Smart station energy management system and method based on Hongmeng system

Abstract

The present invention relates to the field of smart grid technology, and specifically to a smart station energy management system and method based on the Hongmeng system, including a distributed energy acquisition module, a data processing and integration module, an energy demand prediction module, an equipment failure prediction module, an energy optimization and scheduling module, and a user interaction module; wherein: the distributed energy acquisition module collects operating data; the data processing and integration module is used to form a distributed energy data set; the energy demand prediction module predicts future energy demand; the equipment failure prediction module is used to perform failure prediction; the energy optimization and scheduling module optimizes the operating strategy of each energy device in the scheduling station. The present invention optimizes energy scheduling, improves equipment operation reliability and system energy utilization efficiency through real-time data acquisition, intelligent prediction, and fault warning mechanisms, while enhancing operational transparency and convenience through a user-friendly interactive interface.

Description

Technical Field

[0001] The present invention relates to the field of smart grid technology, and in particular to a smart station energy management system and method based on the Hongmeng system. Background Art

[0002] With the development of smart grid technology, energy management in modern power stations is becoming increasingly complex, involving the coordination and optimized use of multiple energy forms, including the integrated management of solar energy, wind energy and energy storage equipment. Existing energy management systems usually rely on centralized control, which makes it difficult to process data from various distributed energy devices in real time and efficiently, and to dynamically adjust according to the real-time environment and needs. In addition, the ability to monitor the operating status of energy equipment, predict faults and accurately predict energy demand is relatively limited, and it is impossible to fully utilize the actual operating status of the equipment and environmental changes, resulting in inflexible energy scheduling and difficulty in achieving optimal energy utilization efficiency.

[0003] Existing technologies present numerous challenges in managing energy demand and equipment failures in power stations. Traditional systems lack real-time monitoring and early warning mechanisms for the operating status of energy equipment, making it impossible to detect potential equipment failures in advance, impacting system stability. Furthermore, energy scheduling decisions are often based on fixed rules, making it difficult to make intelligent adjustments based on real-time data and forecasts. This is especially true when multiple distributed energy devices coexist, making it difficult to achieve the rational allocation and efficient utilization of energy resources. Therefore, there is an urgent need for a smart power station energy management system and method based on the Hongmeng system to improve the overall efficiency and reliability of power station energy management. Summary of the Invention

[0004] Based on the above objectives, the present invention provides a smart station energy management system and method based on the Hongmeng system.

[0005] The smart station energy management system based on the Hongmeng system includes a distributed energy collection module, a data processing and integration module, an energy demand prediction module, an equipment failure prediction module, an energy optimization and scheduling module, and a user interaction module.

[0006] Distributed energy collection module: used to collect real-time operating data of distributed energy equipment in and around the station, including solar power generation equipment, wind power generation equipment and energy storage devices;

[0007] Data processing and integration module: Deployed in the distributed computing framework of the Hongmeng system, it is used to aggregate and pre-process the operating data collected by the distributed energy collection module, and store the processed data in the data center through distributed nodes to form a unified data set of distributed energy;

[0008] Energy Demand Forecasting Module: Based on the real-time computing capabilities of the Hongmeng system, the module uses the data set provided by the data processing and integration module, combined with the station's historical energy consumption records and real-time environmental data, to predict future energy demand and generate an energy usage plan for the next period of time;

[0009] Equipment Failure Prediction Module: Through the IoT connectivity of the Hongmeng system, the module monitors the operating status of various energy equipment in the station in real time, including status data of power equipment, power generation equipment, and energy storage equipment. Based on big data analysis and machine learning algorithms, the module conducts health assessments and fault predictions on the equipment's operating status, generating fault warning information.

[0010] Energy Optimization and Scheduling Module: Based on the energy demand forecast data provided by the energy demand forecast module and the status feedback from the equipment failure prediction module, it optimizes the operation strategy of each energy device in the dispatching station, prioritizes the equipment in good operating status, and rationally allocates and uses distributed energy and external grid resources to maximize energy efficiency.

[0011] User interaction module: Based on the multi-terminal interconnection capabilities of the Hongmeng system, it provides users with a real-time monitoring interface to display energy forecast results, equipment operating status and fault warning information.

[0012] Optionally, the distributed energy collection module includes a sensor unit, a data transmission unit, and a data synchronization unit; wherein:

[0013] Sensor unit: including photovoltaic current sensor, wind speed sensor, power generation sensor and power monitoring sensor, used to monitor and collect operating data of distributed energy equipment in and around the station building, such as the power generation power of solar power generation equipment, wind speed and power generation of wind power generation equipment, and power reserve and discharge status of energy storage devices;

[0014] Data transmission unit: Based on Wi-Fi or 5G communication network, it is connected to the sensor unit and is used to transmit the real-time data collected by the sensor unit;

[0015] Data synchronization unit: connected to the data transmission unit, used to synchronize the data transmitted to the data processing and integration module in real time. Specifically, each data packet is marked with a timestamp and a unique identifier to ensure that the data of different devices are synchronized according to time and source.

[0016] Optionally, the data processing and integration module includes a data aggregation unit, a data preprocessing unit, a distributed storage unit, and a data set integration unit; wherein:

[0017] Data aggregation unit: connected to the distributed energy collection module, used to receive real-time operating data from solar power generation equipment, wind power generation equipment and energy storage devices, and unify the data from different devices;

[0018] Data preprocessing unit: used to preprocess the aggregated data, including data cleaning, formatting, and normalization. Data cleaning is used to remove abnormal data points or incomplete data. Data formatting is used to unify distributed energy data from different sources into a standard format. Normalization is used to adjust values ​​of different dimensions to ensure that data from various energy devices can be compared and processed within a unified framework.

[0019] Distributed storage unit: used to uniformly store pre-processed data on distributed nodes. The distributed storage unit uses multiple storage nodes to partition and store data of different types of energy equipment. At the same time, a consistency algorithm is used to ensure that data on all nodes is kept updated synchronously to prevent data inconsistency or loss.

[0020] Data set integration unit: used to further integrate the data in the storage nodes, classify and aggregate the operating data from different energy devices to form a unified distributed energy data set.

[0021] Optionally, the energy demand forecasting module includes a historical data analysis unit, an environmental data acquisition unit, a Bayesian dynamic linear model unit, and an energy plan generation unit; wherein:

[0022] Historical data analysis unit: used to extract the station building's historical energy consumption records. The historical data includes energy usage, equipment operating status, and load data of various energy equipment. By analyzing historical energy consumption patterns, it identifies peak periods, low periods, and fluctuation patterns of energy usage, providing a basis for energy demand forecasting.

[0023] Environmental data acquisition unit: used to obtain real-time environmental data, including weather conditions, temperature, humidity, wind speed and light intensity information;

[0024] Bayesian dynamic linear model unit: This unit is connected to the historical data analysis unit and the environmental data acquisition unit. Based on the historical energy consumption data and real-time environmental data provided by the two units, it uses the Bayesian dynamic linear model to dynamically predict energy demand for a period of time in the future.

[0025] Energy plan generation unit: used to formulate energy usage plans for a period of time in the future based on the generated energy demand forecast results. Based on the future demand forecast, it reasonably allocates the use priority of solar energy, wind energy and energy storage equipment, and determines the operation strategy of the equipment to ensure the balance of energy supply and demand.

[0026] Optionally, the Bayesian dynamic linear model unit specifically includes:

[0027] Initialize model parameters: Initialize the parameters of the Bayesian dynamic linear model, including the initial state θ0, the state transfer matrix F, and the noise term ε t , and the covariance matrix Q;

[0028] State prediction: The energy demand state θ at the next moment is predicted using the state transition equation of the Bayesian dynamic linear model t Make predictions;

[0029] Observation prediction: The predicted energy demand state θ is transformed into t Converted to actual observable energy demand y t ;

[0030] Forecast update based on historical data: Using historical energy consumption data t , the model parameters are dynamically adjusted through the Bayesian update formula; according to Bayesian theorem, the predicted posterior probability is calculated by the following formula: Among them, p(θ t |y1,y2,…,y t-1 ) represents the posterior distribution of the energy demand state at the current time t; p(y t |θ t ) is the likelihood function of the current observation data; p(θ t |y1,y2,…,y t-1 ) is the prior distribution of the previous moment; p(y t ) is the normalization factor;

[0031] Prediction adjustment based on real-time environmental data: Introducing real-time environmental data X t , the prediction results are corrected by linear regression model, and the corrected energy demand prediction formula is: t ′=y t +βX t , where y t ′ is the revised energy demand forecast value; β is the impact coefficient of environmental data on energy demand; X t Represents the current environment variables;

[0032] Generate future energy demand forecast: Based on the results of the above steps, generate the energy demand forecast value y for a period of time in the future t+1 ,y t+2 ,…,y t+n , the energy demand at each moment in the future is dynamically adjusted by continuously updating the state transfer matrix F, the observation matrix H and the environmental correction coefficient β to ensure that the prediction results can reflect the changes in the environment and demand in real time.

[0033] Optionally, the equipment failure prediction module includes a data acquisition unit, a data processing unit, a health assessment unit, a failure prediction unit, and an early warning generation unit; wherein:

[0034] Data acquisition unit: used to collect real-time operating status data of various energy equipment in the station house, including equipment temperature, vibration, operating current, voltage and equipment operating time information;

[0035] Data processing unit: connected to the data acquisition unit, used to pre-process the collected device status data, including data cleaning and outlier processing;

[0036] Health Assessment Unit: Based on big data analysis technology, it uses the equipment's historical operation data and real-time status data to conduct a health assessment of the equipment's operating status. By analyzing the degree to which the equipment's performance indicators deviate from the normal range, it calculates the equipment's health score H. s , the specific formula is:

[0037] Among them, H s Score the device health, w i is the weight of each performance indicator, M i is the real-time measurement value, B i is the corresponding benchmark value, and n is the number of performance indicators of the equipment;

[0038] Fault prediction unit: It uses a decision tree-based machine learning algorithm to predict faults. It uses the marked equipment fault history data to train the decision tree model, analyzes the real-time status data of the equipment through classification nodes, predicts whether the equipment is about to fail, and outputs the equipment failure probability P. f ;

[0039] Warning generation unit: connected to the fault prediction unit, used to generate fault warning information, specifically when the fault probability P f When the preset threshold is exceeded, the warning generation unit will issue a fault warning message to inform the user of the potential fault type and time range.

[0040] Optionally, the fault prediction unit specifically includes:

[0041] Collect and label historical fault data: Use the equipment’s historical operating data and labeled fault dataset D hist , the dataset contains multi-dimensional features of the equipment operating status, including temperature T, current I, vibration V, and equipment status label L, where L = 1 indicates that the equipment is faulty and L = 0 indicates that the equipment is operating normally;

[0042] Training a decision tree model: Using a labeled dataset D hist The decision tree model is trained. The decision tree recursively divides the device status data space and selects the optimal node to divide the data based on the features. The specific calculation is based on the information gain. The formula is: Where H(L) represents the entropy of the fault label L, p i is feature T i The weight of H(L|T i ) is the characteristic T i Conditional entropy under the condition, by selecting the features with the largest information gain to construct classification nodes, and gradually generate a decision tree model;

[0043] Generate a decision tree model: After the decision tree model is generated, it can associate the device status characteristics with the device failure risk, and finally output the device's predicted failure result L through the device status data from the root node to the leaf node. pred , where L pred =1 indicates fault, L pred =0 means normal;

[0044] Real-time status data is input into the decision tree model: Real-time status data includes real-time temperature T(t), real-time current I(t), and real-time vibration parameter V(t). These data are collected by the acquisition unit and input into the decision tree model. The decision tree model compares the classification nodes of the status data in turn and gradually selects the path:

[0045] Predicting the probability of equipment failure: When real-time status data reaches the leaf node through the decision tree, the decision tree model calculates the equipment failure probability P f , the formula is: Among them, m is the total number of data samples reaching the leaf node, L i is the fault label for each sample.

[0046] Optionally, the energy optimization scheduling module includes an energy status evaluation unit, an equipment status monitoring unit, a priority allocation unit, an energy scheduling unit, and an energy storage management unit; wherein:

[0047] Energy status assessment unit: connected to the energy demand forecast module, used to t 'Evaluate energy demand in the future and assess current and future energy supply and demand;

[0048] Equipment status monitoring unit: used to monitor the operating status of energy equipment in real time and obtain fault prediction feedback information. The equipment status is measured by the fault prediction probability P f To measure, if the failure prediction probability P of a certain device f Below the pre-threshold P thereshold When Pf If the threshold is exceeded, the device will be placed in standby mode first;

[0049] Priority allocation unit: Based on the status monitoring results of the equipment and the energy supply and demand assessment, it assigns scheduling priorities to each device. The priority is determined by the real-time status of the equipment and the power generation efficiency η source Calculation, combined with the equipment failure probability P f Adjust the priority. The specific calculation formula is: U source (t) = η source ×(1-P f ), where U source (t) represents the priority of the device at time t, η source is the power generation efficiency of the equipment, P f Predict the probability of failure for the equipment;

[0050] Energy dispatch unit: According to the results of the priority allocation unit, dynamically adjust the output power of each energy device, specifically give priority to dispatching distributed energy devices with good status and high power generation efficiency to meet the revised energy demand forecast value y t '; If the supply of distributed energy equipment S distributed (t) cannot fully meet the revised demand y t ′, then dispatch external grid resources;

[0051] Energy storage management unit: When distributed energy supply exceeds demand S distributed (t)>y t ′, the energy storage device starts to charge and store excess energy; when the supply is insufficient S distributed (t) <y t ', the energy storage device discharges at a predetermined power P discharge Discharge to make up the difference.

[0052] Optionally, the user interaction module includes a data collection unit, an information display unit, a user feedback and control unit, and a multi-terminal access unit; wherein:

[0053] Data acquisition unit: connected to the energy demand forecast module, equipment failure forecast module and energy optimization scheduling module, used to obtain energy forecast results, equipment operation status and fault warning information in real time;

[0054] Information display unit: provides users with an intuitive monitoring interface through visualization, and generates charts and text information for display by classifying the collected data;

[0055] User feedback and control unit: This provides an interactive interface for users to manually intervene in the energy scheduling strategy. When users determine that the scheduling strategy of energy devices needs to be adjusted based on the information on the monitoring interface, they are allowed to modify the device priority, start and stop the device, or adjust the charging and discharging strategy of the energy storage device.

[0056] Multi-terminal access unit: supports users to access the monitoring interface through different terminal devices, ensuring that users can monitor and manage anytime and anywhere. The different terminal devices include computers, tablets and smartphones.

[0057] The smart station energy management method based on the Hongmeng system is implemented by the above-mentioned smart station energy management system based on the Hongmeng system, and includes the following steps:

[0058] S1: The distributed energy collection module collects real-time operating data of distributed energy equipment in and around the station building. The equipment includes solar power generation equipment, wind power generation equipment, and energy storage devices. The collected data includes power generation, energy storage status, and environmental data.

[0059] S2: preprocessing the operating data collected by S1, including data cleaning, formatting, and normalization; and integrating the preprocessed data to form a unified energy data set;

[0060] S3: Based on the energy dataset generated in S2, combined with historical energy consumption records and real-time environmental data, the Bayesian dynamic linear model is used to predict energy demand for a period of time in the future and generate a revised energy demand forecast value;

[0061] S4: By monitoring the operating status of each energy device in the station in real time and analyzing the equipment's operating data based on a pre-trained decision tree model, it generates equipment failure prediction results;

[0062] S5: Based on the revised energy demand forecast value in S3 and the equipment failure prediction results in S4, the energy optimization scheduling module is used to schedule equipment, giving priority to equipment in good condition and high power generation efficiency to ensure supply and demand balance;

[0063] S6: Provide users with a real-time monitoring interface through the user interaction module to display energy demand forecast results, equipment operating status and fault warning information.

[0064] Beneficial effects of the present invention:

[0065] This invention significantly improves the automation and intelligence level of energy management through the smart station energy management system based on the Hongmeng system. First, through the collection and predictive analysis of real-time data of distributed energy equipment, the system can effectively predict future energy demand and adjust the energy supply strategy in real time according to the prediction results. This not only ensures the stability of energy supply, but also optimizes the utilization efficiency of energy. Especially in the case of large fluctuations in energy supply and demand, it can achieve more flexible energy allocation and use, thereby reducing energy waste.

[0066] The present invention enhances the monitoring capability of the equipment operation status by integrating the equipment failure prediction module, and can timely discover and warn of potential equipment failures. This warning mechanism greatly improves the operation reliability of the equipment and reduces the downtime and maintenance costs caused by failures. At the same time, the implementation of the user interaction module enables the operator to monitor the operation status and scheduling results of the energy system in real time through an intuitive interface, further improving the transparency of management and the convenience of user operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0068] Figure 1 This is a schematic diagram of a smart station energy management system according to an embodiment of the present invention;

[0069] Figure 2 This is a flow chart of the smart station energy management method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0070] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments. It is also noted that, to provide a more detailed description, the following embodiments are best and preferred embodiments, and those skilled in the art may employ alternative methods for implementing certain known technologies. Furthermore, the accompanying drawings are intended only to provide a more detailed description of the embodiments and are not intended to limit the present invention.

[0071] It should be noted that references in the specification to "one embodiment," "an embodiment," "an exemplary embodiment," "some embodiments," etc. indicate that the described embodiments may include specific features, structures, or characteristics, but not every embodiment necessarily includes such specific features, structures, or characteristics. In addition, when specific features, structures, or characteristics are described in conjunction with an embodiment, it is within the knowledge of persons skilled in the relevant art to implement such features, structures, or characteristics in conjunction with other embodiments (whether or not explicitly described).

[0072] In general, terms can be understood, at least in part, from their use in context. For example, depending at least in part on the context, the term "one or more" as used herein can be used to describe any feature, structure, or characteristic in the singular sense, or can be used to describe a combination of features, structures, or characteristics in the plural sense. Additionally, the term "based on" can be understood as not necessarily intended to convey an exclusive set of factors, but can instead, depending at least in part on the context, allow for the presence of other factors that are not necessarily explicitly described.

[0073] like Figure 1 As shown in the figure, the smart station energy management system based on the Hongmeng system includes a distributed energy collection module, a data processing and integration module, an energy demand prediction module, an equipment failure prediction module, an energy optimization and scheduling module, and a user interaction module; among which:

[0074] Distributed energy collection module: used to collect real-time operating data of distributed energy equipment in and around the station, including solar power generation equipment, wind power generation equipment and energy storage devices;

[0075] Data processing and integration module: Deployed in the distributed computing framework of the Hongmeng system, it is used to aggregate and pre-process the operating data collected by the distributed energy acquisition module, and store the processed data in the data center through distributed nodes, forming a unified distributed energy data set, providing a basis for subsequent analysis;

[0076] Energy Demand Forecasting Module: Based on the real-time computing capabilities of the Hongmeng system, the module uses the data set provided by the data processing and integration module, combined with the station's historical energy consumption records and real-time environmental data, to predict future energy demand and generate an energy usage plan for the next period of time;

[0077] Equipment Failure Prediction Module: Through the IoT connectivity of the Hongmeng system, the module monitors the operating status of various energy equipment in the station in real time, including status data of power equipment, power generation equipment, and energy storage equipment. Based on big data analysis and machine learning algorithms, the module conducts health assessments and fault predictions on the equipment's operating status, generating fault warning information.

[0078] Energy Optimization and Scheduling Module: Based on the energy demand forecast data provided by the energy demand forecast module and the status feedback from the equipment failure prediction module, it optimizes the operation strategy of each energy device in the dispatching station, prioritizes the equipment in good operating status, and rationally allocates and uses distributed energy and external grid resources to maximize energy efficiency.

[0079] User interaction module: Based on the multi-terminal interconnection capabilities of the Hongmeng system, it provides users with a real-time monitoring interface to display energy forecast results, equipment operating status and fault warning information.

[0080] The distributed energy collection module includes a sensor unit, a data transmission unit, and a data synchronization unit; wherein:

[0081] Sensor unit: including photovoltaic current sensors, wind speed sensors, power generation sensors, and power monitoring sensors, used to monitor and collect operating data of distributed energy equipment in and around the station building. The operating data includes the power generation of solar power generation equipment, the wind speed and power generation of wind power generation equipment, and the power reserve and discharge status of energy storage devices;

[0082] Data transmission unit: Based on Wi-Fi or 5G communication network, it is connected to the sensor unit and is used to transmit the real-time data collected by the sensor unit with low latency and high bandwidth. Specifically, it uses a distributed network architecture to package and securely transmit data from multiple sensors to the data processing and integration module, ensuring fast and stable data transmission and uninterrupted connection;

[0083] Data synchronization unit: connected to the data transmission unit, used to synchronize the data transmitted to the data processing and integration module in real time. Specifically, each data packet is marked with a timestamp and a unique identifier to ensure that the data of different devices are synchronized according to time and source; through the synergy of the above units, the distributed energy acquisition module can ensure accurate and timely data collection, transmission and synchronization of the operating status of distributed energy equipment in and around the station, ensure the integrity, timeliness and security of the data, and provide basic data support for the energy management of the entire system.

[0084] The data processing and integration module includes a data aggregation unit, a data pre-processing unit, a distributed storage unit, and a data set integration unit; wherein:

[0085] Data aggregation unit: connected to the distributed energy collection module, used to receive real-time operating data from solar power generation equipment, wind power generation equipment and energy storage devices, and uniformly collect data from different devices, including but not limited to power generation power, wind speed, energy storage status and other information, to ensure that all types of data are accurately and completely transmitted to the subsequent processing unit;

[0086] Data preprocessing unit: used to preprocess the aggregated data, including data cleaning, formatting, and normalization. Data cleaning is used to remove abnormal data points or incomplete data. Data formatting is used to unify distributed energy data from different sources into a standard format. Normalization is used to adjust values ​​of different dimensions to ensure that data from various energy devices can be compared and processed within a unified framework.

[0087] Distributed storage unit: used to uniformly store pre-processed data on distributed nodes. The distributed storage unit uses multiple storage nodes to partition and store data of different types of energy equipment. At the same time, a consistency algorithm is used to ensure that data on all nodes is updated synchronously to prevent data inconsistency or loss.

[0088] Dataset integration unit: used to further integrate the data in the storage nodes, classify and aggregate the operating data from different energy devices to form a unified distributed energy data set; through the collaborative work of the above units, the data processing and integration module can efficiently aggregate, pre-process, store and integrate data from distributed energy devices, ensuring the formation of a complete and unified energy data set, providing reliable data support for subsequent energy demand forecasting and optimization scheduling modules.

[0089] The energy demand forecasting module includes a historical data analysis unit, an environmental data acquisition unit, a Bayesian dynamic linear model unit, and an energy plan generation unit; wherein:

[0090] Historical data analysis unit: used to extract the historical energy consumption records of the station building. The historical data includes energy usage, equipment operating status, and load data of various energy equipment. By analyzing the historical energy consumption patterns, the unit can identify the peak and valley periods and fluctuation patterns of energy usage, providing a basis for energy demand forecasting.

[0091] Environmental data acquisition unit: used to obtain real-time environmental data, including weather conditions, temperature, humidity, wind speed and light intensity information; this unit is connected to external meteorological data sources or local sensor systems to ensure that the acquired environmental data can accurately reflect the changes in external conditions at present and in the future;

[0092] Bayesian Dynamic Linear Model Unit: This unit is connected to the Historical Data Analysis Unit and the Environmental Data Acquisition Unit. Based on the historical energy consumption data and real-time environmental data provided by the two, it uses the Bayesian Dynamic Linear Model to dynamically predict energy demand for a period of time in the future. By introducing Bayesian inference, it integrates historical data, real-time environmental factors, and uncertainty factors to generate a probabilistic forecast of future energy demand. At the same time, it can be dynamically updated as new data is introduced to ensure the accuracy and timeliness of the forecast results.

[0093] Energy plan generation unit: used to formulate an energy usage plan for a period of time in the future based on the generated energy demand forecast results, and reasonably allocate the use priority of solar energy, wind energy and energy storage equipment based on future demand forecasts, and determine the equipment operation strategy to ensure the balance of energy supply and demand; through the collaborative work of the historical data analysis unit, environmental data acquisition unit, Bayesian dynamic linear model unit and energy plan generation unit, the energy demand forecast module can make high-precision predictions of future energy demand under various uncertain conditions, and generate a reasonable energy usage plan, providing a reliable basis for the system's energy scheduling and optimization.

[0094] The dynamic prediction of energy demand in the future in the Bayesian dynamic linear model unit specifically includes:

[0095] Initialize model parameters: Initialize the parameters of the Bayesian dynamic linear model, including the initial state θ0, the state transfer matrix F, and the noise term ε t , and the covariance matrix Q; the initial state θ0 represents the initial energy demand state of the system, which is determined according to the average value of historical energy consumption data; the state transition matrix F represents the change in energy demand of the system from the previous moment to the current moment, which is obtained by fitting historical data; the noise term ε t represents the random fluctuation or error in energy demand, which usually follows the normal distribution N(0, Q), where Q is the noise covariance matrix;

[0096] State prediction: The energy demand state θ at the next moment is predicted using the state transition equation of the Bayesian dynamic linear model t To make a prediction, the formula is: θ t =Fθ t-1 +ε t , where θ t represents the predicted energy demand state at the current moment, F is the state transition matrix, θ t-1 is the energy demand state at the previous moment, ε t is the random error at time t; through this formula, the model can dynamically update the energy demand forecast at each moment;

[0097] Observation prediction: The predicted energy demand state θ is transformed into t Converted to actual observable energy demand y t , the formula is: t =Hθ t +v t , where y t Represents the actual energy demand value at the current moment; H is the observation matrix, which is used to transform the state variable θ t Converted to observable quantity; θ t represents the predicted energy demand state at the current moment, v tis the observation noise, usually assuming v t ~N(0,Q),R is the covariance matrix of the observation noise. This step is used to link the internal state values ​​with the actual observable energy demand data;

[0098] Forecast update based on historical data: Using historical energy consumption data t , the model parameters are dynamically adjusted through the Bayesian update formula; according to Bayesian theorem, the predicted posterior probability is calculated by the following formula: Among them, p(θ t |y1,y2,…,y t-1 ) represents the posterior distribution of the energy demand state at the current time t; p(y t |θ t ) is the likelihood function of the current observation data; p(θ t |y1,y2,…,y t-1 ) is the prior distribution of the previous moment; p(y t ) is the normalization factor; through this formula, the model can correct the previous prediction based on the newly observed energy consumption data, so that the prediction is more consistent with the actual situation;

[0099] Prediction adjustment based on real-time environmental data: Introducing real-time environmental data X t (including weather, temperature, wind speed, etc.), the forecast results are corrected by linear regression model, and the revised energy demand forecast formula is: t ′=y t +βX t , where y t ′ is the revised energy demand forecast value; β is the impact coefficient of environmental data on energy demand; X t Represents the environmental variables at the current moment, obtained by fitting historical data. This step ensures that the model can dynamically consider the impact of changes in real-time environmental factors on energy demand;

[0100] Generate future energy demand forecast: Based on the results of the above steps, generate the energy demand forecast value y for a period of time in the future t+1 ,y t+2 ,…,y t+nThe energy demand at each moment in the future is dynamically adjusted by continuously updating the state transfer matrix F, the observation matrix H and the environmental correction coefficient β to ensure that the prediction results can reflect the changes in the environment and demand in real time; by combining the Bayesian dynamic linear model, the energy demand prediction module can dynamically process historical energy consumption data and real-time environmental data, and make accurate dynamic predictions of future energy demand. Bayesian updating and environmental data correction ensure the flexibility and accuracy of the prediction model, which is particularly suitable for situations where energy demand has high uncertainty and environmental dependence, and significantly improves the system's operating efficiency and prediction accuracy.

[0101] The equipment failure prediction module includes a data acquisition unit, a data processing unit, a health assessment unit, a failure prediction unit, and an early warning generation unit; wherein:

[0102] Data acquisition unit: used to collect real-time operating status data of various energy equipment in the station house, including equipment temperature, vibration, operating current, voltage and equipment operating time information. The data acquisition unit ensures that the status information of various equipment is continuously monitored and transmitted to the fault prediction module;

[0103] Data processing unit: connected to the data acquisition unit, used to pre-process the collected equipment status data. Pre-processing includes data cleaning and outlier processing to ensure that the processed data has good quality for subsequent analysis and prediction;

[0104] Health Assessment Unit: Based on big data analysis technology, it uses the equipment's historical operation data and real-time status data to conduct a health assessment of the equipment's operating status. By analyzing the degree to which the equipment's performance indicators (such as temperature, vibration, current, etc.) deviate from the normal range, it calculates the equipment's health score H. s , the specific formula is:

[0105] Among them, H s Score the device health, w i is the weight of each performance indicator, M i is the real-time measurement value, B i is the corresponding benchmark value, n is the number of performance indicators of the device. Through this score, the system can evaluate the health status of the device;

[0106] Fault prediction unit: It uses a decision tree-based machine learning algorithm to predict faults. It uses the marked equipment fault history data to train the decision tree model, analyzes the real-time status data of the equipment through classification nodes, predicts whether the equipment is about to fail, and outputs the equipment failure probability P. f ;

[0107] Warning generation unit: connected to the fault prediction unit, used to generate fault warning information, specifically when the fault probability Pf When the preset threshold is exceeded, the early warning generation unit will issue a fault warning message to inform the user of the potential fault type and time range. The early warning information will be promptly conveyed to the operation and maintenance personnel through the user interaction module to ensure that maintenance operations can be carried out in advance; through the collaborative work of the data acquisition unit, data processing unit, health assessment unit, fault prediction unit and early warning generation unit, the equipment fault prediction module can realize comprehensive health assessment and fault prediction of energy equipment in the station house, ensuring the safe and stable operation of the equipment.

[0108] The fault prediction unit specifically includes:

[0109] Collect and label historical fault data: Use the equipment’s historical operating data and labeled fault dataset D hist ,The dataset contains multi-dimensional features of the equipment operating ,status, including temperature T, current I, vibration V, and the equipment ,status label L, where when L = 1 it indicates that the equipment ,faults occur, and L = 0 indicates that the equipment ,operates normally;

[0110] Training a decision tree model: Using a labeled dataset D hist The decision tree model is trained. The decision tree recursively divides the device status data space and selects the optimal node to divide the data based on the features. The specific calculation is based on the information gain. The formula is: Where H(L) represents the entropy of the fault label L, p i is feature T i The weight of H(L|T i ) is the characteristic T i Conditional entropy under the condition, by selecting the features with the largest information gain to construct classification nodes, and gradually generate a decision tree model;

[0111] Generate a decision tree model: After the decision tree model is generated, it can associate the device status characteristics (such as temperature T, current I, vibration V) with the device failure risk, and finally output the device's predicted failure result L from the root node to the leaf node through the device's status data. pred , where L pred =1 indicates fault, L pred =0 means normal;

[0112] Real-time status data is input into the decision tree model: the real-time status data includes real-time temperature T(t), real-time current I(t) and real-time vibration parameter V(t), which are collected by the acquisition unit and input into the decision tree model. The decision tree model compares the classification nodes of the status data in turn and gradually selects the path; each node will be based on the current status data and the set threshold T threshold ,I threshold ,V tkreskold Compare and select the optimal path. The specific selection expression is ifT(t)>Tthreshold ,goright;else,go left:

[0113] Predicting the probability of equipment failure: When real-time status data reaches the leaf node through the decision tree, the decision tree model calculates the equipment failure probability P f , the formula is: Among them, m is the total number of data samples reaching the leaf node, L i is the fault label for each sample; by calculating the proportion of fault samples in historical data, the probability P of the current fault of the device is obtained f Through the above steps, the fault prediction unit can use historical data to train a decision tree model, and analyze the failure probability of the equipment through real-time data, providing effective protection for the safety of equipment operation.

[0114] The energy optimization and scheduling module includes an energy status assessment unit, an equipment status monitoring unit, a priority allocation unit, an energy scheduling unit, and an energy storage management unit; among which:

[0115] Energy status assessment unit: connected to the energy demand forecast module, used to t ′ (obtained by Bayesian dynamic linear model unit combined with real-time environmental data correction) to evaluate the energy demand in the future period, and at the same time combined with the current output power P of distributed energy equipment (including solar power generation equipment, wind power generation equipment and energy storage equipment) solar (t),P wind (t) and the available power E of the energy storage device storage (t), assess the current and future state of energy supply and demand;

[0116] Equipment status monitoring unit: used to monitor the operating status of energy equipment in real time and obtain fault prediction feedback information. The equipment status is measured by the fault prediction probability P f To measure, if the failure prediction probability P of a certain device f Below the pre-threshold P thereshold When P f If the threshold is exceeded, the device will be placed in standby mode first;

[0117] Priority allocation unit: Based on the status monitoring results of the equipment and the energy supply and demand assessment, it assigns scheduling priorities to each device. The priority is determined by the real-time status of the equipment and the power generation efficiency η source Calculation, combined with the equipment failure probability P f Adjust the priority. The specific calculation formula is: U source (t) = η source ×(1-P f ), where U source(t) represents the priority of the device at time t, η source is the power generation efficiency of the equipment, P f is the predicted probability of equipment failure. Based on this formula, equipment in good condition and with high power generation efficiency is dispatched first.

[0118] Energy dispatch unit: According to the results of the priority allocation unit, dynamically adjust the output power of each energy device, specifically give priority to dispatching distributed energy devices with good status and high power generation efficiency to meet the revised energy demand forecast value y t '; If the supply of distributed energy equipment S distributed (t) cannot fully meet the revised demand y t ′, then dispatch external grid resources, assuming the supply and demand difference is ΔP(t), then the calculation formula is: ΔP(t)=y t ′-S distrbuted (t); if ΔP(t)>0, the external power grid is dispatched to provide supplementary power; if ΔP(t)≤0, the power supply is completely provided by distributed energy devices;

[0119] Energy storage management unit: When distributed energy supply exceeds demand S distributed (t)>y t ′, the energy storage device starts to charge and store excess energy; when the supply is insufficient S distributed (t) <y t ', the energy storage device discharges at a predetermined power P discharge Discharge to make up the difference; the specific strategies are as follows:

[0120]

[0121] Among them, P charwge is the charging power of the energy storage device, P discharge is the discharge power of the energy storage device; through the above units, the energy optimization scheduling module can efficiently realize the dynamic scheduling of energy equipment and the optimal allocation of energy resources based on the revised energy demand forecast data and equipment status feedback.

[0122] The user interaction module includes a data acquisition unit, an information display unit, a user feedback and control unit, and a multi-terminal access unit; wherein:

[0123] Data acquisition unit: connected to the energy demand forecast module, equipment failure forecast module and energy optimization scheduling module, used to obtain energy forecast results, equipment operation status and fault warning information in real time;

[0124] Information display unit: provides users with an intuitive monitoring interface through visualization, and generates charts and text information for display by classifying the collected data;

[0125] User feedback and control unit: This provides an interactive interface for users to manually intervene in the energy scheduling strategy. When users determine that the scheduling strategy of energy devices needs to be adjusted based on the information on the monitoring interface, they are allowed to modify the device priority, start and stop the device, or adjust the charging and discharging strategy of the energy storage device.

[0126] Multi-terminal access unit: supports users to access the monitoring interface through different terminal devices, including computers, tablets and smartphones, ensuring that users can monitor and manage anytime and anywhere. Through the collaborative work of these units, the user interaction module can provide users with comprehensive support for real-time monitoring, information display and operation feedback, ensuring that the operation status of the energy management system is transparent and intuitive, and facilitating effective management and intervention by users.

[0127] like Figure 2 As shown, the smart station energy management method based on the Hongmeng system is implemented by the above-mentioned smart station energy management system based on the Hongmeng system, and includes the following steps:

[0128] S1: The distributed energy collection module collects real-time operating data of distributed energy equipment in and around the station building. The equipment includes solar power generation equipment, wind power generation equipment, and energy storage devices. The collected data includes power generation, energy storage status, and environmental data.

[0129] S2: Preprocesses the operating data collected by S1, including data cleaning, formatting, and normalization; and integrates the preprocessed data to form a unified energy data set for subsequent forecasting and scheduling.

[0130] S3: Based on the energy dataset generated in S2, combined with historical energy consumption records and real-time environmental data, the Bayesian dynamic linear model is used to predict energy demand for a period of time in the future and generate a revised energy demand forecast value;

[0131] S4: By monitoring the operating status of each energy device in the station in real time and analyzing the equipment's operating data based on a pre-trained decision tree model, it generates equipment failure prediction results;

[0132] S5: Based on the revised energy demand forecast value in S3 and the equipment failure prediction results in S4, the energy optimization scheduling module is used to schedule equipment, giving priority to equipment in good condition and high power generation efficiency to ensure supply and demand balance;

[0133] S6: Provide users with a real-time monitoring interface through the user interaction module to display energy demand forecast results, equipment operating status and fault warning information.

[0134] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0135] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. The smart station energy management system based on Hongmeng system is characterized by: It includes distributed energy acquisition module, data processing and integration module, energy demand forecasting module, equipment failure forecasting module, energy optimization scheduling module and user interaction module; among which: Distributed energy collection module: used to collect real-time operating data of distributed energy equipment in and around the station, including solar power generation equipment, wind power generation equipment and energy storage devices; Data processing and integration module: Deployed in the distributed computing framework of the Hongmeng system, it is used to aggregate and pre-process the operating data collected by the distributed energy collection module, and store the processed data in the data center through distributed nodes to form a unified data set of distributed energy; Energy Demand Forecasting Module: Based on the real-time computing capabilities of the Hongmeng system, the module uses the data set provided by the data processing and integration module, combined with the station's historical energy consumption records and real-time environmental data, to predict future energy demand and generate an energy usage plan for the next period of time; The energy demand forecasting module includes a historical data analysis unit, an environmental data acquisition unit, a Bayesian dynamic linear model unit, and an energy plan generation unit; wherein: Historical data analysis unit: used to extract the station building's historical energy consumption records. The historical data includes energy usage, equipment operating status, and load data of various energy equipment. By analyzing historical energy consumption patterns, it identifies peak periods, low periods, and fluctuation patterns of energy usage, providing a basis for energy demand forecasting. Environmental data acquisition unit: used to obtain real-time environmental data, including weather conditions, temperature, humidity, wind speed and light intensity information; Bayesian dynamic linear model unit: This unit is connected to the historical data analysis unit and the environmental data acquisition unit. Based on the historical energy consumption data and real-time environmental data provided by the two units, it uses the Bayesian dynamic linear model to dynamically predict energy demand for a period of time in the future. Energy plan generation unit: used to formulate energy usage plans for a period of time based on the generated energy demand forecast results. Based on the future demand forecast, it reasonably allocates the use priority of solar energy, wind energy and energy storage equipment, and determines the operation strategy of the equipment to ensure the balance of energy supply and demand; The Bayesian dynamic linear model unit specifically includes: Initialize model parameters: Initialize the parameters of the Bayesian dynamic linear model, including the initial state θ0, the state transfer matrix F, and the noise term ε t , and the covariance matrix Q; State prediction: The energy demand state θ at the next moment is predicted using the state transition equation of the Bayesian dynamic linear model t Make predictions; Observation prediction: The predicted energy demand state θ is transformed into t Converted to actual observable energy demand y t ; Forecast update based on historical data: Using historical energy consumption data t , the model parameters are dynamically adjusted through the Bayesian update formula; according to Bayesian theorem, the predicted posterior probability is calculated by the following formula: Among them, p(θ t |y1,y2,…,y t-1 ) represents the posterior distribution of the energy demand state at the current time t; p(y t |θ t ) is the likelihood function of the current observation data; p(θ t |y1,y2,…,y t-1 ) is the prior distribution of the previous moment; p(y t ) is the normalization factor; Prediction adjustment based on real-time environmental data: Introducing real-time environmental data X t , the prediction results are corrected by linear regression model, and the corrected energy demand prediction formula is: t ′=y t +βX t , where y t ′ is the revised energy demand forecast value; β is the impact coefficient of environmental data on energy demand; X t Represents the current environment variables; Generate future energy demand forecast: Based on the results of the above steps, generate the energy demand forecast value y for a period of time in the future t+1 ,y t+2 ,…,y t+n ,The energy demand at each moment in the future is dynamically adjusted by continuously updating the state transfer matrix F, the observation matrix H and the environmental correction coefficient β to ensure that the prediction results can reflect the changes in the environment and demand in real time; Equipment Failure Prediction Module: Through the IoT connectivity of the Hongmeng system, the module monitors the operating status of various energy equipment in the station in real time, including status data of power equipment, power generation equipment, and energy storage equipment. Based on big data analysis and machine learning algorithms, the module conducts health assessments and fault predictions on the equipment's operating status, generating fault warning information. Energy Optimization and Scheduling Module: Based on the energy demand forecast data provided by the energy demand forecast module and the status feedback from the equipment failure prediction module, it optimizes the operation strategy of each energy device in the dispatching station, prioritizes the equipment in good operating status, and rationally allocates and uses distributed energy and external grid resources to maximize energy efficiency. User interaction module: Based on the multi-terminal interconnection capabilities of the Hongmeng system, it provides users with a real-time monitoring interface to display energy forecast results, equipment operating status and fault warning information.

2. The smart station energy management system based on Hongmeng system according to claim 1 is characterized in that: The distributed energy collection module includes a sensor unit, a data transmission unit and a data synchronization unit; wherein: Sensor unit: including photovoltaic current sensor, wind speed sensor, power generation sensor and power monitoring sensor, used to monitor and collect operating data of distributed energy equipment in and around the station building, such as the power generation power of solar power generation equipment, wind speed and power generation of wind power generation equipment, and power reserve and discharge status of energy storage devices; Data transmission unit: Based on Wi-Fi or 5G communication network, it is connected to the sensor unit and is used to transmit the real-time data collected by the sensor unit; Data synchronization unit: connected to the data transmission unit, used to synchronize the data transmitted to the data processing and integration module in real time. Specifically, each data packet is marked with a timestamp and a unique identifier to ensure that the data of different devices are synchronized according to time and source.

3. The smart station energy management system based on Hongmeng system according to claim 1 is characterized in that: The data processing and integration module includes a data aggregation unit, a data pre-processing unit, a distributed storage unit, and a data set integration unit; wherein: Data aggregation unit: connected to the distributed energy collection module, used to receive real-time operating data from solar power generation equipment, wind power generation equipment and energy storage devices, and unify the data from different devices; Data preprocessing unit: used to preprocess the aggregated data, including data cleaning, formatting, and normalization. Data cleaning is used to remove abnormal data points or incomplete data. Data formatting is used to unify distributed energy data from different sources into a standard format. Normalization is used to adjust values ​​of different dimensions to ensure that data from various energy devices can be compared and processed within a unified framework. Distributed storage unit: used to uniformly store pre-processed data on distributed nodes. The distributed storage unit uses multiple storage nodes to partition and store data of different types of energy equipment. At the same time, a consistency algorithm is used to ensure that data on all nodes is kept updated synchronously to prevent data inconsistency or loss. Data set integration unit: used to further integrate the data in the storage nodes, classify and aggregate the operating data from different energy devices to form a unified distributed energy data set.

4. The smart station energy management system based on Hongmeng system according to claim 1 is characterized in that: The equipment failure prediction module includes a data acquisition unit, a data processing unit, a health assessment unit, a failure prediction unit, and an early warning generation unit; wherein: Data acquisition unit: used to collect real-time operating status data of various energy equipment in the station house, including equipment temperature, vibration, operating current, voltage and equipment operating time information; Data processing unit: connected to the data acquisition unit, used to pre-process the collected device status data, including data cleaning and outlier processing; Health Assessment Unit: Based on big data analysis technology, it uses the equipment's historical operation data and real-time status data to conduct a health assessment of the equipment's operating status. By analyzing the degree to which the equipment's performance indicators deviate from the normal range, it calculates the equipment's health score H. s , the specific formula is: Among them, H s Score the device health, w i is the weight of each performance indicator, M i is the real-time measurement value, B i is the corresponding benchmark value, and n is the number of performance indicators of the equipment; Fault prediction unit: It uses a decision tree-based machine learning algorithm to predict faults. It uses the marked equipment fault history data to train the decision tree model, analyzes the real-time status data of the equipment through classification nodes, predicts whether the equipment is about to fail, and outputs the equipment failure probability P. f ; Warning generation unit: connected to the fault prediction unit, used to generate fault warning information, specifically when the fault probability P f When the preset threshold is exceeded, the warning generation unit will issue a fault warning message to inform the user of the potential fault type and time range.

5. The smart station energy management system based on Hongmeng system according to claim 4 is characterized in that: The fault prediction unit specifically includes: Collect and label historical fault data: Use the equipment’s historical operating data and labeled fault dataset D hist , the dataset contains multi-dimensional features of the equipment operating status, including temperature T, current I, vibration V, and equipment status label L, where L = 1 indicates that the equipment is faulty and L = 0 indicates that the equipment is operating normally; Training a decision tree model: Using a labeled dataset D hist The decision tree model is trained. The decision tree recursively divides the device status data space and selects the optimal node to divide the data based on the features. The specific calculation is based on the information gain. The formula is: Where H(L) represents the entropy of the fault label L, p i is feature T i The weight of H(L|T i ) is the characteristic T i Conditional entropy under the condition, by selecting the features with the largest information gain to construct classification nodes, and gradually generate a decision tree model; Generate a decision tree model: After the decision tree model is generated, it can associate the device status characteristics with the device failure risk, and finally output the device's predicted failure result L through the device status data from the root node to the leaf node. pred , where L pred =1 indicates fault, L pred =0 means normal; Real-time status data is input into the decision tree model: Real-time status data includes real-time temperature T(t), real-time current I(t), and real-time vibration parameter V(t). These data are collected by the acquisition unit and input into the decision tree model. The decision tree model compares the classification nodes of the status data in turn and gradually selects the path: Predicting the probability of equipment failure: When real-time status data reaches the leaf node through the decision tree, the decision tree model calculates the equipment failure probability P f , the formula is: Among them, m is the total number of data samples reaching the leaf node, L i is the fault label for each sample.

6. The smart station energy management system based on Hongmeng system according to claim 1 is characterized in that: The energy optimization and scheduling module includes an energy status evaluation unit, an equipment status monitoring unit, a priority allocation unit, an energy scheduling unit, and an energy storage management unit; wherein: Energy status assessment unit: connected to the energy demand forecast module, used to t 'Evaluate energy demand in the future and assess current and future energy supply and demand; Equipment status monitoring unit: used to monitor the operating status of energy equipment in real time and obtain fault prediction feedback information. The equipment status is measured by the fault prediction probability P f To measure, if the failure prediction probability P of a certain device f Below the pre-threshold P thereshold When P f If the threshold is exceeded, the device will be placed in standby mode first; Priority allocation unit: Based on the status monitoring results of the equipment and the energy supply and demand assessment, it assigns scheduling priorities to each device. The priority is determined by the real-time status of the equipment and the power generation efficiency η source Calculation, combined with the equipment failure probability P f Adjust the priority. The specific calculation formula is: U source (t) = η source ×(1-P f ), where U source (t) represents the priority of the device at time t, η source is the power generation efficiency of the equipment, P f Predict the probability of failure for the equipment; Energy dispatch unit: According to the results of the priority allocation unit, dynamically adjust the output power of each energy device, specifically give priority to dispatching distributed energy devices with good status and high power generation efficiency to meet the revised energy demand forecast value y t '; If the supply of distributed energy equipment S distributed (t) cannot fully meet the revised demand y t ′, then dispatch external grid resources; Energy storage management unit: When distributed energy supply exceeds demand S distributed (t)>y t ′, the energy storage device starts to charge and store excess energy; when the supply is insufficient S distributed (t) <y t ', the energy storage device discharges at a predetermined power P discharge Discharge to make up the difference.

7. The smart station energy management system based on Hongmeng system according to claim 1 is characterized in that: The user interaction module includes a data acquisition unit, an information display unit, a user feedback and control unit, and a multi-terminal access unit; wherein: Data acquisition unit: connected to the energy demand forecast module, equipment failure forecast module and energy optimization scheduling module, used to obtain energy forecast results, equipment operation status and fault warning information in real time; Information display unit: provides users with an intuitive monitoring interface through visualization, and generates charts and text information for display by classifying the collected data; User feedback and control unit: This provides an interactive interface for users to manually intervene in the energy scheduling strategy. When users determine that the scheduling strategy of energy devices needs to be adjusted based on the information on the monitoring interface, they are allowed to modify the device priority, start and stop the device, or adjust the charging and discharging strategy of the energy storage device. Multi-terminal access unit: supports users to access the monitoring interface through different terminal devices, ensuring that users can monitor and manage anytime and anywhere. The different terminal devices include computers, tablets and smartphones.

8. The smart station energy management method based on the Hongmeng system is implemented by the smart station energy management system based on the Hongmeng system according to any one of claims 1 to 7, characterized in that: The following steps are involved: S1: The distributed energy collection module collects real-time operating data of distributed energy equipment in and around the station building. The equipment includes solar power generation equipment, wind power generation equipment, and energy storage devices. The collected data includes power generation, energy storage status, and environmental data. S2: preprocessing the operating data collected by S1, including data cleaning, formatting, and normalization; and integrating the preprocessed data to form a unified energy data set; S3: Based on the energy dataset generated in S2, combined with historical energy consumption records and real-time environmental data, the Bayesian dynamic linear model is used to predict energy demand for a period of time in the future and generate a revised energy demand forecast value; S4: By monitoring the operating status of each energy device in the station in real time and analyzing the equipment's operating data based on a pre-trained decision tree model, it generates equipment failure prediction results; S5: Based on the revised energy demand forecast value in S3 and the equipment failure prediction results in S4, the energy optimization scheduling module is used to schedule equipment, giving priority to equipment in good condition and high power generation efficiency to ensure supply and demand balance; S6: Provide users with a real-time monitoring interface through the user interaction module to display energy demand forecast results, equipment operating status and fault warning information.

Images