Intelligent micro-grid control method based on multi-source collaborative optimization

Abstract

The invention provides an intelligent micro-grid control method based on multi-source collaborative optimization, and belongs to the technical field of micro-grid control, and the method comprises the steps: collecting the operation state data of a diesel generator set in real time, including the output power, the fuel consumption rate and the operation temperature of the diesel generator set, and the SOC and the charging and discharging power of an energy storage system; output voltage, current and illumination intensity of the photovoltaic array, and real-time power data of a load end; obtaining the total load demand and the schedulable energy storage capacity of the current micro-grid system; calculating an optimal generator set operation interval according to the load demand and the energy storage scheduling capability; and the output power of the generator set is dynamically adjusted, so that the generator set operates in the optimal fuel efficiency interval. Through real-time sensing and dynamic adjustment, the generator set always operates in the optimal fuel efficiency interval, fuel consumption of unit electric energy is remarkably reduced, the overall energy utilization efficiency is improved, and carbon emission is reduced.

Description

Technical Field

[0001] This invention belongs to the field of microgrid control technology, specifically relating to a smart microgrid control method based on multi-source collaborative optimization. Background Technology

[0002] A microgrid is a small-scale power generation and distribution system composed of distributed power sources (such as photovoltaic and wind power), energy storage devices, loads, monitoring and protection systems, and control systems. It has the ability to operate independently or switch to grid connection and is a key carrier for realizing efficient energy utilization and intelligent management.

[0003] On the other hand, as mentioned in the prior art solution in patent publication number "CN120280938A", relevant technologies already exist for secondary voltage control in microgrids. However, in existing technologies, microgrid systems typically consist of modules such as diesel generator sets, energy storage systems, renewable energy modules, and loads. However, in areas with no or weak electricity, as well as construction sites, the lack of a unified dispatch mechanism leads to information exchange barriers between modules, resulting in low energy dispatch efficiency, frequent generator start-stop cycles, and delayed load response. Furthermore, traditional microgrid systems lack real-time optimization control of generator operating status, resulting in poor fuel economy, high carbon emissions, and reliance on manual inspections for operation and maintenance, making it difficult to achieve efficient, safe, and low-consumption operation goals. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention presents a smart microgrid control method based on multi-source collaborative optimization. This method aims to solve the problem of poor fuel economy of generator sets in existing microgrid systems. By dynamically optimizing and controlling the operating status of diesel generators through a smart microgrid control system, fuel efficiency can be maximized, and operating costs and carbon emissions can be reduced.

[0005] The present invention employs the following technical solution.

[0006] A smart microgrid control method based on multi-source collaborative optimization includes:

[0007] Step 1: Perform multi-source heterogeneous data fusion and state awareness;

[0008] Step 2: Perform intelligent power generation scheduling and fuel efficiency optimization;

[0009] Step 3: Implement a multi-timescale collaborative control strategy;

[0010] Step 4: Implement intelligent operation and maintenance driven by digital twins;

[0011] Step 5: Perform secure collaboration and adaptive communication.

[0012] Furthermore, step 1 specifically includes:

[0013] Data acquisition is performed on the field devices of the microgrid system. This step involves collecting data from the power sensors (Power Sensor 1, Power Sensor 2, Temperature Sensor, SOC Sensor, Power Sensor 3, Voltage Sensor, Current Sensor, Illumination Sensor, and Power Sensor 4) of the edge computing terminal, which is connected to the control unit of the edge computing terminal. The data includes the output power, fuel consumption rate, and operating temperature of the diesel generator; the SOC and charging / discharging power of the energy storage system; the output voltage, current, and illuminance of the photovoltaic array; and the real-time power data of the load. This data is then transmitted to the control unit of the edge computing terminal.

[0014] Furthermore, step 1 specifically includes:

[0015] All collected data is also standardized and encapsulated on the control unit using Modbus-TCP and OPC UA protocols, and the collected data is cleaned and anomaly detected locally on the control unit.

[0016] Furthermore, step 2 specifically includes:

[0017] The control unit constructs a generator set operation optimization model based on a reinforcement learning algorithm. The inputs to this algorithm include current load data, SOC data, operating temperature, and geographical and meteorological parameters of the microgrid system. In other words, it optimizes the power generation plan based on the current state of the microgrid system. Its objective function is:

[0018] ;

[0019] in Indicates at time The output power of the generator set is fuel costs at that time This represents the optimal power generation capacity derived from the SOC of the energy storage system and load forecasting. Represents the smoothing coefficient. This indicates the end time of the set duration.

[0020] Furthermore, the generator set is a diesel generator.

[0021] Furthermore, step 3 specifically includes:

[0022] A hierarchical control architecture is adopted to achieve coordinated execution of control strategies at the second, minute, and hour levels. The bottom layer uses model predictive control to achieve second-level control and regulation of generator output power. The middle layer uses fuzzy logic control units to handle minute-level control and regulation of load changes and energy storage SOC fluctuations. The top layer uses hour-level control and regulation to formulate power generation plans based on load data and geographical and meteorological parameters.

[0023] Furthermore, in step 3, the method for formulating a power generation plan based on load data and geographical and meteorological parameters specifically includes:

[0024] The scheduling strategy is formulated based on a rolling time-domain optimization framework, and the 24-hour power generation plan is updated hourly. The load data and geographical and meteorological parameters are used as inputs to the model predictive control method to generate control commands, which are then sent to the PLC controller connected to the same control unit of the generator unit via the OPC UA protocol. The scheduling execution module of the microgrid system has an anti-maloperation mechanism: when the deviation between the predicted load and the measured value exceeds 10%, the local re-optimization process is automatically triggered. The energy storage system adopts a two-layer SOC control strategy to maintain the SOC in a safe range of 40%-80%.

[0025] Furthermore, step 4 specifically includes:

[0026] A virtual mapping of the microgrid system was constructed to synchronize the operating status of each device in the microgrid system in real time. A graph neural network was used to assess the health status of the microgrid system's devices. Furthermore, the control unit was trained using historical data to obtain the following generator set lifespan degradation model:

[0027] ;

[0028] in Represents the natural constant. Indicates time The output power of the generator set, For the initial lifespan of the generator set, Indicates from time The lifespan of the generator set, The attenuation coefficient is... This indicates the rated power of the generator set.

[0029] Furthermore, step 5 specifically includes:

[0030] The microgrid system adopts a hybrid communication architecture. The diesel generator and energy storage system communicate in real time via industrial Ethernet, and the remote operation and maintenance platform interacts with local edge terminals via the 5G public network. At the same time, the microgrid system introduces a lightweight blockchain method to verify the permissions and record the operation of key control commands and operation and maintenance operations, so as to ensure that the data is tamper-proof and the operation is traceable.

[0031] Furthermore, the photovoltaic array and the load terminal are connected via the LoRaWAN protocol.

[0032] The beneficial effects of the present invention are as follows, compared with the prior art:

[0033] This invention collects real-time operating status data of diesel generator sets, including load rate, speed, fuel consumption rate, output voltage, and frequency; obtains the total load demand and dispatchable energy storage capacity of the current microgrid system; calculates the optimal operating range of the generator set based on the load demand and energy storage dispatch capability; and dynamically adjusts the generator set's output power to ensure it operates within the optimal fuel efficiency range. Through real-time sensing and dynamic adjustment, this invention ensures the generator set always operates within the optimal fuel efficiency range, significantly reducing fuel consumption per unit of electricity, improving overall energy utilization efficiency, and reducing carbon emissions. Attached Figure Description

[0034] Figure 1 This is a flowchart of the intelligent microgrid control method based on multi-source collaborative optimization in this invention. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, other embodiments obtained by those skilled in the art without creative effort are all within the protection scope of this invention.

[0036] like Figure 1 As shown, a smart microgrid control method based on multi-source collaborative optimization includes:

[0037] Step 1: Perform multi-source heterogeneous data fusion and state awareness;

[0038] In step 1, a multi-protocol data acquisition terminal is deployed based on an edge computing architecture to achieve millisecond-level acquisition of operating parameters of the diesel generators, energy storage systems, photovoltaic modules, and load devices in the microgrid system. A hybrid communication protocol stack of OPC UA and Modbus-TCP is adopted to support efficient access of heterogeneous devices and standardized data processing. An embedded AI chip is used for local data preprocessing and anomaly detection, improving data quality and system response speed. This step provides real-time and accurate system status input for subsequent optimized control.

[0039] In a preferred but non-limiting embodiment of the present invention, step 1 specifically includes:

[0040] Data acquisition is performed on the field devices of the microgrid system. This step involves collecting data from the power sensor 1, power sensor 2, temperature sensor, SOC sensor, power sensor 3, voltage sensor, current sensor, illuminance sensor, and power sensor 4 of the edge computing terminal, which is connected to the control unit of the edge computing terminal. These data include the output power, fuel consumption rate, and operating temperature of the diesel generator in the microgrid system; the SOC (State of Charge) and charging / discharging power of the energy storage system; the output voltage, current, and illuminance of the photovoltaic array; and the real-time power data of the load end. This data is then transmitted to the control unit of the edge computing terminal.

[0041] In a preferred but non-limiting embodiment of the present invention, step 1 further includes:

[0042] All collected data is standardized and encapsulated on the control unit using Modbus-TCP and OPC UA protocols. Local data cleaning and anomaly detection are performed on the control unit to ensure the accuracy and timeliness of the collected data. For example, when a voltage drop exceeding 10% is detected in the photovoltaic module (i.e., the photovoltaic array), the control unit triggers a local alarm (e.g., a buzzer-activated voice alarm connected to the control unit) and simultaneously uploads the data to the maintenance platform (such as an industrial control computer) connected to the control unit. Data cleaning and anomaly detection can be performed using existing technologies. The control unit can be a PLC or a microcontroller.

[0043] Step 2: Perform intelligent power generation scheduling and fuel efficiency optimization;

[0044] In step 2, a generator set operation optimization model is constructed based on a reinforcement learning algorithm. Inputs include current load data, SOC data, operating temperature, and geographical and meteorological parameters such as the altitude of the microgrid system. This step dynamically updates the generator set fuel efficiency surface through an online learning mechanism, predicting the optimal output point based on the current system state. A rolling time-domain optimization strategy is employed to achieve generator set start-up, shutdown, and power allocation decisions within a minute-level scheduling cycle, ensuring operation within the high-efficiency fuel range. This step can reduce fuel consumption by 8%–15%, significantly improving energy utilization efficiency.

[0045] In a preferred but non-limiting embodiment of the present invention, step 2 specifically includes:

[0046] The control unit constructs a generator set operation optimization model based on a reinforcement learning algorithm. The inputs to this algorithm include current load data, SOC data, operating temperature, and geographical and meteorological parameters such as altitude, air temperature, air pressure, and wind speed of the microgrid system. In other words, it optimizes the power generation plan based on the current microgrid system state. This step employs a reinforcement learning-based scheduling algorithm, with the objective function being:

[0047] ;

[0048] in Indicates at time The output power of the generator set is fuel costs at that time This represents the optimal power generation capacity derived from the SOC of the energy storage system and load forecasting. This represents a smoothing coefficient, used to balance fuel efficiency and dispatch stability. This indicates the end point of the set duration. For example, during a certain scheduling cycle, the control unit predicts that the current load data is 80kW, the energy storage SOC is 65%, and the ambient temperature is 25℃. The control unit uses an online learning model to predict the fuel efficiency curve of the diesel generator at different power levels, and finally decides to operate the generator set at 75kW to keep the fuel consumption rate below 200g / kWh.

[0049] In a preferred but non-limiting embodiment of the present invention, the generator set is a diesel generator.

[0050] Step 3: Implement a multi-timescale collaborative control strategy;

[0051] In step 3, a hierarchical control architecture is adopted to achieve the coordinated execution of second-level, minute-level, and hour-level control strategies. The bottom layer uses model predictive control (MPC) to achieve rapid adjustment of generator output power; the middle layer uses fuzzy logic control units to handle load surges and energy storage SOC fluctuations; and the top layer formulates generation plans based on load forecasts and geographical and meteorological parameters. Through time-scale decoupling control strategies, the dynamic response capability and long-term operational stability of the microgrid system are improved, ensuring reliable power supply under complex operating conditions in areas with no or weak electricity.

[0052] In a preferred but non-limiting embodiment of the present invention, step 3 specifically includes:

[0053] A hierarchical control architecture is adopted to achieve coordinated execution of second-level, minute-level, and hourly control strategies. The bottom layer uses Model Predictive Control (MPC) to achieve second-level control and regulation of generator output power. The middle layer uses fuzzy logic control units to handle minute-level control and regulation of load surges and energy storage SOC fluctuations. The top layer uses hourly control and regulation based on load data and geographical and meteorological parameters to formulate power generation plans. At the control level, a multi-timescale coordinated control strategy module ensures the stable operation of the microgrid system at different time scales. Second-level control uses Model Predictive Control (MPC) to adjust generator output power in real time to cope with sudden load changes. Minute-level control coordinates energy storage charging and discharging through fuzzy logic control units to maintain system frequency stability. Hourly control formulates a daily power generation plan based on load forecasts and meteorological data. For example, when a sudden load rises to 100kW, the MPC method instructs the energy storage system to release 10kW of power within 2 seconds to prevent the diesel generator from being overloaded instantly. The fuzzy logic control unit coordinates the energy storage to charge and discharge within 5 seconds to maintain the microgrid system frequency stability. At the same time, the scheduling module adjusts the power generation plan within 1 hour to ensure the long-term operating efficiency of the microgrid system.

[0054] In a preferred but non-limiting embodiment of the present invention, step 3, the method for formulating a power generation plan based on load data and geographical meteorological parameters, specifically includes:

[0055] The scheduling strategy is formulated based on a rolling time-domain optimization framework, and the 24-hour power generation plan is updated hourly. The load data and geographical meteorological parameters are used as inputs to the model predictive control (MPC) method to generate control commands, which are then sent to the PLC controller connected to the same control unit of the diesel generator set via the OPC UA protocol. The microgrid system's scheduling execution module has an anti-maloperation mechanism: when the deviation between the predicted load and the measured value exceeds 10%, a local re-optimization process is automatically triggered. The energy storage system adopts a two-layer SOC control strategy to maintain the SOC in a safe range of 40%-80%.

[0056] Step 4: Implement intelligent operation and maintenance driven by digital twins;

[0057] Step 4 involves constructing a microgrid virtual mapping system based on digital twin technology, integrating equipment operation data, historical fault databases, and maintenance strategy models of the microgrid system. A graph neural network (GNN) is used to construct an equipment state evolution map of the microgrid system, enabling lifespan prediction and fault trend early warning for key components. Intelligent operation and maintenance supports remote diagnostics, automated inspections, and intelligent work order generation, combined with AR-assisted operation and maintenance technology to improve on-site handling efficiency. This step 4 achieves visualized management of the entire lifecycle of the microgrid system, reducing the frequency of manual inspections and improving the level of intelligent operation and maintenance.

[0058] In a preferred but non-limiting embodiment of the present invention, step 4 specifically includes:

[0059] A virtual mapping of the microgrid system was constructed to synchronize the operating status of each device in the microgrid system in real time. A graph neural network (GNN) was used to assess the health status of the microgrid system's devices. Furthermore, the control unit was trained using historical data to obtain the following generator set lifespan degradation model:

[0060] ;

[0061] in Represents the natural constant. Indicates time The output power of the generator set, For the initial lifespan of the generator set, Indicates from time The lifespan of the generator set, The attenuation coefficient is... This indicates the rated power of the generator set.

[0062] Step 5: Perform secure collaboration and adaptive communication.

[0063] In step 5, a hybrid communication architecture is adopted, integrating 5G public network, LoRaWAN local area network, and wired industrial Ethernet to ensure stable communication capabilities in areas with no or weak electricity. Blockchain technology is used to dynamically manage data access permissions for each module, constructing a decentralized data collaboration mechanism. A federated learning framework is introduced to achieve multi-site collaborative optimization model training while ensuring data privacy. This step 5 effectively breaks down information sharing barriers and improves the collaborative efficiency and operational security of the microgrid system.

[0064] In a preferred but non-limiting embodiment of the present invention, step 5 specifically includes:

[0065] Ensuring data security and efficient collaboration among modules of the microgrid system is crucial. The microgrid system employs a hybrid communication architecture: diesel generators and energy storage systems communicate in real-time via industrial Ethernet; photovoltaic arrays and load terminals connect via the LoRaWAN protocol; and the remote operation and maintenance platform interacts with local edge terminals via the 5G public network. Simultaneously, the microgrid system incorporates lightweight blockchain technology to verify permissions and record operations for critical control commands and maintenance procedures, ensuring data immutability and operational traceability. For example, when a remote dispatch command is issued, the microgrid system verifies operation permissions through a blockchain smart contract, ensuring the command's legitimacy before executing the dispatch action, preventing unauthorized access and malicious control.

[0066] The beneficial effects of the present invention are as follows, compared with the prior art:

[0067] This invention collects real-time operating status data of diesel generator sets, including load rate, speed, fuel consumption rate, output voltage, and frequency; obtains the total load demand and dispatchable energy storage capacity of the current microgrid system; calculates the optimal operating range of the generator set based on the load demand and energy storage dispatch capability; and dynamically adjusts the generator set's output power to ensure it operates within the optimal fuel efficiency range. Through real-time sensing and dynamic adjustment, this invention ensures the generator set always operates within the optimal fuel efficiency range, significantly reducing fuel consumption per unit of electricity, improving overall energy utilization efficiency, and reducing carbon emissions.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention without departing from the spirit and scope of the present invention. Any modifications or equivalent substitutions should be covered within the scope of protection of the claims of the present invention.

Claims

1. A smart microgrid control method based on multi-source collaborative optimization, characterized in that, include: Step 1: Perform multi-source heterogeneous data fusion and state awareness; Step 2: Perform intelligent power generation scheduling and fuel efficiency optimization; Step 3: Implement a multi-timescale collaborative control strategy; Step 4: Implement intelligent operation and maintenance driven by digital twins; Step 5: Perform secure collaboration and adaptive communication.

2. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 1, characterized in that, Step 1 specifically includes: Data acquisition is performed on the field devices of the microgrid system. This step involves collecting data from the power sensors (Power Sensor 1, Power Sensor 2, Temperature Sensor, SOC Sensor, Power Sensor 3, Voltage Sensor, Current Sensor, Illumination Sensor, and Power Sensor 4) of the edge computing terminal, which is connected to the control unit of the edge computing terminal. The data includes the output power, fuel consumption rate, and operating temperature of the diesel generator; the SOC and charging / discharging power of the energy storage system; the output voltage, current, and illuminance of the photovoltaic array; and the real-time power data of the load. This data is then transmitted to the control unit of the edge computing terminal.

3. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 2, characterized in that, Step 1 also includes: All collected data is also standardized and encapsulated on the control unit using Modbus-TCP and OPC UA protocols, and the collected data is cleaned and anomaly detected locally on the control unit.

4. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 3, characterized in that, Step 2 specifically includes: The control unit constructs a generator set operation optimization model based on a reinforcement learning algorithm. The inputs to this algorithm include current load data, SOC data, operating temperature, and geographical and meteorological parameters of the microgrid system. In other words, it optimizes the power generation plan based on the current state of the microgrid system. Its objective function is: ； in Indicates at time The output power of the generator set is fuel costs at that time This represents the optimal power generation capacity derived from the SOC of the energy storage system and load forecasting. Represents the smoothing coefficient. This indicates the end time of the set duration.

5. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 4, characterized in that, The generator set is a diesel generator.

6. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 5, characterized in that, Step 3 specifically includes: A hierarchical control architecture is adopted to achieve coordinated execution of control strategies at the second, minute, and hour levels. The bottom layer uses model predictive control to achieve second-level control and regulation of generator output power. The middle layer uses fuzzy logic control units to handle minute-level control and regulation of load changes and energy storage SOC fluctuations. The top layer uses hour-level control and regulation to formulate power generation plans based on load data and geographical and meteorological parameters.

7. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 6, characterized in that, In step 3, the method for formulating a power generation plan based on load data and geographical meteorological parameters specifically includes: The scheduling strategy is formulated based on a rolling time-domain optimization framework, and the 24-hour power generation plan is updated hourly. The load data and geographical and meteorological parameters are used as inputs to the model predictive control method to generate control commands, which are then sent to the PLC controller connected to the same control unit of the generator unit via the OPC UA protocol. The scheduling execution module of the microgrid system has an anti-maloperation mechanism: when the deviation between the predicted load and the measured value exceeds 10%, the local re-optimization process is automatically triggered. The energy storage system adopts a two-layer SOC control strategy to maintain the SOC in a safe range of 40%-80%.

8. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 7, characterized in that, Step 4 specifically includes: A virtual mapping of the microgrid system was constructed to synchronize the operating status of each device in the microgrid system in real time. A graph neural network was used to assess the health status of the microgrid system's devices. Furthermore, the control unit was trained using historical data to obtain the following generator set lifespan degradation model: ； in Represents the natural constant. Indicates time The output power of the generator set, For the initial lifespan of the generator set, Indicates from time The lifespan of the generator set, The attenuation coefficient is... This indicates the rated power of the generator set.

9. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 8, characterized in that, Step 5 specifically includes: The microgrid system adopts a hybrid communication architecture. The diesel generator and energy storage system communicate in real time via industrial Ethernet, and the remote operation and maintenance platform interacts with local edge terminals via the 5G public network. At the same time, the microgrid system introduces a lightweight blockchain method to verify the permissions and record the operation of key control commands and operation and maintenance operations, so as to ensure that the data is tamper-proof and the operation is traceable.

10. The intelligent microgrid control method based on multi-source collaborative optimization according to claim 9, characterized in that, The photovoltaic array and the load terminal are connected via the LoRaWAN protocol.

Images