Intelligent Dispatching System for Source-Grid-Load-Storage Integrated Microgrid

Abstract

The present invention discloses an intelligent scheduling system for a source-network-load-storage integrated microgrid, which relates to the technical field of microgrid management and includes an integrated scheduling platform. The integrated scheduling platform is communicatively connected to an energy data acquisition module, an energy prediction module, an intelligent scheduling module, an energy storage management module, and a fault warning module. Among them, the electrical signals are connected between the modules. By integrating sensor networks, data acquisition, and processing technologies, the present invention can monitor the operating states of various energy devices and energy storage devices in the microgrid in real time, significantly improving the reliability and stability of the system. It can detect and warn potential equipment failures in a timely manner, avoid the expansion of faults, reduce the power outage time and economic losses caused by faults, and can flexibly adjust the charge and discharge strategies of energy storage devices according to real-time load demands and renewable energy generation conditions, ensuring that the microgrid can maintain a stable power supply under various working conditions and meet the electricity consumption needs of users.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrid management, and particularly to an intelligent scheduling system for a source-grid-load-storage integrated microgrid. Background Art

[0002] With the accelerating advancement of the global energy transformation, the proportion of new energy sources such as wind power and photovoltaic power is continuously increasing. The problem of the consumption and utilization rate of new energy is one of the important challenges in the current energy field. Digital technologies have emerged. Through the application of digital technologies, the operation of the power system becomes more intelligent and efficient, and the coordinated operation and intelligent scheduling of energy production, transmission, consumption, and storage are realized.

[0003] For example, in a microgrid intelligent monitoring and scheduling method with the Chinese patent publication number: CN118300195A, the method includes: modeling to obtain a power generation power model under different energy forms, collecting the energy information of different power generation energies at the current moment, predicting and calculating the power generation power corresponding to different power generation energies at a future moment by using the power generation power model, modeling the scheduling process of the microgrid multi-energy storage system, constructing a microgrid optimal scheduling objective function, and solving it.

[0004] In the prior art, according to the power generation power corresponding to different power generation energies at a future moment, the scheduling process of the microgrid multi-energy storage system is modeled, a microgrid optimal scheduling objective function with the highest energy storage efficiency of the microgrid multi-energy storage system as the goal is constructed and solved to obtain the optimal scheduling strategy at a future moment. However, in a source-grid-load-storage integrated microgrid, there are various types of energy and energy storage devices, and the complementarity and coordination between devices are poor. Moreover, due to the intermittency and uncertainty of new energy power generation such as wind power and photovoltaic power, it is difficult to predict their power generation, which will lead to a deviation between the microgrid scheduling plan and the actual power generation situation, and thus it is difficult to accurately monitor the abnormal state during the operation of the power grid. Therefore, an intelligent scheduling system for a source-grid-load-storage integrated microgrid is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent scheduling system for a source-grid-load-storage integrated microgrid to solve the problems proposed in the above background art.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is:

[0007] An intelligent scheduling system for a source-grid-load-storage integrated microgrid includes an integrated scheduling platform, and the integrated scheduling platform is communicatively connected to an energy data acquisition module, an energy prediction module, an intelligent scheduling module, an energy storage management module, and a fault warning module, wherein, the modules are electrically connected to each other;

[0008] The energy data acquisition module is used to collect the original operation data and status information of wind power, photovoltaic energy equipment and energy storage equipment in the microgrid, providing accurate and comprehensive data support for subsequent dispatching decisions, and ensuring the accuracy and real-time nature of the dispatching plan;

[0009] The energy prediction module uses the variational mode decomposition algorithm to predict the new energy power generation of wind power and photovoltaic, and analyzes the change trend of the load, improving the accuracy of power generation prediction and reducing the dispatching deviation caused by prediction deviation;

[0010] The intelligent dispatching module, based on the information of the energy data acquisition module and the energy prediction module, uses the sparrow search algorithm to analyze and formulate a dispatching plan, realizing the configuration and dispatching of various types of energy and energy storage equipment in the microgrid, improving the energy utilization efficiency and reducing the operation cost;

[0011] The energy storage management module is used to classify and manage the charging and discharging process of the energy storage equipment, controlling the process of releasing and storing electric energy in the microgrid, improving the response speed and efficiency of the energy storage system, and enhancing the flexibility and stability of the microgrid;

[0012] The fault warning module is used to monitor the operation status of various types of energy equipment and energy storage equipment in the microgrid in real time, identify abnormal patterns during operation, and send an alarm signal when a fault occurs, improving the system's fault response speed and fault handling ability, and reducing the losses caused by faults.

[0013] A further improvement of the technical solution of the present invention lies in: in the energy data acquisition module, the process of obtaining the original operation data and status information is as follows:

[0014] Deploy a data acquisition system and a load monitoring system. By connecting with the sensors and controllers of wind power equipment, photovoltaic equipment, and energy storage batteries, various types of energy equipment and energy storage equipment in the microgrid are connected to the data acquisition system. Among them, wind power equipment and photovoltaic equipment are energy equipment, and energy storage batteries are energy storage equipment, and the load monitoring system is used to monitor the load status in the microgrid;

[0015] The original operation data and status information of energy equipment and energy storage equipment are obtained in real time through sensors. Among them, the operation data of energy equipment include the operation parameters of power generation, power, voltage, and current of wind power and photovoltaic equipment, and the status information of energy storage equipment includes the charge and discharge status and remaining capacity status information of energy storage batteries;

[0016] The collected original data is transmitted to the integrated dispatching platform through a wireless communication network. During the transmission process, it is necessary to ensure the integrity and security of the data to avoid data loss or tampering. After receiving the original data, the integrated dispatching platform stores it classified according to the time sequence and equipment type;

[0017] Preprocess the collected raw data, where the preprocessing includes data cleaning, screening, and formatting steps to remove error data and noise and convert the data into a format suitable for analysis.

[0018] A further improvement of the technical solution of the present invention lies in that: in the energy prediction module, the process of predicting the power generation and analyzing the load change trend is as follows:

[0019] Collect historical power generation data and load data from wind power, photovoltaic equipment, and load monitoring systems, and perform data cleaning and normalization processing on the collected data;

[0020] Input the time series data of the historical power generation data of wind power and photovoltaic into the variational mode decomposition algorithm respectively. Through the iterative optimization process, each time series is decomposed into multiple intrinsic mode functions. Each intrinsic mode function represents a specific frequency component in the power generation time series, and analyze the central frequency and bandwidth characteristics of each intrinsic mode function to extract the characteristic information in the power generation time series;

[0021] Based on the extracted characteristic information and the decomposed intrinsic mode functions, use the historical power generation data and the corresponding intrinsic mode functions as inputs, combine with a convolutional neural network to construct an energy prediction model, and use the trained energy prediction model to predict the power generation of wind power and photovoltaic;

[0022] Use the variational mode decomposition algorithm to decompose the time series data of the load data to obtain multiple intrinsic mode functions, analyze the change trend of each intrinsic mode function, identify the periodic and non-periodic components in the load, and analyze the change trend of the load;

[0023] Combined with the power generation prediction results and load change trend analysis, predict the future power generation and load, which helps the power system to better balance supply and demand and make scheduling decisions.

[0024] A further improvement of the technical solution of the present invention lies in that: the calculation expression for predicting the power generation of wind power and photovoltaic is:

[0025]

[0026]

[0027] Where P t is the predicted power generation, IMF i (t) is the i-th intrinsic mode function at time t during power generation prediction, n is the number of intrinsic mode functions during power generation prediction, and α is an adjustment factor used to adjust the influence of the sum. is the reference value, representing the historical average power generation. γ is the radical parameter used to adjust the non-linear effect of the sum. λ is the reference value of the exponential function. S is the time variable, representing the prediction time point. K i is the number of terms of the i-th IMF during power generation prediction. a k is the amplitude of the k-th term during power generation prediction. f k，i is the central frequency of the k-th term during power generation prediction. θ k，i is the phase of the k-th term during power generation prediction;

[0028] The calculation expression for analyzing the load change trend is:

[0029]

[0030]

[0031] where L t is the predicted load at time t, is the reference value, representing the average of historical load data. w j is the weight of the j-th intrinsic mode function during load prediction, indicating the contribution degree of this mode to load prediction. IMF j (t) is the value of the j-th intrinsic mode function at time t during load prediction, representing the specific frequency component in the load time series. m is the number of intrinsic mode functions during load prediction. ∈ t is the random noise term. θ is the model parameter, representing the non-linear trend of load change. T b is the time variable. K j is the number of terms of the j-th IMF during load prediction. a u is the amplitude of the u-th term during load prediction. f u，j is the central frequency of the u-th term during load prediction. φ u，j is the phase of the u-th term during load prediction.

[0032] A further improvement of the technical solution of the present invention is that in the intelligent scheduling module, the process of configuring and scheduling various energy and energy storage devices in the microgrid is as follows:

[0033] Obtain real-time operation data and historical data from the energy data acquisition module, including the power generation of wind power and photovoltaic power, the status of energy storage devices (electric quantity, power limit), and load demand;

[0034] Use the energy prediction module to predict the power generation of wind power and photovoltaic power and the load demand based on historical data and current weather environment parameters, and integrate the prediction results into the input of the intelligent scheduling module as the basis for scheduling decisions;

[0035] The scheduling problem of the microgrid is defined as an optimization problem using the Sparrow Search Algorithm, and iterative calculations are performed to converge the algorithm to find the optimal scheduling strategy;

[0036] After the algorithm converges, the result with the highest fitness is selected as the optimal solution, which is the optimal energy and energy storage device configuration and scheduling scheme. The optimal solution found by the algorithm is converted into specific operation instructions, including adjusting the output power of the power generation equipment, controlling the charge and discharge of the energy storage device, sending the scheduling instructions to the corresponding energy and energy storage devices through the communication network, and monitoring the execution situation.

[0037] A further improvement of the technical solution of the present invention lies in that: in the energy storage management module, the process of controlling the release and storage of electric energy in the microgrid is as follows:

[0038] Collect data of the energy storage device through sensors and the data acquisition system, monitor the state of the energy storage device, including the battery level, charge and discharge state, and health state, and integrate data from the energy data acquisition module and the energy prediction module, and collect data related to energy storage, including the power generation of wind power and photovoltaic, the state of the energy storage device, and load demand information;

[0039] According to factors such as grid demand, power generation prediction, current market price, and equipment efficiency, formulate specific charge and discharge strategies, analyze the capacity limit of the energy storage device, charge and discharge efficiency, and the overall operation requirements of the microgrid, and convert the formulated charge and discharge strategies into specific operation instructions and send them to the energy storage device for execution;

[0040] Real-time monitor the operation state of the energy storage device and the power supply and demand situation of the microgrid to ensure the smooth progress of the charge and discharge process, pay attention to the limit value of the battery level of the energy storage device, and when it is found that the battery level of the energy storage device is close to the upper limit or the lower limit, adjust the charge and discharge state of the energy storage device to respond to the real-time changes of the grid and avoid overcharging or over-discharging.

[0041] A further improvement of the technical solution of the present invention lies in that: in the fault warning module, the process of identifying the abnormal mode is as follows:

[0042] Continuously collect the operation data of all energy devices and energy storage devices in the microgrid through sensors and the data acquisition system, which are historical fault data and normal operation data respectively, and preprocess the collected operation data, perform data cleaning, remove noise, outliers, etc., to ensure the accuracy and reliability of the data, and extract abnormal features associated with the abnormal mode from the preprocessed data, which are abnormal temperature features, abnormal power features, and abnormal load features respectively, to obtain an abnormal feature set, and divide the abnormal feature set into a training set and a test set;

[0043] Establish a fault warning model by combining the training set data and the decision tree model, and evaluate the trained fault warning model using the test set data, so that the fault warning model can learn the behavior pattern of the device and identify the abnormal patterns that deviate from the normal mode;

[0044] Based on the relevant feature data of the abnormal feature set and the fault warning model, comprehensively analyze to obtain the abnormal warning coefficient, conduct a risk analysis on the abnormal patterns identified by the fault warning model, and judge the severity of the abnormal patterns;

[0045] According to the historical fault data and the abnormal warning coefficient, divide different warning levels for the identified abnormal patterns, namely the first-level warning level, the second-level warning level, and the third-level warning level. Among them, the warning levels increase in severity from the first level to the third level, and match corresponding warning thresholds for each warning level;

[0046] Integrate the fault warning model, input the real-time collected data into the fault warning model for real-time analysis, compare the difference between the current data and the normal behavior pattern, and judge whether there are abnormal patterns. When the calculated abnormal warning coefficient exceeds the warning threshold, respond to the corresponding warning level;

[0047] Based on the determined warning level, send out an alarm signal, generate a risk report and distribute it to relevant personnel. The risk report includes the risk level, the time, location, type, and handling process of the fault, and provides corresponding fault handling suggestions.

[0048] A further improvement of the technical solution of the present invention is that: the calculation expression of the abnormal warning coefficient is:

[0049]

[0050] Where AC is the abnormal warning coefficient, T p is the real-time temperature observation value, T nor is the temperature reference value during normal operation, P cu is the current power, P th is the power safety threshold, L cu is the current load, L max is the historical maximum load, N is the number of observation values, and the larger the value of AC, the more severe the abnormal pattern.

[0051] A further improvement of the technical solution of the present invention is that: multiple warning levels correspond to multiple warning thresholds, where the warning thresholds include upper limit thresholds and lower limit thresholds;

[0052] Multiple warning levels and multiple warning thresholds satisfy the following relationship:

[0053] First-level warning level 0 < AC ≤ AC L; Indicates low risk or initial anomaly, which may not immediately cause a failure but requires attention;

[0054] Secondary warning level AC L <AC ≤ AC M ; Indicates medium risk, with an obvious abnormal pattern and a high possibility of developing into a failure;

[0055] Tertiary warning level AC > AC M ; Indicates high risk, with a severe abnormal pattern and a very high probability of an impending failure;

[0056] Among them, AC is the abnormal warning coefficient, AC L is the lower threshold corresponding to the secondary warning level and the upper threshold corresponding to the primary warning level, AC M is the lower threshold corresponding to the tertiary warning level and the upper threshold corresponding to the secondary warning level.

[0057] Due to the adoption of the above technical solution, the technical progress achieved by the present invention compared with the prior art is:

[0058] 1. The present invention provides an integrated source-network-load-storage microgrid intelligent scheduling system. By integrating sensor networks, data acquisition and processing technologies, it can monitor the operating states of various energy devices and energy storage devices in the microgrid in real time, significantly improving the reliability and stability of the system. It can detect and warn of potential equipment failures in a timely manner, avoid the expansion of failures, reduce the power outage time and economic losses caused by failures, and can flexibly adjust the charge and discharge strategies of energy storage devices according to real-time load demands and renewable energy generation conditions, ensuring stable power supply in the microgrid under various working conditions and meeting the electricity demands of users.

[0059] 2. The present invention provides an integrated source-network-load-storage microgrid intelligent scheduling system. By integrating the data of the energy data acquisition module and the energy prediction module, it realizes accurate prediction of the power generation of renewable energies such as wind power and photovoltaic power. And through real-time analysis and processing capabilities, it can identify abnormal patterns and potential failures in the microgrid in a short time. Once an anomaly is detected, it immediately issues an alarm signal and automatically generates a risk report, providing detailed fault information and treatment suggestions for maintenance personnel, significantly shortening the fault handling time and reducing the impact of faults on the operation of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0061] Figure 1 This is the module composition diagram of the present invention;

[0062] Figure 2 This is the flowchart for predicting power generation and analyzing the load change trend of the present invention;

[0063] Figure 3 This is the flowchart for identifying the abnormal mode of the present invention. Detailed implementation manners

[0064] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0065] Embodiment 1, as shown in Figure 1 and Figure 2 the present invention provides an integrated source-grid-load-storage microgrid intelligent dispatching system, including an integrated dispatching platform, which is communicatively connected to an energy data acquisition module, an energy prediction module, an intelligent dispatching module, a energy storage management module and a fault warning module. Among them, the modules are electrically connected to each other;

[0066] The energy data acquisition module is used to collect the original operation data and status information of wind power, photovoltaic energy devices and energy storage devices in the microgrid, providing accurate and comprehensive data support for subsequent dispatching decisions, ensuring the accuracy and real-time nature of the dispatching plan. Deploy a data acquisition system and a load monitoring system. By connecting to the sensors and controllers of wind power devices, photovoltaic devices, and energy storage batteries, various energy devices and energy storage devices in the microgrid are connected to the data acquisition system. Among them, wind power devices and photovoltaic devices are energy devices, and energy storage batteries are energy storage devices. And use the load monitoring system to monitor the load status in the microgrid. Wind power devices are equipped with wind speed sensors, temperature sensors, and generator status sensors to collect wind speed, temperature, and rotation speed data. Photovoltaic devices are equipped with light intensity sensors, temperature sensors, current and voltage sensors to collect light intensity, temperature, current, and voltage data. Energy storage batteries are equipped with battery status monitoring sensors to collect the charge and discharge status, remaining capacity, and temperature data of the battery. The original operation data and status information of energy devices and energy storage devices are obtained in real time through sensors. Among them, the operation data of energy devices include the operation parameters of power generation, power, voltage, and current of wind power and photovoltaic devices, and the status information of energy storage devices includes the charge and discharge status and remaining capacity status information of energy storage batteries. The collected original data is transmitted to the integrated dispatching platform through a wireless communication network. During the transmission process, it is necessary to ensure the integrity and security of the data to avoid data loss or tampering. After receiving the original data, the integrated dispatching platform classifies and stores it according to time sequence and device type, and preprocesses the collected original data. Among them, the preprocessing includes data cleaning, screening, and formatting steps, removing error data and noise, and converting the data into a format suitable for analysis;

[0067] Energy prediction module, which uses the variational mode decomposition algorithm to predict the new energy power generation of wind power and photovoltaic power, analyzes the changing trend of the load, improves the accuracy of power generation prediction, reduces the scheduling deviation caused by prediction deviation, collects historical power generation data and load data from wind power, photovoltaic equipment and load monitoring systems, and performs data cleaning and normalization processing on the collected data. For data cleaning, outliers, missing values, etc. are removed to ensure the integrity and accuracy of the data. The cleaned data is normalized so that data with different dimensions can be compared and analyzed on the same scale. The time series data of the historical power generation data of wind power and photovoltaic power are respectively input into the variational mode decomposition algorithm. Through an iterative optimization process, each time series is decomposed into multiple intrinsic mode functions. Each intrinsic mode function represents a specific frequency component in the power generation time series, and the central frequency and bandwidth characteristics of each intrinsic mode function are analyzed to extract the characteristic information in the power generation time series. Based on the extracted characteristic information and the decomposed intrinsic mode functions, the historical power generation data and the corresponding intrinsic mode functions are used as inputs, and an energy prediction model is constructed in combination with a convolutional neural network. The trained energy prediction model is used to predict the power generation of wind power and photovoltaic power. The variational mode decomposition algorithm is used to decompose the time series data of the load data to obtain multiple intrinsic mode functions, analyze the changing trend of each intrinsic mode function, identify the periodic and non-periodic components in the load, and analyze the changing trend of the load. Combining the power generation prediction results and the load changing trend analysis, the future power generation and load are predicted, which helps the power system to better balance supply and demand and make scheduling decisions;

[0068] Further, the calculation expression for predicting the power generation of wind power and photovoltaic power is:

[0069]

[0070]

[0071] Among them, P t is the predicted power generation, IMF i (t) is the i-th intrinsic mode function at time t during power generation prediction, n is the number of intrinsic mode functions during power generation prediction, α is an adjustment factor used to adjust the influence of the sum, is the reference value, representing the historical average power generation, γ is a radical parameter used to adjust the non-linear effect of the sum, generally greater than 0, representing the depth of the radical, used to simulate the non-linear trend of power generation growth, λ is the reference value of the exponential function, S is a time variable, representing the prediction time point, which increases as time goes by, indicating the changing trend of the predicted value over time, K i is the number of terms of the i-th IMF during power generation prediction, a k is the amplitude of the k-th term during power generation prediction, fk，i is the central frequency of the k-th term during power generation prediction, θ k，i is the phase of the k-th term during power generation prediction;

[0072] The calculation expression for analyzing the load change trend is:

[0073]

[0074]

[0075] where, L t is the predicted load at time t, is the reference value, representing the average of historical load data, w j is the weight of the j-th intrinsic mode function during load prediction, indicating the contribution degree of this mode to load prediction, IMF j (t) is the value of the j-th intrinsic mode function at time t during load prediction, representing a specific frequency component in the load time series, m is the number of intrinsic mode functions during load prediction, ∈ t is the random noise term, representing the prediction error at time t or the load change that the model fails to capture, which will remain relatively stable over time, θ is the model parameter, indicating the non-linear trend of load change, T b is the time variable, K j is the number of terms of the j-th IMF during load prediction, a u is the amplitude of the u-th term during load prediction, f u，j is the central frequency of the u-th term during load prediction, φ u，j is the phase of the u-th term during load prediction;

[0076] The intelligent scheduling module, based on the information from the energy data acquisition module and the energy prediction module, uses the sparrow search algorithm to analyze and formulate a scheduling plan, realizing the configuration and scheduling of various types of energy and energy storage devices in the microgrid, improving energy utilization efficiency and reducing operating costs. It obtains real-time operation data and historical data from the energy data acquisition module, including the power generation of wind power and photovoltaic, the status of energy storage devices (electricity quantity, power limit), and load demand. Using the energy prediction module, based on historical data and current weather environment parameters, it predicts the power generation of wind power and photovoltaic and the load demand, and integrates the prediction results into the input of the intelligent scheduling module as the basis for scheduling decisions. The sparrow search algorithm is used to define the scheduling problem of the microgrid as an optimization problem, with the goal of minimizing the operating cost while meeting the load demand and system constraints (power balance, equipment capacity limit). A certain number of "sparrows" (i.e., solution sets) are randomly generated, and each sparrow represents a possible configuration and scheduling plan for energy and energy storage devices. According to the current market price, equipment efficiency, and prediction data, the fitness (i.e., operating cost) of each sparrow (solution) is calculated. Some sparrows act as discoverers and conduct random searches near their current positions to explore new solution spaces, while the remaining sparrows act as followers and move according to the quality of the discoverers, tending to better solutions. A vigilant mechanism is introduced to prevent the algorithm from falling into local optima. The diversity of solutions is increased by randomly moving or resetting some sparrows, and it is judged whether the algorithm converges according to the preset number of iterations. If not, it returns to the search and iteration steps and performs iterative calculations to make the algorithm converge to find the optimal scheduling strategy. After the algorithm converges, the result with the highest fitness is selected from all sparrows as the optimal solution, which is the optimal configuration and scheduling plan for energy and energy storage devices. The optimal solution found by the algorithm is converted into specific operation instructions, including adjusting the output power of power generation equipment and controlling the charging and discharging of energy storage devices. The scheduling instructions are sent to the corresponding energy and energy storage devices through the communication network, and the execution situation is monitored;

[0077] Energy storage management module, which is used to classify and manage the charging and discharging processes of energy storage devices, control the processes of releasing and storing electrical energy in the microgrid, improve the response speed and efficiency of the energy storage system, enhance the flexibility and stability of the microgrid, collect data of energy storage devices through sensors and data acquisition systems, monitor the status of energy storage devices, including power, charging and discharging status, and health status, and integrate data from the energy data acquisition module and the energy prediction module, collect data related to energy storage, including the power generation of wind power and photovoltaic, the status of energy storage devices, and load demand information, formulate specific charging and discharging strategies according to factors such as grid demand, power generation prediction, current market price, and device efficiency, analyze the capacity limit, charging and discharging efficiency of energy storage devices, and the overall operation requirements of the microgrid, convert the formulated charging and discharging strategies into specific operation instructions, send them to the energy storage devices for execution, monitor the operation status of energy storage devices and the power supply and demand situation of the microgrid in real time, ensure the smooth progress of the charging and discharging process, pay attention to the limit value of the power of energy storage devices, and adjust the charging and discharging status of energy storage devices when it is found that the power of energy storage devices is close to the upper limit or the lower limit, respond to the real-time changes of the grid, and avoid overcharging or over-discharging phenomena;

[0078] Fault warning module, which is used to monitor the operation status of various energy devices and energy storage devices in the microgrid in real time, identify abnormal patterns during operation, and send alarm signals when a fault occurs, improve the fault response speed and fault handling ability of the system, and reduce losses caused by faults.

[0079] Embodiment 2, as Figure 3 shown, on the basis of Embodiment 1, the present invention provides a technical solution: Preferably, in the fault warning module, the identification process of the abnormal pattern is:

[0080] Continuously collect the operation data of all energy devices and energy storage devices in the microgrid through sensors and data acquisition systems, which are historical fault data and normal operation data respectively. Preprocess the collected operation data, perform data cleaning to remove noise, outliers, etc., to ensure the accuracy and reliability of the data. Extract abnormal features associated with abnormal patterns from the preprocessed data, which are abnormal temperature features, abnormal power features, and abnormal load features respectively, to obtain an abnormal feature set. Divide the abnormal feature set into a training set and a test set. Combine the training set data and a decision tree model to establish a fault warning model. Use the test set data to evaluate the trained fault warning model, enabling the fault warning model to learn the behavior patterns of the devices and identify abnormal patterns that deviate from the normal patterns. Based on the relevant feature data of the abnormal feature set and the fault warning model, comprehensively analyze to obtain an abnormal warning coefficient, conduct a risk analysis on the abnormal patterns identified by the fault warning model, judge the severity of the abnormal patterns, and according to the historical fault data and the abnormal warning coefficient, divide different warning levels for the identified abnormal patterns, which are the first-level warning level, the second-level warning level, and the third-level warning level. Among them, the warning levels increase in severity from the first level to the third level, and match corresponding warning thresholds for each warning level. Integrate the fault warning model, input the real-time collected data into the fault warning model for real-time analysis, compare the current data with the normal behavior patterns, and judge whether there are abnormal patterns. When the calculated abnormal warning coefficient exceeds the warning threshold, respond to the corresponding warning level, based on the determined warning level, send an alarm signal, generate a risk report and distribute it to relevant personnel. The risk report includes the risk level, the time, location, type, and handling process of the fault occurrence, and provides corresponding fault handling suggestions;

[0081] Furthermore, the calculation expression of the abnormal warning coefficient is:

[0082]

[0083] where AC is the abnormal warning coefficient, T p is the real-time temperature observation value, T nor is the temperature reference value during normal operation, P cu is the current power, P th is the power safety threshold, L cu is the current load, L max is the historical maximum load, N is the number of observation values, T p -T nor represents the temperature deviation, that is, the difference between the real-time observation value and the normal value, is the exponential function, which is used to evaluate the abnormality degree of the power relative to the safety threshold. When the power exceeds the threshold, this part will approach 0, is the load ratio, representing the ratio of the current load to the historical maximum load, used to evaluate the abnormality degree of the load. The larger the value of AC, the more serious the abnormal mode. T p -T nor indicates that the observed value is lower than the normal value, and a positive value indicates that the observed value is higher than the normal value. The larger the abnormal warning coefficient, the more likely the system is in a high-risk state;

[0084] Furthermore, multiple warning levels correspond to multiple warning thresholds. Among them, the warning thresholds include an upper threshold and a lower threshold;

[0085] The multiple warning levels and the multiple warning thresholds satisfy the following relationship:

[0086] The first-level warning level 0 < AC ≤ AC L ; indicating low risk or initial abnormality, which may not immediately cause a failure but requires attention;

[0087] The second-level warning level AC L < AC ≤ AC M ; indicating medium risk, the abnormal mode is relatively obvious, and there is a high possibility of developing into a failure;

[0088] The third-level warning level AC > AC M ; indicating high risk, the abnormal mode is serious, and it is very likely that a failure will occur soon;

[0089] Among them, AC is the abnormal warning coefficient, and AC L is the lower threshold corresponding to the second-level warning level and the upper threshold corresponding to the first-level warning level, and AC M is the lower threshold corresponding to the third-level warning level and the upper threshold corresponding to the second-level warning level.

[0090] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. The source-grid-load-storage integrated microgrid intelligent dispatching system includes a comprehensive dispatching platform, which is characterized by: The integrated dispatching platform is communicatively connected with an energy data acquisition module, an energy prediction module, an intelligent dispatching module, an energy storage management module and a fault warning module, wherein electrical signals are connected between the modules; The energy data acquisition module is used to collect the original data of the operation data and status information of the wind power, photovoltaic energy equipment and energy storage equipment in the microgrid; The energy prediction module uses the variational mode decomposition algorithm to predict the power generation of new energy sources such as wind power and photovoltaic power, and analyze the change trend of the load. The process of predicting power generation and analyzing the change trend of the load is as follows: Collect historical power generation data and load data from wind power, photovoltaic equipment and load monitoring systems, and perform data cleaning and normalization on the collected data; The time series data of historical power generation data of wind power and photovoltaic power generation are respectively input into the variational mode decomposition algorithm. Each time series is decomposed into multiple intrinsic mode functions through an iterative optimization process. The center frequency and bandwidth characteristics of each intrinsic mode function are analyzed to extract the characteristic information in the power generation time series. Based on the extracted feature information and the decomposed intrinsic mode function, the historical power generation data and the corresponding intrinsic mode function are used as input, and the energy prediction model is built in combination with the convolutional neural network. The trained energy prediction model is used to predict the power generation of wind power and photovoltaic power. Use the variational mode decomposition algorithm to decompose the time series data of the load data to obtain multiple intrinsic mode functions, analyze the change trend of each intrinsic mode function, identify the periodic components and non-periodic components in the load, and analyze the change trend of the load; Combine power generation forecast results with load change trend analysis to forecast future power generation and load; The intelligent scheduling module uses the sparrow search algorithm to analyze and formulate a scheduling plan based on the information of the energy data acquisition module and the energy prediction module, so as to realize the configuration and scheduling of various energy and energy storage devices in the microgrid; The energy storage management module is used to classify and manage the charging and discharging process of the energy storage device, and control the process of releasing and storing electric energy in the microgrid; The fault warning module is used to monitor the operating status of various energy devices and energy storage devices in the microgrid in real time, identify abnormal modes during operation, and send out an alarm signal when a fault occurs.

2. The source-grid-load-storage integrated microgrid intelligent dispatching system according to claim 1 is characterized by: In the energy data acquisition module, the process of acquiring the original data of the operating data and status information is as follows: Deploy data acquisition systems and load monitoring systems, and connect various energy devices and energy storage devices in the microgrid to the data acquisition system by connecting them to sensors and controllers of wind power equipment, photovoltaic equipment, and energy storage batteries. Among them, wind power equipment and photovoltaic equipment are energy devices, and energy storage batteries are energy storage devices. The load monitoring system is used to monitor the load status in the microgrid; The original data of the operation data and status information of energy equipment and energy storage equipment are obtained in real time through sensors, wherein the operation data of energy equipment includes the operation parameters of power generation, power, voltage and current of wind power and photovoltaic equipment, and the status information of energy storage equipment includes the charge and discharge status of energy storage batteries and the status information of remaining capacity; The collected raw data is transmitted to the integrated dispatching platform through the wireless communication network. After receiving the raw data, the integrated dispatching platform classifies and stores them according to time sequence and equipment type; The collected raw data is preprocessed, where the preprocessing includes data cleaning, screening and formatting steps.

3. The source-grid-load-storage integrated microgrid intelligent dispatching system according to claim 1 is characterized in that: The calculation expression for predicting the power generation of wind power and photovoltaic power is: Among them, P t For the predicted electricity generation, IMF i (t) is the i-th intrinsic mode function at time t in power generation prediction, n is the number of intrinsic mode functions in power generation prediction, α is the adjustment factor, is the benchmark value, representing the historical average power generation, γ is the root parameter, λ is the benchmark value of the exponential function, S is the time variable, representing the forecast time point, K i is the number of items in the ith IMF when predicting power generation, a k is the amplitude of the kth item in power generation forecast, f k，i is the center frequency of the kth item in power generation forecast, θ k，i is the phase of the kth item in power generation forecast; The calculation expression for analyzing the load change trend is: Among them, L t is the predicted load at time t, is the benchmark value, representing the average value of historical load data, w j is the weight of the jth intrinsic mode function in load forecasting, IMF j (t) is the value of the jth intrinsic mode function at time t during load forecasting, m is the number of intrinsic mode functions during load forecasting, ∈ t is a random noise term, θ is a model parameter, representing the nonlinear trend of load change, T b is the time variable, K j is the number of items in the jth IMF during load forecasting, a u is the amplitude of the uth item in load forecasting, f u，j is the center frequency of the uth item in load forecasting, φ u，j is the phase of the u-th item in load forecasting.

4. The source-grid-load-storage integrated microgrid intelligent dispatching system according to claim 3 is characterized by: In the intelligent dispatching module, the process of configuring and dispatching various energy sources and energy storage devices in the microgrid is as follows: Obtain real-time operation data and historical data from the energy data acquisition module, including wind power, photovoltaic power generation, energy storage equipment status, and load demand; The energy forecasting module is used to forecast the power generation and load demand of wind power and photovoltaic power based on historical data and current weather and environmental parameters, and the forecast results are integrated into the input of the intelligent scheduling module as the basis for scheduling decisions; The microgrid dispatch problem is defined as an optimization problem using the sparrow search algorithm, and iterative calculations are performed to converge the algorithm in order to find the optimal dispatch strategy. After the algorithm converges, the result with the highest fitness is selected as the optimal solution, that is, the optimal energy and energy storage equipment configuration and scheduling plan. The optimal solution found by the algorithm is converted into specific operation instructions, including adjusting the output power of the power generation equipment, controlling the charging and discharging of the energy storage equipment, sending the scheduling instructions to the corresponding energy and energy storage equipment through the communication network, and monitoring the execution status.

5. The source-grid-load-storage integrated microgrid intelligent dispatching system according to claim 4 is characterized in that: In the energy storage management module, the process of controlling the release and storage of electric energy in the microgrid is as follows: Collect data from energy storage devices through sensors and data acquisition systems, monitor the status of energy storage devices, including power, charging and discharging status, and health status, and integrate data from energy data acquisition modules and energy prediction modules to collect data related to energy storage, including wind power and photovoltaic power generation, the status of energy storage devices, and load demand information; Formulate specific charging and discharging strategies based on grid demand, power generation forecasts, current market prices, and equipment efficiency factors, analyze the capacity limitations of energy storage equipment, charging and discharging efficiency, and the overall operation requirements of the microgrid, and convert the formulated charging and discharging strategies into specific operating instructions, which are sent to the energy storage equipment for execution; Monitor the operating status of energy storage equipment and the power supply and demand of the microgrid in real time, pay attention to the power limit of energy storage equipment, adjust the charging and discharging status of energy storage equipment, and respond to real-time changes in the power grid.

6. The source-grid-load-storage integrated microgrid intelligent dispatching system according to claim 5 is characterized by: In the fault warning module, the abnormal mode recognition process is as follows: The operating data of all energy devices and energy storage devices in the microgrid are continuously collected through sensors and data acquisition systems, which are historical fault data and normal operating data, and the collected operating data are preprocessed. Abnormal features associated with abnormal modes are extracted from the preprocessed data, which are abnormal temperature features, abnormal power features and abnormal load features, to obtain an abnormal feature set, which is divided into a training set and a test set; A fault warning model is established by combining the training set data and the decision tree model. The trained fault warning model is evaluated using the test set data so that the fault warning model can learn the behavior pattern of the equipment and identify abnormal patterns that deviate from the normal pattern. Based on the relevant feature data of the abnormal feature set and the fault warning model, the abnormal warning coefficient is obtained through comprehensive analysis, and the risk analysis is performed on the abnormal pattern identified by the fault warning model to determine the severity of the abnormal pattern; According to the historical fault data and abnormal warning coefficient, different warning levels are divided for the identified abnormal patterns, namely, the first warning level, the second warning level and the third warning level. The severity of the warning level increases from level one to level three, and the corresponding warning threshold is matched for each warning level; Integrated fault warning model, input the real-time collected data into the fault warning model, conduct real-time analysis, compare the difference between the current data and the normal behavior pattern, determine whether there is an abnormal pattern, and respond to the corresponding warning level when the calculated abnormal warning coefficient exceeds the warning threshold; Based on the determined warning level, an alarm signal is issued, a risk report is generated and distributed to relevant personnel, and corresponding fault handling suggestions are provided.

7. The source-grid-load-storage integrated microgrid intelligent dispatching system according to claim 6 is characterized by: The calculation expression of the abnormal warning coefficient is: Among them, AC is the abnormal warning coefficient, T p is the real-time temperature observation value, T nor is the temperature reference value during normal operation, P cu is the current power, P th is the power safety threshold, L cu is the current load, L max is the historical maximum load, N is the number of observations, and the larger the AC value is, the more serious the abnormal pattern is.

8. The source-grid-load-storage integrated microgrid intelligent dispatching system according to claim 7 is characterized by: A plurality of the warning levels correspond to a plurality of the warning thresholds, wherein the warning thresholds include an upper threshold and a lower threshold; The multiple warning levels and the multiple warning thresholds satisfy the following relationship: Level 1 Warning 0 <AC≤AC L ; Level 2 warning level AC L <AC≤AC M ; Level 3 warning level AC>AC M ; Among them, AC is the abnormal warning coefficient, AC L is the lower threshold corresponding to the second warning level and the upper threshold corresponding to the first warning level, AC M It is the lower threshold corresponding to the third warning level and the upper threshold corresponding to the second warning level.

Images