New energy multi-energy complementary monitoring system and method based on artificial intelligence

Through the new energy multi-energy complementary monitoring system based on artificial intelligence, the operation of new energy is monitored and optimized in real time, and the problems of low resource volatility and energy conversion efficiency are solved, achieving efficient, stable operation and robustness of the system.

CN118971344APending Publication Date: 2024-11-15深圳市宏智智能装备有限公司
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411004412.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-25
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

New energy multi-energy complementary systems face problems such as resource volatility and instability, low energy conversion efficiency, difficult system optimization and insufficient data analysis, which makes it difficult for the system to achieve efficient and stable operation.

Method used

Adopt a new energy multi-energy complementary monitoring system based on artificial intelligence, and through data acquisition, processing, intelligent analysis and system control modules, the operating status of new energy is monitored and optimized in real time, and deep learning and machine learning algorithms are used to predict potential problems and generate optimized scheduling. The solution is to achieve multi-energy complementarity between new energy sources.

Benefits of technology

It improves the stability and conversion efficiency of energy output, enhances the robustness and reliability of the system, can respond to abnormal situations in a timely manner, optimizes the system operation strategy, and improves overall performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118971344A_ABST
    Figure CN118971344A_ABST
Patent Text Reader

Abstract

The invention discloses a new energy multi-energy complementary monitoring system and method based on artificial intelligence, and the system comprises a data collection module which is used for collecting various types of power generation data, meteorological data and equipment operation state data in real time through various types of sensors disposed on various types of new energy power generation equipment; the data processing module is used for cleaning, storing and preprocessing a large amount of collected original data; constructing a sample data set; the intelligent analysis module performs analysis and training by using the sample data set based on an artificial intelligence algorithm to obtain a prediction model; on the basis of the prediction model, predicting potential problems in system operation, and providing an optimal scheduling scheme; and the system control module is used for controlling the operation states of various corresponding power generation equipment according to the scheduling scheme provided by the intelligent analysis module, so that multi-energy complementation among new energy sources is realized. According to the system, through artificial intelligence prediction and optimal scheduling, the fluctuation of new energy power generation is effectively reduced, and the energy output stability of the system is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of new energy power stations and artificial intelligence data processing technology, and more specifically to a new energy multi-energy complementary monitoring system and method based on artificial intelligence. Background Art

[0002] At present, the new energy multi-energy complementary system mainly includes the integrated utilization of various renewable energy sources such as solar energy, geothermal energy, wind energy, ocean energy, biomass energy and nuclear fusion energy. However, due to the significant differences in the power generation characteristics and meteorological conditions of various new energy sources, the system faces many challenges in resource scheduling, energy conversion and comprehensive utilization:

[0003] 1. Resource volatility and instability: Wind and solar energy are greatly affected by meteorological conditions and have significant volatility and instability, which makes the energy output of the system difficult to predict and control.

[0004] 2. Low energy conversion efficiency: Different forms of energy conversion efficiency vary, and traditional multi-energy complementary systems are prone to inefficiency during the energy conversion process.

[0005] 3. Difficulty in system optimization: Due to the mutual coupling of multiple energy sources, the optimization and scheduling of the system are highly complex, and existing optimization methods often cannot meet the needs of efficient scheduling and stable operation.

[0006] 4. Insufficient data analysis: Traditional systems lack the ability to analyze and process large-scale real-time data and are unable to identify and respond to system anomalies in a timely manner. Summary of the invention

[0007] In view of this, in order to solve the above problems, the present invention provides a new energy multi-energy complementary monitoring system and method based on artificial intelligence, which uses artificial intelligence technology to perform real-time monitoring, data analysis and intelligent scheduling of multiple new energy sources to achieve efficient and stable operation of the system.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] The embodiment of the present invention provides a new energy multi-energy complementary monitoring system based on artificial intelligence, comprising:

[0010] The data acquisition module is used to collect various power generation data, meteorological data and equipment operation status data in real time through various sensors installed on various new energy power generation equipment;

[0011] The data processing module cleans, stores and preprocesses the large amount of raw data collected; and constructs a sample data set;

[0012] The intelligent analysis module uses sample data sets for analysis and training based on artificial intelligence algorithms to obtain a prediction model. Based on the prediction model, it predicts potential problems in system operation and provides an optimized scheduling solution.

[0013] The system control module controls the operating status of various corresponding power generation equipment according to the scheduling plan provided by the intelligent analysis module, realizes multi-energy complementarity among new energy sources, and ensures stable and efficient operation of the system.

[0014] Furthermore, the data acquisition module includes:

[0015] Sensor units, which are a variety of sensors installed on various types of renewable energy power generation equipment, including temperature sensors, humidity sensors, wind speed and direction sensors, light intensity sensors, image sensors, position sensors, and current and voltage sensors;

[0016] Data collector: used to receive data collected by the sensor unit, perform preliminary processing and data format conversion;

[0017] A data transmission unit transmits data to a central processing unit or a cloud server via a wireless network or a wired network;

[0018] The power management unit is used to supply power to the sensor unit and the data logger.

[0019] Furthermore, the data processing module includes:

[0020] Cleaning cells by using mean interpolation, interpolation, or deleting rows containing missing values; using statistical methods to detect and remove outliers; and using wavelet or Fourier transforms to remove noise from the data;

[0021] The preprocessing unit performs standardization, normalization and dimension reduction on the cleaned data in sequence;

[0022] The storage unit organizes structured data into data tables according to their features and stores them in relational databases; it stores unstructured data in the form of documents in NoSQL databases.

[0023] Furthermore, the data processing module further includes:

[0024] The fusion unit performs feature-level fusion or decision-level fusion processing on multi-source data.

[0025] Furthermore, the intelligent analysis module includes:

[0026] Data preparation and feature engineering unit, used for data preparation including obtaining sample data sets from the data processing module and performing feature engineering;

[0027] Model selection and training unit, used to adopt deep learning models, suitable for processing time series data, to predict multi-energy complementary systems; or adopt machine learning models, suitable for classification and regression tasks, to identify potential problems in the system;

[0028] The prediction and analysis unit uses the trained model to predict the new data obtained and analyze the results to identify potential problems in system operation;

[0029] The optimization scheduling plan generation unit sets the optimization goal of maximizing energy output and minimizing energy loss based on the prediction and analysis results, and uses the optimization algorithm to generate a scheduling plan.

[0030] Furthermore, in the data preparation and feature engineering module, feature engineering includes:

[0031] The feature selection unit is used to select the features that have a greater impact on the system operation, including temperature, humidity, wind speed and light intensity;

[0032] Feature construction unit, constructs new features based on existing features, including: moving average and cumulative value of historical data;

[0033] The feature scaling unit is used to standardize or normalize the features to ensure that the data ranges of different features are similar, so as to improve the model convergence speed and prediction accuracy.

[0034] The model selection and training unit adopts a deep learning model, which is suitable for processing time series data and predicting a multi-energy complementary system, specifically including:

[0035] Select LSTM as the prediction network to predict time series data;

[0036] Improve LSTM by increasing the number of LSTM layers to three, setting 100 units in each layer, and adding a Dropout layer after each layer; and add an Attention layer after the LSTM layer, and use the Concatenate layer to combine the LSTM output and the Attention output;

[0037] The historical power generation data, equipment status data and corresponding environmental parameters of various new energy systems are used as sample data for training and optimization of the improved LSTM, and finally the optimized prediction model is obtained.

[0038] Furthermore, the system control module includes:

[0039] The optimization scheduling execution unit is used to receive the optimization scheduling plan generated by the intelligent analysis module, convert it into specific control instructions, and send it to the corresponding various types of new energy power generation equipment, and record the execution status of the scheduling instructions and provide feedback;

[0040] Real-time monitoring unit is used to monitor the system operation status in real time to ensure the stable operation of the system under various working conditions; when an abnormal situation is detected, an alarm is issued in time and corresponding measures are given;

[0041] Adaptive control unit, based on real-time monitoring data and feedback, automatically adjusts the system's power generation and operating parameters of start-stop equipment to adapt to environmental changes and load fluctuations;

[0042] The human-machine interaction unit is used to provide a friendly user interface, allowing operators to easily monitor the system operating status, adjust control parameters and handle abnormal situations.

[0043] In the second aspect, the present invention also provides a new energy multi-energy complementary monitoring method based on artificial intelligence, using the new energy multi-energy complementary monitoring system based on artificial intelligence as described in any one of the first aspects to realize real-time monitoring, data analysis and intelligent scheduling of multiple new energy sources.

[0044] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a new energy multi-energy complementary monitoring system based on artificial intelligence, which has the following advantages:

[0045] Improve energy output stability: Through artificial intelligence prediction and optimized scheduling, the volatility of renewable energy power generation is effectively reduced and the energy output stability of the system is improved.

[0046] Improve energy conversion efficiency: Intelligent scheduling optimizes the conversion efficiency of different types of energy, maximizes the advantages of various new energy sources, and improves the energy conversion efficiency of the overall system.

[0047] Enhance the robustness and reliability of the system: Through real-time monitoring and anomaly detection, the system can respond to and handle various abnormal situations in a timely manner to ensure the stable operation of the system.

[0048] Data-driven decision support: Based on big data analysis and artificial intelligence algorithms, data-driven decision support is provided to optimize the system's operating strategy and improve the system's overall performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0050] Figure 1 This is a block diagram of the new energy multi-energy complementary monitoring system based on artificial intelligence provided by the present invention.

[0051] Figure 2 This is a block diagram of the data acquisition module provided by the present invention.

[0052] Figure 3 A block diagram of the data processing module provided by the present invention.

[0053] Figure 4 A block diagram of the intelligent analysis module provided by the present invention.

[0054] Figure 5 A block diagram of the feature engineering provided by the present invention.

[0055] Figure 6 This is a block diagram of the system control module provided by the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] The embodiment of the present invention discloses a new energy multi-energy complementary monitoring system based on artificial intelligence, referring to Figure 1 As shown, including:

[0058] The data acquisition module is used to collect various power generation data, meteorological data and equipment operation status data in real time through various sensors installed on various new energy power generation equipment;

[0059] The data processing module cleans, stores and preprocesses the large amount of raw data collected; and constructs a sample data set;

[0060] The intelligent analysis module uses sample data sets for analysis and training based on artificial intelligence algorithms to obtain a prediction model. Based on the prediction model, it predicts potential problems in system operation and provides an optimized scheduling solution.

[0061] The system control module controls the operating status of various corresponding power generation equipment according to the scheduling plan provided by the intelligent analysis module, realizes multi-energy complementarity among new energy sources, and ensures stable and efficient operation of the system.

[0062] For example:

[0063] Example 1

[0064] A new energy power plant has installed a variety of new energy power generation equipment, including wind turbines, photovoltaic panels and biomass power generation equipment. The system architecture is as follows:

[0065] Data acquisition module: Temperature sensors, humidity sensors, wind speed and direction sensors are installed on wind turbines and biomass power generation equipment, light intensity sensors are installed on photovoltaic panels, and voltage and current detection sensors are installed separately, each of which is used to collect environmental data and equipment operation data in real time.

[0066] Data processing module: The collected data is transmitted to the central processing unit via the wireless network for data cleaning, storage and preprocessing, and multi-source data fusion.

[0067] Intelligent analysis module: Based on historical data and real-time data, it uses deep learning algorithms to predict the power generation situation in the next 24 hours, and uses reinforcement learning algorithms to optimize the power generation scheduling plan.

[0068] System control module: According to the optimized scheduling plan, dynamically adjust the speed of wind turbines, the angle of photovoltaic panels and the operating status of biomass power generation equipment to ensure the optimal operating efficiency of the system.

[0069] Finally, this new energy power plant, through the prediction and scheduling of the volatility of wind and photovoltaic power generation, the system can effectively respond to the power generation fluctuations caused by weather changes and ensure stable energy output.

[0070] Example 2

[0071] The new energy power generation system built in a coastal area includes three new energy power generation equipment: wind energy, ocean energy and solar energy. The system architecture is as follows:

[0072] Data acquisition module: Install water flow velocity sensors and temperature sensors on ocean energy power generation equipment, install wind speed and direction sensors on wind turbines, and install light intensity sensors on photovoltaic panels to collect environmental data and equipment operation data in real time.

[0073] Data processing module: Through data fusion technology, the power generation data of ocean energy, wind energy and solar energy are comprehensively processed and stored in the cloud database.

[0074] Intelligent analysis module: Based on deep learning algorithms, it predicts the power generation trend in the next week, and combines weather forecast data with reinforcement learning algorithms to develop optimized power generation scheduling plans.

[0075] System control module: According to the scheduling plan provided by the intelligent analysis module, dynamically adjust the operating status of various power generation equipment to ensure the optimal operating efficiency of the system under different meteorological conditions.

[0076] In this embodiment, the new energy power generation system constructed in the coastal area can effectively respond to changes in the marine environment and meteorological conditions and ensure stable energy output through comprehensive prediction and scheduling of ocean energy, wind energy and solar power generation.

[0077] In one embodiment, the data acquisition module, referring to Figure 2 As shown, including:

[0078] Sensor units, a variety of sensors installed on various types of new energy power generation equipment, including temperature sensors, humidity sensors, wind speed and direction sensors, light intensity sensors, image sensors, position sensors and current and voltage sensors.

[0079] Depending on the different types of renewable energy power generation equipment, the configuration of the sensor unit is different:

[0080] For example, wind power generation equipment can be equipped with wind speed sensors to measure wind speed and help predict wind power output. Wind direction sensors can be installed to measure wind direction and help optimize the angle adjustment of wind turbines. Temperature sensors can be installed to monitor ambient temperature or equipment temperature to prevent equipment from overheating and ensure optimal operation.

[0081] For example, solar power generation equipment is equipped with light intensity sensors, temperature sensors and tilt sensors to measure the intensity of solar radiation and predict the power output of photovoltaic power generation; monitor the temperature of photovoltaic panels to prevent overheating from affecting power generation efficiency; and monitor the tilt angle of photovoltaic panels to optimize light reception.

[0082] For another example, biomass power generation equipment is equipped with temperature sensors, humidity sensors and flue gas sensors to monitor the combustion temperature of biomass fuel to ensure the efficiency and safety of the combustion process; monitor the humidity of biomass materials to determine the impact on combustion efficiency; and monitor the composition of flue gas produced by combustion to control pollution emissions.

[0083] The data collector is used to receive the data collected by the sensor unit, perform preliminary processing and data format conversion. Receive the raw data sent by the sensor unit, including analog signals and digital signals; and perform preliminary processing such as filtering, amplification, and A / D conversion on the received data. Then convert the processed data into a standard data format (such as JSON, XML) for easy transmission and storage. When the network is disconnected or the signal is unstable, temporarily store the data and upload it after the network is restored.

[0084] The data transmission unit transmits data to the central processing unit or cloud server through a wireless network or a wired network; for example, low-power wide area network technologies such as Wi-Fi, LoRa, and NB-IoT are used to achieve long-distance, low-power data transmission. In scenarios that require high reliability and large bandwidth, Ethernet is used for data transmission. In addition, the transmitted data can be encrypted to ensure the security and integrity of the data during transmission.

[0085] The power management unit is used to power the sensor unit and data collector. For example, in the wild or in scenarios where there is no access to the power grid, solar cells are used to power the sensor unit and data collector. In the case of continuous lack of solar energy, batteries are used to provide backup power to ensure the continuity of data collection. The power status can also be monitored in real time, low-battery alarms can be issued in time, and the backup power supply can be switched.

[0086] In this embodiment, the data acquisition module can realize comprehensive monitoring and data acquisition of various types of new energy power generation equipment, and provide reliable data support for the intelligent analysis and optimized scheduling of the system.

[0087] In one embodiment, the data processing module, referring to Figure 3 As shown, including:

[0088] Cleaning cells by using mean interpolation, interpolation, or deleting rows containing missing values; using statistical methods to detect and remove outliers; and using wavelet or Fourier transforms to remove noise from the data;

[0089] Assume that the collected data set is Among them, x i =[x i1 , x i2 , ..., x im ], represents the i-th sample, containing m features;

[0090] (1) Missing value processing (mean interpolation method):

[0091]

[0092] Among them, N j Indicates the number of samples with non-missing values ​​for the jth feature.

[0093] (2) Outlier detection (Z-Score method):

[0094]

[0095] Among them, μ j and σ j are the mean and standard deviation of the jth feature respectively; if |z ij∣>τ (usually τ=3), then it is considered that x ij Outliers can be replaced by the mean or median.

[0096] (3) Noise removal:

[0097]

[0098] Among them, W and W- 1 They represent wavelet transform and inverse transform respectively, and T is the threshold processing function used to remove noise.

[0099] The preprocessing unit performs standardization, normalization and dimension reduction on the cleaned data in sequence;

[0100] standardization:

[0101]

[0102] where μj and σj are the mean and standard deviation of the j-th feature, respectively.

[0103] Normalization:

[0104]

[0105] Where min(x j ) and max(x j ) are the minimum and maximum values ​​of the j-th feature, respectively.

[0106] Dimensionality reduction (Principal component analysis PCA):

[0107] X new =Xw

[0108] Among them, W is the projection matrix, which is obtained by eigenvalue decomposition.

[0109] The fusion unit performs feature-level fusion or decision-level fusion processing on multi-source data. Multi-source data fusion technology is used to comprehensively process data from different sources to form a consistent and comprehensive data set.

[0110] Assume that there are two data sources: and Data fusion is required.

[0111] 1. Feature-level fusion:

[0112]

[0113] The features of different data sources are concatenated to form a new sample feature vector.

[0114] 2. Decision-level fusion:

[0115]

[0116] Where f and g are the prediction functions of different data sources, respectively, and ⊕ represents the decision fusion operation (such as weighted average, voting, etc.).

[0117] The storage unit separates structured data into data tables according to their features and stores them in relational databases; it stores unstructured data in the form of documents in NoSQL databases. Data storage needs to consider the structured and unstructured characteristics of data and choose a suitable storage solution, such as relational databases (MySQL, PostgreSQL) or non-relational NoSQL databases (MongoDB, Cassandra).

[0118] In this embodiment, the data processing module can process the collected raw data comprehensively and accurately, providing a solid foundation for building a high-quality sample data set, and further providing support for subsequent intelligent analysis and optimized scheduling.

[0119] In one embodiment, the intelligent analysis module analyzes and predicts the processed data based on artificial intelligence algorithms, identifies potential problems in system operation, and provides optimized scheduling solutions. This module uses deep learning and machine learning technologies, combined with data from actual applications, to build an efficient analysis and prediction model. The following is a detailed technical solution and specific implementation steps.

[0120] Reference Figure 4 As shown, the intelligent analysis module includes:

[0121] The data preparation and feature engineering unit is used for data preparation, including obtaining sample data sets from the data processing module and performing feature engineering. Among them, data preparation includes obtaining cleaned and pre-processed data from the data processing module and performing feature engineering to improve the prediction ability of the model.

[0122] Among them, refer to Figure 5 As shown, the feature engineering steps are as follows:

[0123] Feature selection: Select features that have a greater impact on system operation, such as temperature, humidity, wind speed, light intensity, etc.

[0124] Feature construction: construct new features based on existing features, such as moving average and cumulative value of historical data.

[0125] Feature scaling: Standardize or normalize features to ensure that the data ranges of different features are similar, thereby improving model convergence speed and prediction accuracy.

[0126] In addition, feature learning can be achieved by analyzing the real-time operation data of the equipment to learn the correlation between fault features. Based on the learned feature relationship, when equipment A fails, related features that may cause other faults are automatically recommended. The recommended features are monitored in detail to improve the fault warning capability of the system.

[0127] The model selection and training unit is used to adopt deep learning models, which are suitable for processing time series data and making predictions on multi-energy complementary systems; or adopt machine learning models, which are suitable for classification and regression tasks and identifying potential problems in the system.

[0128] Among them, the deep learning model (LSTM, long short-term memory network) is suitable for processing time series data and predicting multi-energy complementary systems.

[0129] Machine learning models (such as random forests or support vector machines): suitable for classification and regression tasks to identify potential problems in the system.

[0130] The optimization scheduling plan generation unit sets the optimization goal of maximizing energy output and minimizing energy loss based on the prediction and analysis results, and uses the optimization algorithm to generate a scheduling plan.

[0131] Based on the prediction and analysis results, an optimized scheduling plan for the system is provided to improve energy conversion efficiency and system operation stability.

[0132] Optimization goals: maximize energy output, minimize energy loss, and improve system stability.

[0133] Optimization algorithm: Use optimization algorithms (such as genetic algorithm and particle swarm optimization) to generate scheduling plans.

[0134] Optimization algorithm steps (taking genetic algorithm as an example):

[0135] 1) Initialize the population: Generate the initial scheduling solution population.

[0136] 2) Fitness evaluation: Calculate the fitness of each scheduling scheme and evaluate its performance.

[0137] 3) Selection operation: Select the solution with higher fitness for crossover and mutation operations to generate a new solution.

[0138] 4) Crossover operation: Combine some features of two solutions to generate a new solution.

[0139] 5) Mutation operation: Randomly change some features of the solution to increase diversity.

[0140] 6) Iterative update: Repeat the above steps until the termination condition is met.

[0141] In the specific implementation, in the multi-energy complementary system, the long short-term memory network (LSTM) in deep learning is used to predict time series data, and the multi-energy complementary scheme is generated in combination with the optimization algorithm. The following is a detailed implementation process, including network selection, network improvement, input parameters, output results and the implementation steps of the complementary scheme.

[0142] 1. Network selection and improvement

[0143] Since the data in multi-energy systems has strong temporal sequence, Long Short-Term Memory (LSTM) is an ideal choice for processing such data. LSTM can capture the long-term dependencies of data and make accurate predictions of future energy production.

[0144] LSTM Network Architecture

[0145] Input layer: Receives historical data of various energy sources and environmental parameters.

[0146] LSTM layer: captures long-term dependencies of time series data.

[0147] Fully connected layer: maps the output of the LSTM layer to specific prediction values.

[0148] Output layer: Outputs the predicted energy production.

[0149] The improvements are as follows:

[0150] 1.1 Increase the number of LSTM layers and units

[0151] By increasing the number of LSTM layers and the number of units in each layer, the representation capability of the model is improved, enabling it to better capture complex patterns in the data.

[0152] Implementation steps:

[0153] Increase the number of LSTM layers to three, with 100 units in each layer.

[0154] A Dropout layer is added after each layer to prevent overfitting.

[0155] 1.2 Introducing the Attention Mechanism

[0156] By introducing the Attention mechanism, the model can focus on important time steps and improve the accuracy of prediction.

[0157] Implementation steps:

[0158] Add an Attention layer after the LSTM layer.

[0159] Use the Concatenate layer to combine the LSTM output and the Attention output.

[0160] 1.3 Using a hybrid model (LSTM+CNN)

[0161] By combining convolutional neural network (CNN) and LSTM, the local features and temporal characteristics in the data are captured to further improve the prediction accuracy.

[0162] Implementation steps:

[0163] A 1D convolution layer and a pooling layer are added after the input layer to extract local features.

[0164] The output of the convolutional layer is combined with the LSTM layer to capture the temporal features.

[0165] 1.4 Data Augmentation and Generative Adversarial Networks (GAN)

[0166] More sample data is generated through data enhancement and generative adversarial networks (GANs) to improve the generalization ability of the model.

[0167] Implementation steps:

[0168] Generate additional training data using GANs.

[0169] Combine the generated data with the original data to enhance the model training effect.

[0170] 1.5 Hyperparameter Optimization

[0171] Use hyperparameter optimization tools (such as Grid Search or Bayesian Optimization) to find the optimal hyperparameter combination and improve model performance.

[0172] Implementation steps:

[0173] Define the hyperparameter grid.

[0174] Use Grid Search to search for hyperparameters and find the optimal parameter combination.

[0175] 1.6 Loss Function

[0176] By introducing adaptive weights, the loss function can dynamically adjust the weights according to the prediction error, thereby improving the adaptability of the model.

[0177] Mathematical expression:

[0178]

[0179] Where k is the adjustment coefficient, N is the number of samples; y i is the true value; is the predicted value.

[0180] 2. Train the improved LSTM model:

[0181] Prepare training and test data sets, including historical data and environmental parameters of various new energy systems: iteratively optimize the improved LSTM model.

[0182] 3. Based on the trained prediction model, input the historical data and environmental parameters of various new energy systems:

[0183] Historical power generation data: such as solar power generation, wind power generation, biomass power generation, etc.

[0184] Meteorological data: such as temperature, humidity, wind speed, light intensity, etc.

[0185] Equipment status data: such as equipment operating status, fault records, etc.

[0186] The sample input data is shown in Table 1:

[0187]

[0188] 4. The output result is the power generation forecast value of various energy sources in a certain period of time in the future and the optimized complementary scheduling plan:

[0189] Forecasted power generation: power generation from solar energy, wind energy, and biomass energy in the future.

[0190] Optimize the scheduling plan: Generate a multi-energy complementary scheduling plan based on the prediction results to maximize energy utilization efficiency. The sample output data is shown in Table 2:

[0191]

[0192] Through the above-mentioned improvement scheme, the accuracy of power generation prediction in the multi-energy complementary system can be improved, thereby better optimizing energy scheduling and improving the overall efficiency and stability of the system.

[0193] In this embodiment, the intelligent analysis module can realize efficient analysis and prediction of the new energy multi-energy complementary system, identify potential problems in the operation of the system, and provide optimized scheduling solutions to improve the overall performance and operating efficiency of the system.

[0194] In one embodiment, the system control module is responsible for executing the optimized scheduling plan generated by the intelligent analysis module, monitoring the operating status of the new energy multi-energy complementary system, and making real-time adjustments and feedback to ensure the efficient, safe and stable operation of the system. The following is a detailed description and technical details:

[0195] Reference Figure 6 As shown, the system control module includes:

[0196] The optimization scheduling execution unit is used to receive the optimization scheduling plan generated by the intelligent analysis module, convert it into specific control instructions, and send it to the corresponding various types of new energy power generation equipment, and record the execution status of the scheduling instructions and provide feedback;

[0197] Real-time monitoring unit is used to monitor the system operation status in real time to ensure the stable operation of the system under various working conditions; when an abnormal situation is detected, an alarm is issued in time and corresponding measures are given;

[0198] Adaptive control unit, based on real-time monitoring data and feedback, automatically adjusts the system's power generation and operating parameters of start-stop equipment to adapt to environmental changes and load fluctuations;

[0199] The human-machine interaction unit is used to provide a friendly user interface, allowing operators to easily monitor the system operating status, adjust control parameters and handle abnormal situations.

[0200] In this embodiment, the system control module can efficiently, stably and safely manage the operation of the new energy multi-energy complementary system, ensure the execution of the optimized scheduling plan, and achieve efficient operation of the system and maximize energy utilization.

[0201] Based on the same inventive concept, the present invention also provides a new energy multi-energy complementary monitoring method based on artificial intelligence, using the new energy multi-energy complementary monitoring system based on artificial intelligence as described above, to achieve real-time monitoring, data analysis and intelligent scheduling of multiple new energy sources. It can achieve effective coordination and optimized scheduling among multiple new energy sources, ensuring the stability and efficiency of energy supply.

[0202] Based on the power generation forecast results and fault diagnosis information, the intelligent dispatching system adjusts the power generation configuration of each renewable energy power station in real time to achieve wind-solar complementarity and coordinated dispatching of energy storage systems. During peak power demand periods, renewable energy power stations with higher power generation are prioritized; during off-peak periods, energy is stored for emergency use. In addition, the system also optimizes power generation strategies and improves overall power generation efficiency based on historical data and real-time environmental parameters.

[0203] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0204] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A new energy multi-energy complementary monitoring system based on artificial intelligence, characterized in that: include: The data acquisition module is used to collect various power generation data, meteorological data and equipment operation status data in real time through various sensors installed on various new energy power generation equipment; The data processing module cleans, stores and preprocesses the large amount of raw data collected; and constructs a sample data set; The intelligent analysis module uses sample data sets for analysis and training based on artificial intelligence algorithms to obtain a prediction model. Based on the prediction model, it predicts potential problems in system operation and provides an optimized scheduling solution. The system control module controls the operating status of various corresponding power generation equipment according to the scheduling plan provided by the intelligent analysis module, realizes multi-energy complementarity among new energy sources, and ensures stable and efficient operation of the system.

2. According to the artificial intelligence-based new energy multi-energy complementary monitoring system of claim 1, it is characterized in that: The data acquisition module comprises: Sensor units, which are a variety of sensors installed on various types of renewable energy power generation equipment, including temperature sensors, humidity sensors, wind speed and direction sensors, light intensity sensors, image sensors, position sensors, and current and voltage sensors; Data collector: used to receive data collected by the sensor unit, perform preliminary processing and data format conversion; A data transmission unit transmits data to a central processing unit or a cloud server via a wireless network or a wired network; The power management unit is used to supply power to the sensor unit and the data logger.

3. According to the artificial intelligence-based new energy multi-energy complementary monitoring system of claim 1, it is characterized in that: The data processing module comprises: Cleaning cells by interpolating, imputing, or deleting rows containing missing values; using statistical methods to detect and remove outliers; and removing noise from the data by using wavelet or Fourier transforms; The preprocessing unit performs standardization, normalization and dimension reduction on the cleaned data in sequence; The storage unit organizes structured data into columns according to features to form data tables and stores them in relational databases; it stores unstructured data in the form of documents in NoSQL databases.

4. According to claim 3, a new energy multi-energy complementary monitoring system based on artificial intelligence is characterized in that: The data processing module further includes: The fusion unit performs feature-level fusion or decision-level fusion processing on multi-source data.

5. According to the artificial intelligence-based new energy multi-energy complementary monitoring system of claim 1, it is characterized in that: The intelligent analysis module comprises: Data preparation and feature engineering unit, used for data preparation including obtaining sample data sets from the data processing module and performing feature engineering; Model selection and training unit, used to adopt deep learning models, suitable for processing time series data, to predict multi-energy complementary systems; or adopt machine learning models, suitable for classification and regression tasks, to identify potential problems in the system; The prediction and analysis unit uses the trained model to predict the new data obtained and analyze the results to identify potential problems in system operation; The optimization scheduling plan generation unit sets the optimization goal of maximizing energy output and minimizing energy loss based on the prediction and analysis results, and uses the optimization algorithm to generate a scheduling plan.

6. The artificial intelligence-based new energy multi-energy complementary monitoring system according to claim 5 is characterized in that: In the data preparation and feature engineering module, feature engineering includes: The feature selection unit is used to select the features that have a greater impact on the system operation, including temperature, humidity, wind speed and light intensity; Feature construction unit, constructs new features based on existing features, including: moving average and cumulative value of historical data; The feature scaling unit is used to standardize or normalize the features to ensure that the data ranges of different features are similar, so as to improve the model convergence speed and prediction accuracy.

7. The new energy multi-energy complementary monitoring system based on artificial intelligence according to claim 5 is characterized in that: The model selection and training unit adopts a deep learning model, which is suitable for processing time series data and predicting the multi-energy complementary system, specifically including: Select LSTM as the prediction network to predict time series data; Improve LSTM by increasing the number of LSTM layers to three, setting 100 units in each layer, and adding a Dropout layer after each layer; and add an Attention layer after the LSTM layer, and use the Concatenate layer to combine the LSTM output and the Attention output; The historical power generation data, equipment status data and corresponding environmental parameters of various new energy systems are used as sample data for training and optimization of the improved LSTM, and finally the optimized prediction model is obtained.

8. The new energy multi-energy complementary monitoring system based on artificial intelligence according to claim 1 is characterized in that: The system control module comprises: The optimization scheduling execution unit is used to receive the optimization scheduling plan generated by the intelligent analysis module, convert it into specific control instructions, and send it to the corresponding various types of new energy power generation equipment, and record the execution status of the scheduling instructions and provide feedback; Real-time monitoring unit is used to monitor the system operation status in real time to ensure the stable operation of the system under various working conditions; when an abnormal situation is detected, an alarm is issued in time and corresponding measures are given; Adaptive control unit, based on real-time monitoring data and feedback, automatically adjusts the system's power generation and operating parameters of start-stop equipment to adapt to environmental changes and load fluctuations; The human-machine interaction unit is used to provide a friendly user interface, allowing operators to easily monitor the system operating status, adjust control parameters and handle abnormal situations.

9. A new energy multi-energy complementary monitoring method based on artificial intelligence, characterized in that: By using the artificial intelligence-based new energy multi-energy complementary monitoring system as described in any one of claims 1 to 8, real-time monitoring, data analysis and intelligent scheduling of multiple new energy sources can be achieved.

Citation Information

Cited By

  • New energy power generation prediction system and method

    CN119253621A

  • Pulse power supply system parameter selection method and system

    CN119675488A

  • New energy power transmission state monitoring method and system based on artificial intelligence algorithm

    CN119944951A