data to fill in missing values, x i-1 Refers to the data of the day before the missing value, x i+1 Refers to the data for the day after the missing value.

[0026] Specifically, the process of photovoltaic power generation prediction involves features with different dimensions, such as temperature, irradiance, and cloud density. To prevent these features of different magnitudes from interfering with the prediction results, normalization is required before the data is input into the model to convert each feature to the same numerical range. In this process, the maximum and minimum value normalization method is used, as shown in the following formula:

[0027] ; Among them, 𝑥 is the original data, 𝑥 ∗ is the normalized data, 𝑚𝑖𝑛 is the minimum value in the original data, 𝑚𝑎𝑥 is the maximum value in the original data.

[0028] In this embodiment, a convolutional neural network (CNN) is used to extract spatial features from weather data (such as cloud distribution and radiometric image features), and local features are extracted from time series data. The features extracted by the CNN are used as input variables of the SARIMA model to assist SARIMA in processing input related to weather images (such as radiometric distribution maps), thereby improving the accuracy of seasonal feature extraction.

[0029] This embodiment uses the SARIMA model to capture the linear trend and seasonal characteristics of buoy photovoltaic power generation, and uses the LSTM model to learn the nonlinear relationship and long-term dependence between buoy photovoltaic power generation and historical weather. It can supplement the nonlinear part that the SARIMA model fails to capture and improve the prediction accuracy of buoy photovoltaic power generation.

[0030] The SARIMA model is in the form of , where the parameters are determined by the autocorrelation function (ACF) and partial autocorrelation function (PACF) graphs. p is the autoregressive order, d is the difference order, and q is the sliding average order, which are used to eliminate data non-stationarity and capture short-term trends; P is the seasonal autoregressive order, D is the seasonal difference order, Q is the seasonal sliding average order, and m is the seasonal period. For example, when the unit is "day", m=7 represents a weekly period, which is used to capture long-term seasonal patterns.

[0031] S30: The photovoltaic power generation of the buoy in a future preset period is predicted by the power generation prediction model and the residual prediction model to obtain a prediction result.

[0032] In this embodiment, the photovoltaic power generation of the buoy in the future preset period is predicted by the power generation prediction model and the residual prediction model to obtain the prediction results, which specifically include: Get weather data for a preset period in the future.

[0033] According to the length of the future preset period, the historical data of the buoy's photovoltaic power generation is input into the power generation prediction model to obtain the predicted value of the buoy's photovoltaic power generation in the future preset period; the weather data of the future preset period is input into the residual prediction model to obtain the residual of the predicted value of the buoy's photovoltaic power generation in the future preset period.

[0034] A prediction result is obtained based on the predicted value of the photovoltaic power generation of the buoy in the future preset period and the residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period.

[0035] S40: Obtain a prediction result based on the predicted value of the photovoltaic power generation of the buoy in the future preset period and the residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period.

[0036] In this embodiment, the prediction result is obtained based on the predicted value of the photovoltaic power generation of the buoy in the future preset period and the residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period, specifically including: The predicted value of photovoltaic power generation and the residual of the predicted value of photovoltaic power generation are summed by weighted average to obtain the prediction result. The calculation expression of weighted average is shown as follows:

[0037] ; The weight coefficient α is dynamically adjusted based on the model's performance on the validation set. This fusion approach fully leverages the SARIMA model's ability to capture linear trends and seasonality, and the LSTM model's ability to model nonlinear relationships and long-term dependencies, thereby achieving a highly accurate forecast of photovoltaic power generation over the next 14 days and determining daily power generation for the next 14 days.

[0038] based on Figure 1 The photovoltaic-based energy consumption optimization method for offshore buoys shown in the figure obtains historical data on photovoltaic power generation of the buoy in the target sea area, as well as historical weather data of the target sea area; and classifies the importance of each power-consuming component of the buoy to obtain an importance level sequence of each power-consuming component. The historical data of the buoy's photovoltaic power generation is input into a SARIMA model, and the SARIMA model is used to capture the linear trend and seasonal characteristics of the buoy's photovoltaic power generation. The parameters of the SARIMA model are determined based on the linear trend and seasonal characteristics to obtain a power generation prediction model; the residual between the daily predicted value of the buoy's photovoltaic power generation output by the power generation prediction model and the actual value of the power generation on that day is calculated to obtain a residual sequence, and the residual sequence and historical weather data are used as input, and the residual of the photovoltaic power generation prediction value is used as output to train an LSTM model. During the training process, the LSTM model is used to learn the nonlinear relationship and long-term dependence between the buoy's photovoltaic power generation and historical weather to obtain a residual prediction model; the photovoltaic power generation of the buoy in a future preset period is predicted using the power generation prediction model and the residual prediction model to obtain a prediction result; the power consumption of the buoy is allocated to each day in the future preset period based on the prediction result, and the working mode of each power-consuming component is set according to the importance level sequence and the power consumption allocated to the buoy on each day in the future preset period. This approach uses the SARIMA model to capture the linear trends and seasonal characteristics of a buoy's photovoltaic power generation, and the LSTM model to learn the nonlinear relationship and long-term dependency between the buoy's photovoltaic power generation and historical weather conditions, improving the accuracy of photovoltaic power generation forecasts for offshore buoys. Based on the buoy's photovoltaic power generation over a preset future period, power consumption is allocated to each power-consuming component within the buoy. The operating mode of each power-consuming component is then set based on its importance level and power consumption, improving the continuity and stability of offshore buoy operations.

[0039] When applying the method for optimizing energy consumption of offshore buoys based on photovoltaic power generation provided by the present invention, it is not necessary to Figure 1 The steps are executed in the order shown. The specific execution order of the steps can be determined according to needs, and the present invention does not limit this.

[0040] Furthermore, in one or more embodiments of the present invention, allocating power consumption to the buoy on each day within a future preset period according to the prediction results specifically includes: Based on the prediction results, determine whether there will be a day in the future preset period when the buoy's battery can be fully charged.

[0041] If there is a k-th day when the buoy's battery can be fully charged, calculate the total power generation of the buoy from the first day to the k-th day in the future preset period, where k is a positive integer.

[0042] The power consumption allocated to the buoy on each day in a future preset period is calculated based on the battery power and the total power generation.

[0043] For each day within a future preset period, determine whether the sum of the buoy's daily power generation and battery power is less than the power consumption allocated for the day. If the sum of the buoy's daily power generation and battery power is less than the power consumption allocated for the day, the power consumption allocated for the day will be determined as the sum of the buoy's daily power generation and battery power.

[0044] In this embodiment, reference Figure 2 , in the case that the preset period in the future is the next two weeks, the specific steps of daily power allocation are as follows: 1) Determine whether the battery can be fully charged one day in the next 14 days: traverse the power generation in the next 14 days E k , to determine whether there is a day k , making E k ≥ E a .

[0045] 2) If there is a day when the battery can be fully charged: Set up the first k Day is a day that can fully charge the battery, not taking into account k -14-day electricity distribution.

[0046] Calculate from the current to the k The total power generation forecast for the day is: ; Calculate total power E s : ; Calculate the allocated electricity consumption for each day E d : ; Determine whether the power generation on that day plus the battery power is less than E d, if so, then E d Equal to the amount of electricity generated on that day plus the battery capacity E c ,otherwise E d constant.

[0047] 3) If there is no day when the battery can be fully charged: Calculate the total predicted power generation from now to the 14th day E p : ; Calculate total power E s : ; Calculate the allocated electricity consumption for each day E d : ; Determine whether the power generation on that day plus the battery power is less than E d , if so, then E d Equal to the amount of electricity generated on that day plus the battery capacity E c ,otherwise E d constant.

[0048] Specifically, in Example 1 of this solution, a simulation model was built using the Matlab / Simulink platform to simulate changes in photovoltaic power generation under different weather conditions over the next 14 days. First, historical data (PV power generation data and meteorological data from the past five years) was input, and a machine learning model was used to predict power generation over the next 14 days. Then, based on the predicted power generation and energy consumption requirements, energy consumption was allocated to each device using constraints, and an optimization algorithm was used to find the optimal energy allocation solution.

[0049] For example, in a rainy weather scenario, the predicted photovoltaic power generation value is low. The energy consumption management system will prioritize ensuring the power supply of the buoy's most important communication system and data transmission equipment, while adjusting the energy consumption of other auxiliary equipment according to the energy consumption level to prevent the system from running out of power prematurely.

[0050] Through simulation testing, we were able to verify how the system optimizes energy consumption and improves energy efficiency in different weather scenarios, while ensuring that the buoy's basic functions are not affected. On sunny days, the buoy's battery is fully charged, and excess energy can be used for other auxiliary equipment. In rainy weather, the system can effectively dispatch energy and ensure the operation of key equipment such as the buoy's communications, preventing the buoy's functions from being affected by insufficient photovoltaic power generation due to prolonged rainy weather.

[0051] Specifically, in Example 2 of this solution, a buoy located in a certain sea area was selected. This buoy was equipped with a photovoltaic power generation system to provide energy for the buoy. First, photovoltaic power generation data for the past five years was obtained from a data center. Simultaneously, historical meteorological data interfaces provided by a meteorological service provider were used to collect data on sunlight, irradiance, cloud cover, and temperature for the same sea area over the past five years. All data was time-aligned and organized into a structured dataset on an hourly basis.

[0052] After obtaining the real-time weather data of the sea area where the buoy is located, the parameters (p, d, q) and seasonal parameters (P, D, Q, m) of the SARIMA model are determined through ACF and PACF analysis, and the SARIMA model is fitted with historical data to capture the linear trend and seasonal characteristics of the buoy's photovoltaic power generation. Next, an LSTM model is constructed to process the nonlinear relationship in the historical data, and the photovoltaic power generation data is divided into time series samples. A neural network structure containing an LSTM layer and a fully connected layer is designed, and trained on the training set. At the same time, the hyperparameters are adjusted on the validation set to optimize the performance. The residual learning method is adopted, and the prediction results of the SARIMA model are used as the benchmark to calculate the residual between the actual value and the predicted value. The LSTM model is used to model the residual to supplement the nonlinear part that SARIMA fails to capture. Finally, the prediction results of the two models are fused by weighted averaging, and the weight coefficients are dynamically adjusted. The SARIMA-LSTM model is used to achieve high-precision photovoltaic power generation forecasts for the next 14 days. The forecast results are as follows: Figure 3 As shown, the horizontal axis is time, the unit is day, the vertical axis is power, the unit is KWh, according to Figure 3 Comparing the curves of predicted power generation and actual power generation, it can be seen that the power generation prediction results of this scheme for future dates have the same trend as the actual power generation, and the results are close to fitting. In particular, the values ​​of actual power generation from the 6th to the 14th day are almost consistent with the predicted power generation, indicating that this scheme is relatively accurate in predicting photovoltaic power generation for future dates.

[0053] In this embodiment, different energy consumption levels are assigned to various energy-consuming components on the buoy (such as communications equipment, sensors, and battery management systems). Specifically, these components are categorized into four levels: "high priority," "medium priority," "low priority," and "emergency," and assigned scores based on their impact on the buoy's normal operation. Using the power allocation method described in this patent, the available power for the next 14 days is allocated. Within the constraints of daily available power, the highest-scoring solution is selected to maximize power allocation to high-priority tasks.

[0054] After several months of testing and optimization, the offshore buoys in this implementation are now able to proactively adjust their energy allocation strategy based on predicted PV power generation during most rainy weather conditions, effectively preventing the risk of power outages. In actual operation, compared to unoptimized energy management methods, the optimized buoys prioritize critical functions when PV power generation is insufficient, avoiding buoy failures caused by prolonged rainy weather and improving system reliability and sustainability.

[0055] This embodiment determines whether there is a day within the next two weeks that can fully charge the buoy's battery, based on the buoy's photovoltaic power generation over the next two weeks. If there is a k-th day that can fully charge the buoy's battery, the total power generation of the buoy from the first to the k-th day of the next two weeks is calculated, where k is a positive integer. Based on the battery power and the total power generation, the buoy's allocated power consumption for each day within the next two weeks is calculated. For each day within the next two weeks, a determination is made as to whether the sum of the buoy's daily power generation and battery power is less than the allocated power consumption for each day. If the sum of the buoy's daily power generation and battery power is less than the allocated power consumption for each day, the allocated power consumption for that day is determined to be the sum of the buoy's daily power generation and battery power. This improves the rationality of the buoy's power allocation over the next 14 days, enabling a reasonable power allocation to each of the next 14 days, preventing the buoy from running out of power on a given day and improving the stability of the buoy's operation.

[0056] Furthermore, in one or more embodiments of the present invention, setting the operating mode of each power-consuming component according to the importance level sequence and the power consumption allocated to the buoy on each day within a future preset period specifically includes: The various power-consuming components of the buoy are divided into working modes to obtain multiple working modes for each power-consuming component, and each working mode corresponds to a different working frequency and working time.

[0057] According to the importance level sequence, each power-consuming component is assigned a score in each working mode.

[0058] All combination schemes of each power-consuming component in different working modes are determined by exhaustive method, and the scheme in which the sum of the power consumption of each power-consuming component in all combination schemes is less than or equal to the power consumption allocated each day is determined as the target scheme.

[0059] According to the scores of each power-consuming component in each working mode, the total scores corresponding to each scheme in the target scheme are obtained, the scheme with the largest total score in the target scheme is determined as the optimal scheme, and the working mode of each power-consuming component is set according to the optimal scheme.

[0060] In this embodiment, the operating modes of each power-consuming component are divided into multiple operating modes. Specifically, the multiple operating modes include four gear operating modes, as shown in Table 1.

[0061] Table 1 Power consumption component gear settings In this embodiment, different levels of different components are assigned points according to their importance, as shown in Table 2.

[0062] Table 2 Scoring of different power consumption components Specifically, let the first working mode of element 1 be E 11 , the mode is assigned S 11 , the second working mode is E 12 , the mode is assigned S 12 , and so on, the nth working mode of component m is E mn , the mode is assigned S mn .

[0063] The mathematical constraints are: E 1i +E 2p +…+E mq ≤Ed, under this constraint, by enumerating all possible solutions, we can use the exhaustive method to obtain a brute force search, calculate the score of each solution, and select the optimal solution that meets the constraint.

[0064] The above is a method for optimizing energy consumption of an offshore buoy based on photovoltaic power generation provided by one or more embodiments of the present invention. Based on the same idea, the present invention also provides a corresponding device for optimizing energy consumption of an offshore buoy based on photovoltaic power generation, such as Figure 4 Shown, including: The acquisition module is used to obtain the historical data of the photovoltaic power generation of the buoy in the target sea area, as well as the historical weather data of the target sea area, to obtain a historical data set; the importance level of each power-consuming component of the buoy is graded and the working mode is divided, to obtain the importance level sequence of each power-consuming component and multiple working modes, each of the multiple working modes corresponds to a different working frequency and working time.

[0065] A construction module is used to input historical datasets into the SARIMA model, use the SARIMA model to capture the linear trend and seasonal characteristics of buoy photovoltaic power generation, and obtain a power generation prediction model; input historical datasets into the LSTM model, use the LSTM model to learn the nonlinear relationship and long-term dependence between buoy photovoltaic power generation and historical weather, and obtain a residual prediction model.

[0066] The prediction module is used to input the acquired weather data of the target sea area in the future preset period into the power generation prediction model and the residual prediction model respectively, to obtain the first prediction value and the second prediction value of the photovoltaic power generation of the buoy in the future preset period.

[0067] The energy consumption optimization module is used to determine the photovoltaic power generation of the buoy in a future preset period based on the first prediction value and the second prediction value; allocate power consumption to each power-consuming component in the buoy according to the photovoltaic power generation of the buoy in the future preset period, and set the working mode of each power-consuming component according to the importance level sequence and the power consumption of each power-consuming component.

[0068] The specific definition of the photovoltaic-based offshore buoy energy consumption optimization device can be found in the definition of the photovoltaic-based offshore buoy energy consumption optimization method above, and will not be repeated here. The various modules in the photovoltaic-based offshore buoy energy consumption optimization device can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of each of the above modules.

[0069] The present invention also provides a computer-readable storage medium, which stores a computer program, which can be used to execute the Figure 1 A method for optimizing energy consumption of offshore buoys based on photovoltaic power generation is provided.

[0070] The present invention also provides Figure 5 The structural diagram of the computer equipment shown in FIG. Figure 5As shown, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and may also include other hardware required for the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it to implement the Figure 1 A method for optimizing energy consumption of offshore buoys based on photovoltaic power generation is provided.

[0071] Those skilled in the art will appreciate that all or part of the processes in the described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the described method embodiments. Any reference to memory, storage, database, or other media used in the embodiments provided herein may include at least one of non-volatile and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0072] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present invention.

Claims

1. A method for optimizing energy consumption of offshore buoys based on photovoltaic power generation, characterized in that: include: Acquiring historical data on photovoltaic power generation of a buoy in a target sea area and historical weather data of the target sea area; grading the importance of each power-consuming component of the buoy to obtain an importance level sequence of each power-consuming component; Inputting the historical data of the buoy's photovoltaic power generation into a SARIMA model, using the SARIMA model to capture the linear trend and seasonal characteristics of the buoy's photovoltaic power generation, and determining the parameters of the SARIMA model based on the linear trend and seasonal characteristics to obtain a power generation prediction model; calculating the residual between the daily predicted value of the buoy's photovoltaic power generation output by the power generation prediction model and the actual value of the power generation on that day to obtain a residual sequence, using the residual sequence and the historical weather data as input, and using the residual of the photovoltaic power generation prediction value as output to train an LSTM model, and using the LSTM model to learn the nonlinear relationship and long-term dependency between the buoy's photovoltaic power generation and historical weather during the training process to obtain a residual prediction model; The photovoltaic power generation of the buoy in a future preset period is predicted by the power generation prediction model and the residual prediction model to obtain a prediction result; The power consumption of the buoy is allocated on each day in a future preset period according to the prediction result, and the working mode of each power-consuming component is set according to the importance level sequence and the power consumption allocated to the buoy on each day in the future preset period.

2. The method for optimizing energy consumption of offshore buoys based on photovoltaic power generation according to claim 1, characterized in that: The photovoltaic power generation of the buoy in a future preset period is predicted by the power generation prediction model and the residual prediction model to obtain a prediction result, which specifically includes: Obtaining weather data for the future preset period; Inputting historical data of the photovoltaic power generation of the buoy into the power generation prediction model according to the time length of the future preset period to obtain a predicted value of the photovoltaic power generation of the buoy in the future preset period; inputting weather data of the future preset period into the residual prediction model to obtain a residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period; The prediction result is obtained based on the predicted value of the photovoltaic power generation of the buoy in the future preset period and the residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period.

3. The method for optimizing energy consumption of offshore buoys based on photovoltaic power generation according to claim 1, characterized in that: Allocating power consumption to the buoy on each day within a future preset period according to the prediction results specifically includes: Determining, based on the prediction results, whether there is a day in a future preset period when the battery of the buoy can be fully charged; If there is a k-th day when the battery of the buoy can be fully charged, calculate the total power generation of the buoy from the first day to the k-th day in the future preset period, where k is a positive integer; Calculating the power consumption allocated to the buoy on each day within a future preset period based on the battery power and the total power generation; For each day within a future preset period, determine whether the sum of the buoy's daily power generation and the battery power is less than the power consumption allocated for the day. If the sum of the buoy's daily power generation and the battery power is less than the power consumption allocated for the day, the power consumption allocated for the day will be determined as the sum of the buoy's daily power generation and the battery power.

4. The method for optimizing energy consumption of offshore buoys based on photovoltaic power generation according to claim 1, wherein: The operation mode of each power-consuming component is set according to the importance level sequence and the power consumption allocated to the buoy on each day in a future preset period, specifically including: Dividing the working modes of the power-consuming components of the buoy to obtain multiple working modes of the power-consuming components, each working mode corresponding to a different working frequency and working duration; assigning points to each of the power-consuming components in each operating mode according to the importance level sequence; Determine all combination schemes of the power-consuming components under different operating modes by an exhaustive method, and determine as a target scheme a scheme in which the sum of the power consumption of the power-consuming components among all the combination schemes is less than or equal to the power consumption allocated for each day; According to the scores of each power-consuming component in each working mode, the total scores corresponding to each solution in the target solution are obtained, the solution with the largest total score in the target solution is determined as the optimal solution, and the working mode of each power-consuming component is set according to the optimal solution.

5. The method for optimizing energy consumption of offshore buoys based on photovoltaic power generation according to claim 2, wherein: The prediction result is obtained based on the predicted value of the photovoltaic power generation of the buoy in the future preset period and the residual of the predicted value of the photovoltaic power generation of the buoy in the future preset period, specifically including: The photovoltaic power generation prediction value and the residual of the photovoltaic power generation prediction value are summed in a weighted average manner to obtain the prediction result.

6. The method for optimizing energy consumption of offshore buoys based on photovoltaic power generation according to claim 1, characterized in that: Before inputting the historical data set into the SARIMA model, the method further includes performing missing value processing, outlier processing and normalization processing on the historical data set.

7. The method for optimizing energy consumption of offshore buoys based on photovoltaic power generation according to claim 1, characterized in that: The weather data includes temperature, radiometry and cloud cover.

8. An offshore buoy energy consumption optimization device based on photovoltaic power generation, characterized in that: include: an acquisition module for acquiring historical data on photovoltaic power generation of the buoy in the target sea area and historical weather data of the target sea area; grading the importance of each power-consuming component of the buoy to obtain an importance level sequence of each power-consuming component; A construction module is provided for inputting the historical data of the buoy's photovoltaic power generation into a SARIMA model, using the SARIMA model to capture the linear trend and seasonal characteristics of the buoy's photovoltaic power generation, and determining the parameters of the SARIMA model based on the linear trend and seasonal characteristics to obtain a power generation prediction model; calculating the residual between the daily predicted value of the buoy's photovoltaic power generation output by the power generation prediction model and the actual value of the power generation on that day to obtain a residual sequence, using the residual sequence and the historical weather data as input, and using the residual of the photovoltaic power generation prediction value as output to train an LSTM model, and using the LSTM model to learn the nonlinear relationship and long-term dependence between the buoy's photovoltaic power generation and historical weather during the training process to obtain a residual prediction model; A prediction module, configured to predict the photovoltaic power generation of the buoy in a future preset period by using the power generation prediction model and the residual prediction model to obtain a prediction result; The energy consumption optimization module is used to allocate the power consumption of the buoy on each day in the future preset period according to the prediction results, and to set the working mode of each power-consuming component according to the importance level sequence and the power consumption allocated to the buoy on each day in the future preset period.

9. A computer device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the program, the method for optimizing energy consumption of an offshore buoy based on photovoltaic power generation according to any one of claims 1 to 5 is implemented.

Images