A micro-grid scheduling control method and device based on policy optimization, equipment and medium

Abstract

The application provides a micro-grid scheduling control method and device based on policy optimization, equipment and medium, belonging to the power grid scheduling technical field, the method: collecting real-time operation data of new energy power generation equipment, power load equipment and energy storage system; receive user power generation and power consumption plan, combine real-time operation data and historical operation data, extract multi-resource dynamic response characteristics through LSTM model, output adjustment capacity evaluation matrix; receive power grid scheduling plan, based on the adjustment capacity evaluation matrix, match the optimal operation mode from the knowledge base, generate the scheduling instruction set; issue the scheduling instruction set to the target equipment and execute, collect the execution data in real time and calculate the operation deviation; when the operation deviation is over the threshold, the instruction weight is updated by using the online Q learning algorithm, and the optimization strategy is fed back to the knowledge base. The application realizes fine scheduling of micro-grid through real-time data collection and intelligent optimization algorithm, improves resource utilization and operation flexibility, and reduces operation cost.

Description

Technical Field

[0001] This application belongs to the field of power grid dispatching technology, specifically relating to a microgrid dispatching and control method, device, equipment, and medium based on strategy optimization. Background Technology

[0002] With the development of distributed new energy sources (such as solar photovoltaic and wind power) and adjustable loads (electric vehicles), new power systems have emerged. However, the large-scale integration of these distributed power sources and adjustable loads has significantly enhanced the random fluctuations on both the source and load sides of the power system, posing a huge challenge to the traditional power grid that uses the source-follow-load dynamic regulation mode.

[0003] Smart microgrids, as a carrier of new power grids, promote energy structure transformation, enhance renewable energy absorption capacity, and improve the flexibility and reliability of the power system through energy management. Traditional microgrid dispatching has the following drawbacks:

[0004] First, the dispatching model is too simplistic. It focuses solely on peak shaving and valley filling, failing to develop targeted dispatching strategies for different electricity user characteristics and resource scenarios, resulting in limited resource utilization. For example, in industrial park scenarios, different enterprises have significantly different production schedules and electricity consumption characteristics; the traditional single dispatching model cannot fully leverage the potential of distributed power sources and adjustable loads. Second, resource coordination is insufficient. It merely controls new energy sources, energy storage, or loads in isolation, lacking dynamic coupling of multiple resources. This makes the microgrid lack flexibility and reliability when dealing with complex electricity demand scenarios, easily leading to power shortages or wind and solar curtailment. Finally, the dispatching strategy is rigid and fixed. It typically relies on day-ahead manual commands, lacking an execution feedback loop, resulting in a high deviation rate in execution performance.

[0005] While virtual power plants can currently be used for microgrid dispatch, they rely too heavily on macro-level dispatch and neglect personalized user needs (such as comfort and production scheduling). This may reduce users' enthusiasm for participating in virtual power plant dispatch and affect the synergistic interaction between microgrids and virtual power plants. At the same time, the evaluation indicators are too simplistic, only assessing grid deviations and failing to establish a strategy self-optimization mechanism. This makes it difficult for microgrids to adapt to the dynamic changes in market rules when participating in electricity market transactions. Summary of the Invention

[0006] In a first aspect, embodiments of this application provide a microgrid scheduling and control method based on strategy optimization, comprising the following steps:

[0007] S1. Collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store them in a comprehensive database;

[0008] S2. Receive user power generation and consumption plans, combine real-time operation data and historical operation data, extract multi-resource dynamic response features through a pre-trained LSTM model, and output a regulation capability evaluation matrix;

[0009] S3. Receive the power grid dispatch plan, and based on the regulation capacity assessment matrix, match the optimal operating mode from the knowledge base to generate a set of dispatch instructions for new energy power generation equipment, power load equipment and energy storage system;

[0010] S4. Issue a set of scheduling instructions to the target device and execute them, collect execution data in real time and calculate the running deviation;

[0011] S5. When the operational deviation exceeds the threshold, the instruction weights are updated using an online Q-learning algorithm, and the optimization strategy is fed back to the knowledge base.

[0012] Furthermore, the specific steps of step S1 are as follows:

[0013] S11. Collect three types of resource data and environmental parameters through a multi-protocol interface:

[0014] Obtain the active power of new energy power generation equipment via Modbus / TCP protocol. Electricity generation and running status ;

[0015] Real-time power of electrical load equipment is collected via IoT network. Electricity consumption and controllable identifier ;

[0016] The charging and discharging power of the energy storage system is collected via CAN bus. SOC status and rated capacity ;

[0017] Ambient temperature is collected by sensors. and irradiance ;

[0018] S12. Store the collected real-time operational data in the integrated database according to the timestamp, and perform spatiotemporal alignment:

[0019] ;

[0020] in, This is a vector of real-time running data, where t represents a timestamp;

[0021] S13. Preprocess the collected real-time data, including missing value imputation and outlier handling.

[0022] Furthermore, the specific steps of step S2 are as follows:

[0023] S21. Construct the LSTM model:

[0024] The LSTM model is set to input a feature sequence with a time window length of k. ;

[0025] Wherein, the feature vector at time t ;

[0026] The LSTM network architecture consists of an input layer, a first LSTM layer, a second LSTM layer, a fully connected layer, and an output layer connected in sequence.

[0027] The output is set as a regulation capability assessment matrix represented by a normalized regulation capability index vector:

[0028]

[0029] in, This indicates power regulation capability, specifically the power response speed regulation capability. This indicates the power regulation capability, specifically the ability to regulate the duration of continuous regulation. This refers to the SOC state regulation capability, specifically the SOC change rate regulation capability index.

[0030] S22. Divide the historical running data into feature sequences and construct a training dataset;

[0031] Set a loss function and pre-train the LSTM model until the loss function converges or the maximum number of iterations is met;

[0032] The loss function is as follows:

[0033]

[0034] Where N represents the number of samples in the training dataset, This represents the actual rate of change of power. This is the actual energy regulation amount. The rate of change of SOC For power conversion time, Typical adjustment duration, For response time;

[0035] S23. Divide the real-time running data according to the time window k to obtain the real-time data window. The pre-trained LSTM model is input, and forward propagation is performed to obtain the output conditioning evaluation capability matrix:

[0036]

[0037] In this matrix, the rows of the adjustment assessment capability matrix represent new energy power generation equipment, energy storage systems, and power load equipment, respectively, and the columns represent power regulation, power regulation, and SOC regulation, respectively.

[0038] S24. Calculate the resource integration adjustment index and quantify the response capability as follows:

[0039]

[0040] in, To adjust the preset weights corresponding to the elements in the assessment capability matrix, Indicates the energy storage system adjustment index. Indicates the new energy adjustment index, This represents the load adjustment index.

[0041] Furthermore, the specific steps of step S3 are as follows:

[0042] S31. Based on grid connection point power constraints in power grid dispatching plans and daily electricity consumption constraints Combining the adjustment capability assessment matrix C, the optimal operating mode is selected through the preset decision tree in the knowledge base;

[0043] The specific decisions are as follows:

[0044] If the power grid dispatch plan requires that the power at the grid connection point not exceed the power constraint at the grid connection point. If so, select the power-limited operation mode;

[0045] If the power grid dispatch plan requires that daily electricity consumption not exceed the daily electricity consumption constraint If so, select the power-limited operation mode;

[0046] If the power grid dispatch plan requires zero-carbon operation, then the zero-carbon operation mode shall be selected;

[0047] If the power grid dispatch plan has no constraints on adjustment, then the economic operation mode shall be selected;

[0048] If the power grid is disconnected from the main grid, the microgrid autonomous operation mode is selected;

[0049] Otherwise, select normal operating mode;

[0050] S32. Based on the selected optimal operating mode, combined with the adjustment capability assessment matrix C and the resource fusion adjustment index... Generate scheduling instruction set ;

[0051] in, It is an instruction on new energy power generation equipment. It is an instruction from the energy storage system. It is an instruction for electrical load equipment;

[0052] S33. Store the generated scheduling instruction set in the knowledge base and record the basis for selecting the optimal operating mode and the instruction generation parameters.

[0053] Furthermore, the specific process for generating the scheduling instruction set in step S32 is as follows:

[0054] S321. Directive on New Energy Power Generation Equipment The generation process is as follows:

[0055] In power-limited operation mode, set ;

[0056] in, This indicates the maximum available power of the new energy power generation equipment. Indicates the new energy adjustment index, It is the maximum available power allowed by the power grid;

[0057] Setting up a zero-carbon operation mode with the goal of maximizing new energy sources ;

[0058] In power-limited operation mode, economic operation mode, microgrid autonomous operation mode, and normal operation mode, active power priority is set as the objective. ;

[0059] S322. Energy Storage System Command The generation process is as follows:

[0060] Setting the power-limited operation mode ;

[0061] in, Indicates the actual power of the load. Indicates the energy storage system adjustment index. It is the maximum available power allowed by the power grid;

[0062] Setting the zero-carbon operation mode ;

[0063] in, It is the total electricity demand of the load. It is renewable energy power generation. It is purchased green electricity. Indicates the energy storage system adjustment index. It is the security verification time window for instruction execution;

[0064] In the economic operation mode, charging is set up during off-peak hours according to time-of-use pricing. Peak discharge settings ;

[0065] in, Indicates the charging power of the energy storage system. Indicates the discharge power of the energy storage system;

[0066] S323. Electrical Load Equipment Instructions The generation process is as follows:

[0067] In power-limited operation mode, controllable loads are adjusted according to the load regulation index. The power is reduced proportionally, that is ;

[0068] in, Indicates the actual power of the load;

[0069] In zero-carbon operation mode, only non-essential loads are adjusted;

[0070] In economic operation models, peak shifting and valley filling are used to move the load from peak hours to valley hours;

[0071] No power load equipment instructions are generated in power-limited operation mode, microgrid autonomous operation mode, and normal operation mode.

[0072] Furthermore, the specific steps of step S4 are as follows:

[0073] S41. Verify the instructions in the scheduling instruction set using security verification rules:

[0074] For new energy power generation equipment, the commanded power is limited to not exceeding its maximum available power, i.e. ;

[0075] in, This indicates the maximum available power of the new energy power generation equipment. It is an instruction for new energy power generation equipment;

[0076] For energy storage systems, the charging and discharging power should not exceed the rated power, and the state of charge (SOC) variation should be within the allowable range, i.e.:

[0077]

[0078]

[0079] in, Indicates the rated power of the energy storage system. Indicates the rated capacity of the energy storage system. Indicates the minimum permissible SOC of the energy storage system. Indicates the maximum permissible SOC of the energy storage system. Indicates commands for energy storage systems. It is the security verification time window for instruction execution;

[0080] For electrical load equipment, the commanded power should not exceed the current actual power and should not change the critical production load, i.e. ;

[0081] in, Indicates the actual power of the load. Indicates instructions for electrical load equipment;

[0082] S42. Real-time acquisition of data after instruction execution from the scheduling instruction set, and calculation of operational deviations. :

[0083]

[0084] in, Indicates the deviation in power generation from new energy sources. , This indicates the actual power generation capacity of the new energy equipment. Indicates the rated power of the new energy power generation system. It is an instruction for new energy power generation equipment;

[0085] Indicates the power deviation of the energy storage system. , Indicates the actual power of the energy storage system. Indicates the rated power of the energy storage system. Indicates commands for the energy storage system;

[0086] Indicates load power deviation. , Indicates the actual load power. The reference power representing the electrical load. Indicates instructions for electrical load equipment;

[0087] The weighting coefficients satisfy ;

[0088] S43. Determine if a State of Control (SOC) target exists;

[0089] If so, proceed to step S44;

[0090] If not, the operating deviation will be... As the final operating deviation Proceed to step S5;

[0091] S44. Calculate the SOC deviation term :

[0092]

[0093] in, Indicates the actual SOC. Indicates the target SOC;

[0094] Final running deviation ,in, The weight of the SOC deviation term. It's an operational deviation.

[0095] Furthermore, the specific steps of step S5 are as follows:

[0096] S51. Determine the final operating deviation Does it exceed the preset deviation threshold? :

[0097] like If so, Q-learning optimization is triggered, and step S52 is entered;

[0098] like If the current execution result is not found, the result will be stored as a success case in the knowledge base, and the process will end.

[0099] S52.Q Learning and Optimization Process:

[0100] The state space is defined as follows:

[0101]

[0102] Where mode is the operating mode code. Indicates the energy storage system adjustment index. Indicates the new energy adjustment index, Indicates the load adjustment index. It is an operational deviation;

[0103] Define the action space as a set of actions whose instruction weights are adjusted;

[0104] The rewards are defined as follows:

[0105]

[0106] in, This is a new operational deviation. It is the original operational deviation;

[0107] Table Q is updated as follows:

[0108]

[0109] in, For the new state, For new actions, α is the learning rate, and γ is the discount factor;

[0110] Based on the updated Q value, select the action that maximizes the Q value. To adjust instruction weights;

[0111] S53. Feed the adjusted instruction weights back to the knowledge base as the optimized strategy, and update the instruction generation rules for the corresponding operating mode.

[0112] Secondly, embodiments of this application also provide a microgrid dispatch control device based on strategy optimization, comprising:

[0113] The aggregated resource information collection module is used to collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store them in a comprehensive database;

[0114] The response capability assessment module is used to receive user power generation and consumption plans, combine real-time operation data and historical operation data, extract multi-resource dynamic response features through a pre-trained LSTM model, and output a regulation capability assessment matrix.

[0115] The scheduling control strategy selection module is used to receive the power grid scheduling plan, match the optimal operating mode from the knowledge base based on the regulation capability evaluation matrix, and generate a set of scheduling instructions for new energy power generation equipment, power load equipment and energy storage system.

[0116] The scheduling control instruction execution module is used to issue and execute scheduling instruction sets to the target device, collect execution data in real time, and calculate the running deviation.

[0117] The strategy optimization module is used to update the instruction weights using an online Q-learning algorithm when the running deviation exceeds the threshold, and then feeds the optimization strategy back to the knowledge base.

[0118] Thirdly, embodiments of this application also provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the microgrid scheduling and control method based on strategy optimization as described in the first aspect.

[0119] Fourthly, embodiments of this application also provide a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the microgrid scheduling and control method based on strategy optimization as described in the first aspect.

[0120] As can be seen from the above technical solutions, this application has the following advantages:

[0121] The microgrid dispatch control method, device, equipment, and medium based on strategy optimization provided in this application collect real-time operating data from new energy power generation equipment, power load equipment, and energy storage systems. Combined with user power generation and consumption plans, a pre-trained LSTM model is used to extract dynamic response features of multiple resources and generate a regulation capability assessment matrix. Based on the regulation capability assessment matrix, the optimal operating mode is matched from a knowledge base to generate a dispatch instruction set. An online Q-learning algorithm is then used to optimize operating deviations. This enables refined dispatching of the microgrid, improves resource utilization, enhances the flexibility and reliability of the microgrid, reduces operating costs, and improves the competitiveness of the microgrid in the electricity market. Attached Figure Description

[0122] To more clearly illustrate the technical solution of this application, the accompanying drawings used in the description will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0123] Figure 1 This is a flowchart illustrating the microgrid scheduling and control method based on strategy optimization according to the present invention.

[0124] Figure 2 This is a schematic diagram of the microgrid dispatch and control system based on strategy optimization according to the present invention. Detailed Implementation

[0125] The various embodiments of this disclosure will be described more fully in the following detailed description of the specific steps of the strategy-optimized microgrid dispatch control method. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0126] For example, with the rapid development of distributed new energy sources (such as solar photovoltaic and wind power) and adjustable loads (such as electric vehicles), new power systems have emerged. However, the large-scale integration of these distributed power sources and adjustable loads has led to a significant increase in the random fluctuations on both the source and load sides of the power system, posing unprecedented challenges to the traditional power grid that primarily uses a "source follows load" regulation mode.

[0127] As a crucial component of the new power system, smart microgrids leverage advanced energy management technologies to drive energy structure transformation, enhance the absorption capacity of renewable energy, and improve the flexibility and reliability of the power system. However, traditional microgrid dispatching technologies suffer from the following major drawbacks:

[0128] First, the dispatching model is relatively simplistic. Current dispatching mainly focuses on peak shaving and valley filling, and arbitrage using peak-valley electricity prices. It fails to develop targeted dispatching strategies based on the characteristics of different electricity users and resource scenarios, thus limiting resource utilization. For example, in industrial parks, different enterprises have significantly different production schedules and electricity consumption characteristics. The traditional single dispatching model cannot fully leverage the potential of distributed power sources and adjustable loads.

[0129] Secondly, resource coordination capabilities are insufficient. Existing dispatching methods mostly regulate renewable energy generation, energy storage systems, or adjustable loads in isolation, lacking dynamic coupling and collaborative optimization among multiple resources. This makes microgrids inflexible and unreliable when dealing with complex power demand scenarios, easily leading to power shortages or curtailment of wind and solar power.

[0130] Finally, scheduling strategies suffer from rigidity and inflexibility. Traditional scheduling typically relies on day-to-day manual instructions and lacks a feedback loop mechanism, resulting in a high deviation rate in the execution of scheduling instructions.

[0131] Furthermore, while virtual power plants (VPS) can perform microgrid dispatch, their current dispatch methods rely excessively on macro-level dispatch commands, neglecting users' personalized needs (such as comfort and production scheduling). This may reduce users' enthusiasm for participating in VPS dispatch, thereby affecting the synergistic interaction between microgrids and VPS. Simultaneously, the dispatch evaluation indicators for VPS are relatively singular, using only grid deviation as the assessment standard, and an effective strategy self-optimization mechanism has not yet been established. This makes it difficult for microgrids to adapt to dynamic changes in market rules when participating in electricity market transactions, hindering their full realization of potential in terms of flexibility and economic benefits.

[0132] To address the aforementioned issues, this embodiment provides a microgrid scheduling and control method based on strategy optimization. Through real-time data acquisition and intelligent optimization algorithms, it achieves refined scheduling of the microgrid, improves resource utilization and operational flexibility, and reduces operating costs.

[0133] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0134] Please see Figure 1 The diagram shows a flowchart of a microgrid scheduling and control method based on strategy optimization in a specific embodiment. The method includes the following steps:

[0135] S1. Collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store them in a comprehensive database;

[0136] It should be noted that the data collection provides a data foundation for subsequent adjustment capability assessment and dispatch instruction generation, ensuring the accuracy and integrity of the data and improving the reliability of dispatch control;

[0137] S2. Receive user power generation and consumption plans, combine real-time operation data and historical operation data, extract multi-resource dynamic response features through a pre-trained LSTM model, and output a regulation capability evaluation matrix;

[0138] It should be noted that by extracting the dynamic response features of multiple resources through the LSTM model, the adjustment capabilities of each resource can be accurately assessed, providing a scientific basis for the generation of scheduling instructions and improving the accuracy and adaptability of scheduling.

[0139] S3. Receive the power grid dispatch plan, and based on the regulation capacity assessment matrix, match the optimal operating mode from the knowledge base to generate a set of dispatch instructions for new energy power generation equipment, power load equipment and energy storage system;

[0140] It should be noted that by selecting the optimal operating mode through decision trees, the grid dispatching needs can be responded to quickly, reasonable dispatching instructions can be generated, and the operating efficiency and flexibility of the microgrid can be improved.

[0141] S4. Issue a set of scheduling instructions to the target device and execute them, collect execution data in real time and calculate the running deviation;

[0142] It should be noted that by monitoring the execution effect of scheduling instructions in real time and calculating the operational deviation, data support is provided for subsequent strategy optimization, thereby improving the accuracy and reliability of scheduling control.

[0143] S5. When the running deviation exceeds the threshold, the instruction weights are updated using the online Q-learning algorithm, and the optimization strategy is fed back to the knowledge base;

[0144] It should be noted that by updating the command weights, the scheduling strategy can be dynamically adjusted according to the real-time operating conditions, continuously optimizing the scheduling commands and improving the operating efficiency and stability of the microgrid.

[0145] This embodiment achieves refined scheduling of microgrids through real-time data acquisition, intelligent regulation capability assessment, and optimization algorithms, thereby improving resource utilization and operational flexibility, reducing operating costs, and enhancing the competitiveness of microgrids in the electricity market.

[0146] Furthermore, as a refinement and extension of the specific implementation methods of the above embodiments, in order to fully illustrate the specific implementation process of this embodiment, another microgrid scheduling and control method based on strategy optimization is provided, which includes the following steps:

[0147] S1. Collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store them in a comprehensive database;

[0148] The specific steps of step S1 are as follows:

[0149] S11. Collect three types of resource data and environmental parameters through a multi-protocol interface:

[0150] Obtain the active power of new energy power generation equipment via Modbus / TCP protocol. Electricity generation and running status ;

[0151] Real-time power of electrical load equipment is collected via IoT network. Electricity consumption and controllable identifier ;

[0152] The charging and discharging power of the energy storage system is collected via CAN bus. SOC status and rated capacity ;

[0153] Ambient temperature is collected by sensors. and irradiance ;

[0154] S12. Store the collected real-time operational data in the integrated database according to the timestamp, and perform spatiotemporal alignment:

[0155] ;

[0156] in, This is a vector of real-time running data, where t represents a timestamp;

[0157] S13. Preprocess the collected real-time data, including missing value imputation and outlier handling;

[0158] Missing value supplementation:

[0159] Outlier handling: If Then replace with the moving average. ;

[0160] S2. Receive user power generation and consumption plans, combine real-time operation data and historical operation data, extract multi-resource dynamic response features through a pre-trained LSTM model, and output a regulation capability evaluation matrix;

[0161] The specific steps of step S2 are as follows:

[0162] S21. Construct the LSTM model:

[0163] The LSTM model is set to input a feature sequence with a time window length of k. ;

[0164] Wherein, the feature vector at time t ;

[0165] The LSTM network architecture consists of an input layer, a first LSTM layer, a second LSTM layer, a fully connected layer, and an output layer connected in sequence.

[0166] The input layer adopts a fully connected layer architecture, mapping the 5×k-dimensional input to a 5k-dimensional output. The first LSTM layer includes 64 LSTM units and has a forget gate to map the 5k-dimensional input to a 64-dimensional output. The second LSTM layer includes 32 LSTM units, mapping the 64-dimensional input to a 32-dimensional output. The fully connected layer linearly maps to the output dimension, mapping the 32-dimensional dimension to 3-dimensional dimension. The output layer uses Sigmoid activation to generate a modulating vector.

[0167] The output is set as a regulation capability assessment matrix represented by a normalized regulation capability index vector:

[0168]

[0169] in, This indicates power regulation capability, specifically the power response speed regulation capability. This indicates the power regulation capability, specifically the ability to regulate the duration of continuous regulation. This refers to the SOC state regulation capability, specifically the SOC change rate regulation capability index.

[0170] S22. Divide the historical running data into feature sequences and construct a training dataset;

[0171] Set a loss function and pre-train the LSTM model until the loss function converges or the maximum number of iterations is met;

[0172] The loss function is as follows:

[0173]

[0174] Where N represents the number of samples in the training dataset, This represents the actual rate of change of power. This is the actual energy regulation amount. The rate of change of SOC For power conversion time, Typical adjustment duration, For response time;

[0175] S23. Divide the real-time running data according to the time window k to obtain the real-time data window. The pre-trained LSTM model is input, and forward propagation is performed to obtain the output conditioning evaluation capability matrix:

[0176]

[0177] In this matrix, the rows of the adjustment assessment capability matrix represent new energy power generation equipment, energy storage systems, and power load equipment, respectively, and the columns represent power regulation, power regulation, and SOC regulation, respectively.

[0178] It should be noted that new energy power generation equipment and electricity load do not have SOC regulation, so they are set to 0 in the matrix;

[0179] Perform forward propagation:

[0180]

[0181]

[0182] S24. Calculate the resource integration adjustment index and quantify the response capability as follows:

[0183]

[0184] in, To adjust the preset weights corresponding to the elements in the assessment capability matrix, Indicates the energy storage system adjustment index. Indicates the new energy adjustment index, Indicates the load adjustment index;

[0185] For example, if , , ,but: , , ;

[0186] S3. Receive the power grid dispatch plan, and based on the regulation capacity assessment matrix, match the optimal operating mode from the knowledge base to generate a set of dispatch instructions for new energy power generation equipment, power load equipment and energy storage system;

[0187] The specific steps of step S3 are as follows:

[0188] S31. Based on grid connection point power constraints in power grid dispatching plans and daily electricity consumption constraints Combining the adjustment capability assessment matrix C, the optimal operating mode is selected through the preset decision tree in the knowledge base;

[0189] The specific decisions are as follows:

[0190] If the power grid dispatch plan requires that the power at the grid connection point not exceed the power constraint at the grid connection point. If so, select the power-limited operation mode;

[0191] If the power grid dispatch plan requires that daily electricity consumption not exceed the daily electricity consumption constraint If so, select the power-limited operation mode;

[0192] If the power grid dispatch plan requires zero-carbon operation, then the zero-carbon operation mode shall be selected;

[0193] If the power grid dispatch plan has no constraints on adjustment, then the economic operation mode shall be selected;

[0194] If the power grid is disconnected from the main grid, the microgrid autonomous operation mode is selected;

[0195] Otherwise, select normal operating mode;

[0196] It should be noted that zero-carbon operation means that the amount of renewable energy generated plus the amount of purchased green electricity is greater than or equal to the amount of electricity consumed.

[0197] S32. Based on the selected optimal operating mode, combined with the adjustment capability assessment matrix C and the resource fusion adjustment index... Generate scheduling instruction set ;

[0198] in, It is an instruction on new energy power generation equipment. It is an instruction from the energy storage system. It is an instruction for electrical load equipment;

[0199] The specific process of generating the scheduling instruction set in step S32 is as follows:

[0200] S321. Directive on New Energy Power Generation Equipment The generation process is as follows:

[0201] In power-limited operation mode, set ;

[0202] in, This indicates the maximum available power of the new energy power generation equipment. Indicates the new energy adjustment index, It is the maximum available power allowed by the power grid;

[0203] Setting up a zero-carbon operation mode with the goal of maximizing new energy sources ;

[0204] In power-limited operation mode, economic operation mode, microgrid autonomous operation mode, and normal operation mode, active power priority is set as the objective. ;

[0205] S322. Energy Storage System Command The generation process is as follows:

[0206] Setting the power-limited operation mode ;

[0207] in, Indicates the actual power of the load. Indicates the energy storage system adjustment index. It is the maximum available power allowed by the power grid;

[0208] Setting the zero-carbon operation mode ;

[0209] in, It is the total electricity demand of the load. It is renewable energy power generation. It is purchased green electricity. Indicates the energy storage system adjustment index. It is the security verification time window for instruction execution;

[0210] In the economic operation mode, charging is set up during off-peak hours according to time-of-use pricing. Peak discharge settings ;

[0211] in, Indicates the charging power of the energy storage system. Indicates the discharge power of the energy storage system;

[0212] S323. Electrical Load Equipment Instructions The generation process is as follows:

[0213] In power-limited operation mode, controllable loads are adjusted according to the load regulation index. The power is reduced proportionally, that is ;

[0214] in, This indicates the actual load power; in zero-carbon operation mode, only non-essential loads (such as delayed charging pile charging) are adjusted.

[0215] In economic operation models, peak shifting and valley filling are used to move the load from peak hours to valley hours;

[0216] No power load equipment instructions are generated in power-limited operation mode, microgrid autonomous operation mode, and normal operation mode.

[0217] S33. Store the generated scheduling instruction set in the knowledge base and record the basis for selecting the optimal operating mode and the instruction generation parameters;

[0218] S4. Issue a set of scheduling instructions to the target device and execute them, collect execution data in real time and calculate the running deviation;

[0219] The specific steps of step S4 are as follows:

[0220] S41. Verify the instructions in the scheduling instruction set using security verification rules:

[0221] For new energy power generation equipment, the commanded power is limited to not exceeding its maximum available power, i.e. ;

[0222] in, This indicates the maximum available power of the new energy power generation equipment. It is an instruction for new energy power generation equipment;

[0223] For energy storage systems, the charging and discharging power should not exceed the rated power, and the state of charge (SOC) variation should be within the allowable range, i.e.:

[0224]

[0225]

[0226] in, Indicates the rated power of the energy storage system. Indicates the rated capacity of the energy storage system. Indicates the minimum permissible SOC of the energy storage system. Indicates the maximum permissible SOC of the energy storage system. Indicates commands for energy storage systems. It is the security verification time window for instruction execution;

[0227] For electrical load equipment, the commanded power should not exceed the current actual power and should not change the critical production load, i.e. ;

[0228] in, Indicates the actual power of the load. Indicates instructions for electrical load equipment;

[0229] S42. Real-time acquisition of data after instruction execution from the scheduling instruction set, and calculation of operational deviations. :

[0230]

[0231] in, Indicates the deviation in power generation from new energy sources. , This indicates the actual power generation capacity of the new energy equipment. Indicates the rated power of the new energy power generation system. It is an instruction for new energy power generation equipment;

[0232] Indicates the power deviation of the energy storage system. , Indicates the actual power of the energy storage system. Indicates the rated power of the energy storage system. Indicates commands for the energy storage system;

[0233] Indicates load power deviation. , Indicates the actual load power. This represents the reference power of the electrical load. The reference power representing the electrical load. Indicates instructions for electrical load equipment;

[0234] The weighting coefficients satisfy ;

[0235] S43. Determine if a State of Control (SOC) target exists;

[0236] If so, proceed to step S44;

[0237] If not, the operating deviation will be... As the final operating deviation Proceed to step S5;

[0238] For example, a zero-carbon operating mode has a SOC control target;

[0239] S44. Calculate the SOC deviation term :

[0240]

[0241] in, Indicates the actual SOC. Indicates the target SOC;

[0242] Final running deviation ,in, The weight of the SOC deviation term. It is an operational deviation;

[0243] S5. When the operational deviation exceeds the threshold, the instruction weights are updated using an online Q-learning algorithm, and the optimization strategy is fed back to the knowledge base; the specific steps of step S5 are as follows:

[0244] S51. Determine the final operating deviation Does it exceed the preset deviation threshold? :

[0245] like If so, Q-learning optimization is triggered, and step S52 is entered;

[0246] like If the current execution result is not found, the result will be stored as a success case in the knowledge base, and the process will end.

[0247] S52.Q Learning and Optimization Process:

[0248] The state space is defined as follows:

[0249]

[0250] Where mode is the operating mode code. Indicates the energy storage system adjustment index. Indicates the new energy adjustment index, Indicates the load adjustment index. It is an operational deviation;

[0251] Define the action space as a set of actions whose instruction weights are adjusted;

[0252] For example: To increase the energy storage regulation weight , To reduce load adjustment weight , To adjust the proportion of new energy vehicle emissions restricted;

[0253] The rewards are defined as follows:

[0254]

[0255] in, This is a new operational deviation. It is the original operational deviation;

[0256] Table Q is updated as follows:

[0257]

[0258] in, For the new state, For new actions, α is the learning rate, and γ is the discount factor;

[0259] Based on the updated Q value, select the action that maximizes the Q value. To adjust instruction weights;

[0260] S53. Feed the adjusted instruction weights back to the knowledge base as the optimized strategy, and update the instruction generation rules for the corresponding operating mode.

[0261] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0262] like Figure 2 As shown, the following are embodiments of the microgrid dispatch control device based on strategy optimization provided in this disclosure. This device and the microgrid dispatch control method based on strategy optimization in the above embodiments belong to the same inventive concept. For details not described in detail in the embodiments of the microgrid dispatch control device based on strategy optimization, please refer to the embodiments of the microgrid dispatch control method based on strategy optimization described above.

[0263] The device includes:

[0264] The aggregated resource information collection module is used to collect real-time operating data of new energy power generation equipment, power load equipment and energy storage systems, and store them in a comprehensive database;

[0265] The response capability assessment module is used to receive user power generation and consumption plans, combine real-time operation data and historical operation data, extract multi-resource dynamic response features through a pre-trained LSTM model, and output a regulation capability assessment matrix.

[0266] The scheduling control strategy selection module is used to receive the power grid scheduling plan, match the optimal operating mode from the knowledge base based on the regulation capability evaluation matrix, and generate a set of scheduling instructions for new energy power generation equipment, power load equipment and energy storage system.

[0267] The scheduling control instruction execution module is used to issue and execute scheduling instruction sets to the target device, collect execution data in real time, and calculate the running deviation.

[0268] The strategy optimization module is used to update the instruction weights using an online Q-learning algorithm when the running deviation exceeds the threshold, and then feeds the optimization strategy back to the knowledge base.

[0269] This embodiment achieves refined scheduling of microgrids by integrating the resource information acquisition module, response capability assessment module, scheduling control strategy selection module, scheduling control command execution module, and strategy optimization module. This improves resource utilization and operational flexibility, reduces operating costs, and enhances the competitiveness of microgrids in the electricity market.

[0270] The microgrid scheduling and control method based on strategy optimization provided in this application can be applied to electronic devices. Those skilled in the art will understand that the electronic device structure involved in the embodiments of this invention does not constitute a limitation on the electronic device. An electronic device may include more or fewer components than illustrated, or combine certain components, or have different component arrangements. In the embodiments of this invention, the electronic device includes, but is not limited to, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the embodiments of this application described and / or claimed herein.

[0271] Electronic devices may include processors, external memory interfaces, internal memory, universal serial bus (USB) interfaces, charging management modules, power management modules, batteries, wireless communication modules, audio modules, speakers, microphones, sensor modules, buttons, cameras, displays, and SIM card interfaces, etc.

[0272] It is understood that the structures illustrated in the embodiments of this application do not constitute a specific limitation on the electronic device. In other embodiments of this application, the electronic device may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0273] A processor may include one or more processing units, such as a central processing unit (CPU), an application processor (AP), a modem processor, a graphics processing unit (GPU), an image signal processor (ISP), a controller, memory, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural network processing unit (NPU). Different processing units may be independent devices or integrated into one or more processors.

[0274] The processor can serve as the nerve center and command center of an electronic device. The controller can generate operation control signals based on the instruction opcode and timing signals to control the fetching and execution of instructions.

[0275] The processor may also include memory for storing instructions and data. In some embodiments, the memory in the processor is a cache memory. This memory can store instructions or data that the processor has just used or that are used repeatedly. If the processor needs to use the instruction or data again, it can retrieve it directly from this memory. This avoids repeated accesses, reduces processor latency, and thus improves system efficiency.

[0276] The aforementioned electronic equipment implements the strategy optimization-based microgrid dispatch control method of this application, which collects real-time operating data of new energy power generation equipment, power load equipment, and energy storage systems, and stores it in a comprehensive database; receives user power generation and consumption plans, combines real-time operating data and historical operating data, extracts multi-resource dynamic response features through a pre-trained LSTM model, and outputs a regulation capability evaluation matrix; receives grid dispatch plans, matches the optimal operating mode from the knowledge base based on the regulation capability evaluation matrix, and generates a dispatch instruction set for new energy power generation equipment, power load equipment, and energy storage systems; issues and executes the dispatch instruction set to the target equipment, collects execution data in real time, and calculates operating deviations; when the operating deviation exceeds a threshold, it uses an online Q-learning algorithm to update the instruction weights and feeds the optimization strategy back to the knowledge base. This technical solution achieves the beneficial effects of realizing refined microgrid dispatch, improving resource utilization and operational flexibility, reducing operating costs, and enhancing the competitiveness of microgrids in the electricity market.

[0277] The storage medium provided in this application stores a program product capable of implementing a microgrid scheduling and control method based on strategy optimization.

[0278] The microgrid dispatch control method based on strategy optimization includes: collecting real-time operating data of new energy power generation equipment, power load equipment, and energy storage systems, and storing them in a comprehensive database; receiving user power generation and consumption plans, combining real-time operating data and historical operating data, extracting multi-resource dynamic response features through a pre-trained LSTM model, and outputting a regulation capability evaluation matrix; receiving grid dispatch plans, matching the optimal operating mode from a knowledge base based on the regulation capability evaluation matrix, and generating a dispatch instruction set for new energy power generation equipment, power load equipment, and energy storage systems; issuing and executing the dispatch instruction set to the target equipment, collecting execution data in real time, and calculating operating deviations; when the operating deviation exceeds a threshold, updating the instruction weights using an online Q-learning algorithm, and feeding the optimization strategy back to the knowledge base.

[0279] In some possible implementations, the strategy-optimized microgrid dispatch control method of this disclosure can be implemented as a program product comprising program code that, when the program product is run on a terminal device, causes the terminal device to perform the steps described in the "Exemplary Methods" section of this specification according to various exemplary embodiments of this disclosure.

[0280] The storage medium disclosed herein may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0281] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for microgrid dispatching control based on policy optimization, characterized in that, The method comprises the following steps: S1. Collecting real-time operation data of new energy power generation equipment, power load equipment and energy storage system, and storing them into a comprehensive database; S2. Receiving user power generation and power consumption plans, combining real-time operation data and historical operation data, extracting multi-resource dynamic response features through a pre-trained LSTM model, and outputting an adjustment capacity evaluation matrix; S3. Receiving a power grid dispatching plan, matching an optimal operation mode from a knowledge base based on the adjustment capacity evaluation matrix, and generating a set of dispatching instructions for the new energy power generation equipment, power load equipment and energy storage system; S4. Issuing the set of dispatching instructions to the target equipment and executing them, collecting execution data in real time and calculating operation deviations; Real-time acquisition of data after execution of instructions in the dispatch instruction set, calculation of operation deviation : wherein, represents the new energy power deviation, , represents the actual new energy equipment power generation, represents the new energy power generation system rated power, is the new energy power generation equipment instruction; PES_Power_Offset represents the power deviation of the energy storage system, , PES_Actual_Power represents the actual power of the energy storage system, PES_Rated_Power represents the rated power of the energy storage system, PES_Command represents the command of the energy storage system; represents a load power deviation, , represents an actual load power, represents a reference power of an electric load, represents an electric load device instruction; The weight coefficients satisfy ; Judging whether there is a SOC control target; If yes, calculate SOC bias term : wherein, represents the actual SOC, represents the target SOC; final run bias wherein, is a weight for the SOC bias term, is a run bias; If not, run deviation As final run deviation Enter step S5; S5. When the operation deviation exceeds a threshold, an online Q-learning algorithm is used to update instruction weights, and the optimization strategy is fed back to the knowledge base. 2.The microgrid dispatch control method based on policy optimization of claim 1, wherein, The specific steps of step S1 are as follows: S11. Collecting three types of resource data and environmental parameters through a multi-protocol interface: Obtain active power of new energy power generation equipment through Modbus / TCP protocol , power generation and running state ; Collecting real-time power of power load equipment through IOT network , power consumption and controllable identification ; Collecting charge and discharge power of energy storage system through CAN bus , SOC state and rated capacity ; Collecting environmental temperature by sensor and irradiance ; S12. Storing the collected real-time operation data into the comprehensive database according to the time stamp, and performing time and space alignment: ； wherein, is a vector of real-time running data, t denotes a time stamp; S13. Preprocessing the collected real-time data, performing missing value filling and outlier processing. 3.The microgrid dispatch control method based on policy optimization of claim 2, wherein, The specific steps of step S2 are as follows: S21. Constructing an LSTM model: Setting the LSTM model input to be a sequence of features of length k ; wherein the feature vector at time t is ; The LSTM network architecture comprises an input layer, a first LSTM layer, a second LSTM layer, a full connection layer and an output layer connected in sequence; The output is set as an adjustment capacity evaluation matrix represented by a normalized adjustment capacity index vector: wherein, represents the power adjustment capability, i.e. the power response speed adjustment capability index, represents the power adjustment capability, i.e. the power response speed adjustment capability index, represents the power adjustment capability, i.e. the power response speed adjustment capability index, S22. Dividing the historical operation data into feature sequences to construct a training data set; Setting a loss function, pre-training the LSTM model until the loss function converges or the maximum number of iterations is met; The loss function is as follows: wherein N represents the number of samples in the training data set, is the actual power change rate, is the actual energy adjustment rate, is the SOC change rate, is the power conversion duration, is the typical adjustment duration, is the response time; S23. The real-time running data is divided according to the time window k to obtain a real-time data window and input the pre-trained LSTM model, perform forward propagation, and obtain an output adjustment capability evaluation matrix: Wherein, the rows of the adjustment capacity evaluation matrix represent the new energy power generation equipment, the energy storage system and the power load equipment in sequence, and the columns represent the power adjustment, the power adjustment and the SOC adjustment; S24. Calculating the resource fusion adjustment index by the following method to quantify the response capacity: wherein, is a preset weight corresponding to an element in the regulation capability evaluation matrix, denotes a resource integration regulation index, denotes a energy storage system regulation index, denotes a new energy regulation index, denotes a load regulation index. 4.The microgrid dispatch control method based on policy optimization of claim 3, wherein, The specific steps of step S3 are as follows: S31. Based on the grid connection point power constraints in the grid dispatch plan and daily electricity consumption constraints , combined with the regulation capacity evaluation matrix C, the optimal operation mode is selected through the preset decision tree in the knowledge base; The specific decision is as follows: If the grid dispatch plan requires that the point of grid connection power does not exceed the point of grid connection power constraint , then the limited power operation mode is selected; If the grid dispatching plan requires that the daily power consumption does not exceed the daily power consumption constraint , the power consumption limiting mode is selected; If the power grid dispatching plan requires zero-carbon operation, select the zero-carbon operation mode; If the power grid dispatching plan has no constraint adjustment, select the economic operation mode; If the power grid is disconnected from the main grid, select the microgrid autonomous operation mode; Otherwise, select the normal operation mode; S32. According to the selected optimal operation mode, the regulation capacity evaluation matrix C and the resource fusion regulation index are combined to generate a scheduling instruction set ; wherein, is a new energy power generation device instruction, is an energy storage system instruction, is an electrical load device instruction; S33. Storing the generated set of dispatching instructions into the knowledge base, and recording the optimal operation mode selection basis and instruction generation parameters.

5. The method of claim 4, wherein, The specific generation process of the set of dispatching instructions in step S32 is as follows: S321. New energy power generation equipment instruction The generation process is as follows: In the limited power operation mode, set ; wherein, represents the maximum available power of the new energy power generation equipment, represents the new energy regulation index, is the grid-connected point power constraint; In a zero-carbon operation mode, set a target of maximizing new energy ; In the limited power operation mode, the economic operation mode, the micro-grid autonomous operation mode and the normal operation mode, the active power is prioritized ; S322. Energy storage system instructions The generation process is as follows: In a power-limited mode of operation ; wherein, represents the load actual power, represents the energy storage system regulation index, is the grid point power constraint; In a zero-carbon operation mode setting ; Wherein, is the total load demand power, is the new energy power generation, is the green power purchase amount, represents the energy storage system adjustment index, is the safety check time window of instruction execution; In the economic operation mode, according to the time-of-use electricity price, the valley period is set for charging , and the peak period is set for discharging ; wherein, represents the charging power of the energy storage system, represents the discharging power of the energy storage system; S323. Use electrical load device instructions The generation process is as follows: In the limited power operation mode, the controllable load is adjusted according to the load adjustment index The power is proportionally reduced, i.e. ; wherein, represents the actual load power; In the zero-carbon operation mode, only unnecessary loads are adjusted; In the economic operation mode, peak shifting is performed to move the load from the peak period to the valley period; In the limited power consumption mode, the microgrid autonomous operation mode and the normal operation mode, no power load equipment instructions are generated. 6.The microgrid dispatch control method based on policy optimization of claim 4, wherein, Step S4 further comprises the following steps: Verify the instructions in the set of dispatching instructions through safety verification rules: For new energy power generation equipment, the limit instruction power is not more than its maximum available power, that is ; wherein, represents the maximum available power of the new energy power generation equipment, is the new energy power generation equipment instruction; For the energy storage system, set the charging and discharging power not to exceed the rated power, and the SOC change to be within the allowed range, i.e. wherein, represents the rated power of the energy storage system, represents the rated capacity of the energy storage system, represents the minimum allowed SOC of the energy storage system, represents the maximum allowed SOC of the energy storage system, represents the instruction of the energy storage system, is the safety check time window of the instruction execution; For the electrical load device, set the command power not more than the current actual power, and do not change the production key load, that is ; wherein, represents the load actual power, represents the power consumption load device command. 7.The microgrid dispatch control method based on policy optimization of claim 6, wherein, The specific steps of step S5 are as follows: S51. Determine final operating deviation whether the preset deviation threshold is exceeded : If Q-learning optimization is triggered and step S52 is entered. If the current execution result is stored as a success case in the knowledge base, and the process ends. S52. Q-learning optimization process: Define the state space as follows: wherein mode is a mode of operation code, denotes a storage system regulation index, denotes a new energy regulation index, denotes a load regulation index, is an operating deviation; Define the action space as a set of actions for adjusting the instruction weights; The reward is defined as follows: wherein is a new operating deviation, is an original operating deviation; The Q table is updated as follows: wherein, is the new state, is the new action, a is the learning rate, and g is the discount factor; selecting an action that maximizes the Q value according to the updated Q value to adjust the instruction weight; S53. The adjusted instruction weight is fed back to the knowledge base as the optimized strategy, and the instruction generation rule of the corresponding operation mode is updated.

8. A microgrid dispatch control device based on policy optimization, adopting the microgrid dispatch control method based on policy optimization in any one of claims 1-7, characterized in that, Comprise: The aggregated resource information collection module is configured to collect real-time operation data of the new energy power generation equipment, the power consumption load equipment, and the energy storage system, and store the real-time operation data into the comprehensive database; The response capability evaluation module is configured to receive a user power generation and power consumption plan, combine real-time operation data and historical operation data, extract multi-resource dynamic response features through a pre-trained LSTM model, and output an adjustment capability evaluation matrix; The dispatching control strategy selection module is configured to receive a power grid dispatching plan, match an optimal operation mode from the knowledge base based on the adjustment capability evaluation matrix, and generate a set of dispatching instructions for the new energy power generation equipment, the power consumption load equipment, and the energy storage system; The dispatching control instruction execution module is configured to issue and execute the set of dispatching instructions to the target equipment, collect execution data in real time, and calculate an operation deviation; The strategy optimization module is configured to update the instruction weight using an online Q learning algorithm when the operation deviation exceeds a threshold value, and feed back the optimized strategy to the knowledge base.

9. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the program to implement the steps of the microgrid dispatching control method based on strategy optimization according to any one of claims 1 to 7.

10. A storage medium having stored thereon a computer program, characterized in that The computer program is executed by the processor to implement the steps of the microgrid dispatching control method based on strategy optimization according to any one of claims 1 to 7.

Images