Virtual power plant intelligent regulation and control method and system based on dynamic operation monitoring

Abstract

The invention discloses a virtual power plant intelligent regulation and control method and system based on dynamic operation monitoring, and the method specifically comprises the steps: integrating a real-time data collection network of a distributed power supply, an energy storage device and a load, carrying out the data preprocessing through edge calculation and AI, and setting a multi-dimensional optimization target; solving the constructed comprehensive optimization model by adopting a fast non-dominated sorting multi-target particle swarm algorithm with an elitist strategy, and obtaining an optimal energy efficiency improvement scheme; an optimal energy efficiency improvement scheme is deployed, an integrated intelligent power distribution network is constructed, a block chain technology is applied to record an operation state, an intelligent monitoring system is established, the energy efficiency trend is predicted by using the Internet of Things and machine learning, the scheme is dynamically adjusted and optimized, and an effect is evaluated; through a user feedback mechanism and big data analysis, load side management is optimized, and energy-saving potential is mined; according to the invention, the energy efficiency of the power distribution network is improved, and intelligent and efficient energy management is realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of virtual power plants, and specifically relates to an intelligent control method and system for virtual power plants based on dynamic operation monitoring. Background Art

[0002] With the growing demand for energy and the urgent need for clean energy utilization, virtual power plants have gradually emerged as a new power system solution. Virtual power plants integrate distributed energy resources, such as distributed photovoltaic power generation, wind power generation, small hydropower, various energy storage devices, controllable loads, etc., and treat them as a unified power plant for coordinated management. However, virtual power plants currently face many challenges in their operation. On the one hand, the output of distributed energy is intermittent and volatile. For example, photovoltaic power generation is greatly affected by light intensity and weather changes, and wind power generation fluctuates with unstable wind speed, which makes it difficult for the overall output power of virtual power plants to be stable. On the other hand, traditional control methods are difficult to adapt to the complex and changeable operating conditions of virtual power plants in real time and accurately, and cannot give full play to the efficiency of each component, resulting in low energy utilization efficiency, and may even affect the safe and stable operation of the power grid.

[0003] For example, the Chinese patent application with the authorization announcement number CN112016838B discloses a method, system and terminal equipment for calculating the contribution rate of the distribution network energy efficiency index system, the method includes: establishing an energy efficiency index system from a medium-voltage distribution network, a flexible substation and a low-voltage distribution network, combining the third-level index and the second-level index in the energy efficiency index system to calculate the second subjective weight of the third-level index, and also calculating the second objective weight of the third-level index for the third-level index in the energy efficiency index system, and applying weighting and fusion processing to the second subjective weight and the second objective weight of the third-level index to obtain the contribution rate of the third-level index, and proposing an improvement plan for improving the energy efficiency level of the power electronic distribution network based on the third-level index with a large contribution rate. This technical solution can establish a comprehensive and appropriate energy efficiency index system for energy efficiency influencing factors, and adopts a more scientific and effective calculation of the contribution rate of the third-level index corresponding to the influencing factors by combining weighting, so as to provide accurate guidance and reasonable planning for improving the energy efficiency level of the power electronic distribution network and improving the planning, transformation and construction of the distribution network.

[0004] The above existing technologies all have the following problems: 1) Although a fusion process based on subjective and objective weights is proposed to calculate the contribution rate, it is more inclined to static analysis and lacks real-time and dynamic performance; 2) The contribution rate is calculated to identify the tertiary indicators that have a greater impact on energy efficiency, and improvement plans are proposed accordingly, but this optimization process is relatively indirect and requires multiple iterations and adjustments to find the optimal energy efficiency improvement plan; 3) The system integration and intelligence level are low. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a virtual power plant intelligent control method and system based on dynamic operation monitoring, which integrates a real-time data acquisition network of distributed power sources, energy storage equipment and loads, and sets multi-dimensional optimization goals through edge computing and AI pre-processing data; adopts a fast non-dominated sorting multi-objective particle swarm algorithm with an elite strategy to solve the constructed comprehensive optimization model to obtain the optimal energy efficiency improvement plan; deploys the optimal energy efficiency improvement plan, builds an integrated intelligent distribution network, and applies blockchain technology to record the operating status, establishes an intelligent monitoring system, uses the Internet of Things and machine learning to predict energy efficiency trends, dynamically adjusts the optimization plan, and evaluates the effect; optimizes load-side management and taps energy-saving potential through user feedback mechanism and big data analysis; the present invention improves the energy efficiency of the distribution network and realizes intelligent and efficient energy management.

[0006] To achieve the above object, the present invention provides the following technical solutions: The intelligent control method of virtual power plant based on dynamic operation monitoring includes: Step S1: Establish an intelligent data acquisition system based on the Internet of Things, integrate the sensor network of distributed power sources, energy storage equipment, and loads, collect data in real time, and pre-process the collected data through edge computing and AI algorithms. Based on the pre-processed data, set the multi-dimensional optimization goals and decision variables of the distribution network; Step S2: According to the set multi-dimensional optimization objectives and decision variables of the distribution network, a comprehensive optimization model of the distribution network is constructed, and the fast non-dominated sorting multi-objective particle swarm algorithm with elite strategy is used to solve the comprehensive optimization model of the distribution network, and the optimal energy efficiency improvement plan is obtained according to the optimal solution set; Step S3: Deploy the obtained optimal energy efficiency improvement solution to the distribution network for infrastructure construction, build an integrated smart distribution network system, and use blockchain technology to record the operating status of the facilities; Step S4: Establish an intelligent monitoring system, use IoT sensors and AI analysis algorithms to monitor the energy efficiency of the distribution network in real time, use machine learning algorithms to build an energy efficiency prediction model, predict the trend of energy efficiency changes, and dynamically adjust the optimal energy efficiency improvement plan through an adaptive learning mechanism, combining monitoring data and prediction results. Use a neural network algorithm to regularly evaluate the energy efficiency improvement effect, compare and analyze it with the set target, and adjust the optimal energy efficiency improvement plan based on the comparison results; Step S5: Establish a user feedback mechanism and use big data analysis methods to conduct in-depth mining of user electricity usage behaviors, identify electricity usage patterns and energy-saving potential, and optimize load-side management based on the identification results.

[0007] Specifically, the specific steps of step S2 include: S2.1: Set multi-dimensional optimization goals and construct comprehensive target optimization functions At the same time, establish constraints, x represents the decision variable; S2.2: Set the parameters of the particle swarm algorithm and randomly initialize the position and velocity of the particles; S2.3: Calculate the objective function value of each particle according to its position, and apply fast non-dominated sorting to stratify the population and calculate the value of each particle. i The crowding distance .

[0008] Specifically, the specific steps of step S2 also include: S2.4: Update according to the speed and position update formula of the particle swarm algorithm to obtain the updated position and speed, and perform fast non-dominated sorting on the updated population according to S2.3 to evaluate the quality of the updated particles; S2.5: Merge the non-dominated solutions in the current population with the non-dominated solutions in the parent population to form an extended solution set, and then select the solution based on the fast non-dominated sorting and crowding distance. Select a new parent population to continue the next round of iteration; S2.6: Save the set of non-dominated solutions found through an external archive, and update the external archive in each iteration, remove old solutions dominated by new solutions, and add new non-dominated solutions; S2.7: Check whether the maximum number of iterations has been reached. If so, stop the iteration process and output the optimal solution set in the external archive.

[0009] Specifically, the specific formula of the constraint condition in S2.1 is: ,in, represents the equality constraint, represents the inequality constraint, p represents the number of equality constraints, q represents the number of inequality constraints, represents the minimum function, , .

[0010] Specifically, the specific steps of S2.3 include: S2.31: Calculate each particle i The objective function value of ,in, Indicates n The particle in the optimization target i The objective function of S2.32: For each particle i , set the dominance i The number of particles and quilt i Dominated particle set , and traverse all particle pairs in the population, updating each particle and , among which, if , it is represented as a non-dominated particle and is also a non-dominated solution in the current population.

[0011] Specifically, the specific steps of S2.3 also include: S2.33: All The particles are placed in the first layer F and the particles are updated Each particle in , repeat this process until all particles are delaminated; S2.34: Initialize each particle i The crowding distance makes , for each optimization objective, Sort by, Represents particles k The objective function value at the same layer.

[0012] Specifically, the specific steps of S2.3 also include: S2.35: For the sorted boundary particles, set their crowding distance to ; For the sorted non-boundary particles, the formula Calculate its crowding distance, where represents the accumulated crowding distance, represents the scaling factor, and Respectively represent the optimization objectives j On the particle i The objective function value of the adjacent particles before and after is, and Respectively represent the optimization objectives in the current layer j The maximum and minimum values ​​of the objective function value, represents a positive number, represents the time variation factor, t represents the number of iterations, Represents particles i Its neighboring particles h The interaction strength between Represents the impact factor of the constraint condition.

[0013] The virtual power plant intelligent control system based on dynamic operation monitoring includes: data processing module, model solving module, system recording module, intelligent monitoring module and optimization module; The data processing module is used to collect data of distributed power sources, energy storage devices and loads in real time and perform preprocessing; The model solving module constructs a comprehensive optimization model of the distribution network according to the set multi-dimensional optimization objectives and decision variables, and solves to obtain the optimal energy efficiency improvement solution; The system recording module deploys the obtained optimal energy efficiency improvement solution to the distribution network, builds an integrated intelligent distribution network system, and records the operating status of the facilities; The intelligent monitoring module is used to monitor the energy efficiency of the distribution network in real time, predict the energy efficiency change trend, and dynamically adjust the optimal energy efficiency improvement plan based on the monitoring data and prediction results; The optimization module is used to collect user opinions and suggestions, optimize the performance of the smart distribution network system, and conduct in-depth mining of user electricity consumption behavior to optimize load-side management.

[0014] Specifically, the model solving module includes: a model building unit and a solving unit; The model building unit sets optimization objectives and decision variables according to the preprocessed data to build a comprehensive optimization model for the distribution network; The solving unit uses a fast non-dominated sorting multi-objective particle swarm algorithm with an elite strategy to solve the comprehensive optimization model of the distribution network and obtain an optimal solution set.

[0015] Specifically, the intelligent monitoring module includes: an intelligent monitoring unit, an energy efficiency prediction unit, and an optimization and adjustment unit; The intelligent monitoring unit uses IoT sensors and AI analysis algorithms to monitor the energy efficiency of the distribution network in real time; The energy efficiency prediction unit uses a machine learning algorithm to build an energy efficiency prediction model to predict the energy efficiency change trend; The optimization and adjustment unit dynamically adjusts the optimal energy efficiency improvement plan through an adaptive learning mechanism in combination with monitoring data and prediction results, and uses a neural network algorithm to regularly evaluate the energy efficiency improvement effect.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention proposes an intelligent control method for a virtual power plant based on dynamic operation monitoring. Through real-time data collection and intelligent analysis, it can accurately identify energy efficiency bottlenecks and optimization potential in the distribution network. Combining multi-dimensional optimization objectives and decision variables, the constructed comprehensive optimization model can find the optimal energy efficiency improvement plan, thereby improving the energy efficiency level of the distribution network; the Internet of Things technology is used to realize real-time data collection, and pre-processing is performed through edge computing and AI algorithms, so that the system can respond quickly to changes in the distribution network. At the same time, the intelligent monitoring system and adaptive learning mechanism can dynamically adjust the energy efficiency improvement plan according to real-time monitoring data and prediction results to ensure that the system always maintains the optimal operating state.

[0017] 2. The present invention proposes an intelligent control method for virtual power plants based on dynamic operation monitoring, which integrates multiple methods such as the Internet of Things, edge computing, AI algorithms, machine learning, and neural networks, and constructs an intelligent data collection, analysis, prediction, and optimization system, which not only reduces intervention and errors and improves work efficiency, but also can realize the automatic management and optimization of distribution networks; by integrating a sensor network of distributed power sources and energy storage devices, the method can monitor and optimize the power generation and energy storage processes of clean energy in real time, thereby improving the clean energy absorption capacity and the utilization efficiency of the energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a schematic diagram of the intelligent control method of a virtual power plant based on dynamic operation monitoring of the present invention; Figure 2 It is a principle flow chart of the virtual power plant intelligent control method based on dynamic operation monitoring of the present invention; Figure 3 It is a flowchart for realizing the optimal energy efficiency improvement solution of the virtual power plant intelligent control method based on dynamic operation monitoring of the present invention; Figure 4 A flow chart of congestion distance acquisition of the virtual power plant intelligent control method based on dynamic operation monitoring of the present invention; Figure 5 This is an architecture diagram of the virtual power plant intelligent control system based on dynamic operation monitoring of the present invention. DETAILED DESCRIPTION

[0019] Example 1

[0020] See also Figure 1-Figure 2 , an embodiment provided by the present invention: a virtual power plant intelligent control method based on dynamic operation monitoring, comprising the following steps: Step S1: Establish an intelligent data acquisition system based on the Internet of Things, integrate a sensor network of distributed power sources, energy storage equipment, and loads, collect data in real time, and pre-process the collected data through edge computing and AI algorithms. Based on the pre-processed data, set the multi-dimensional optimization objectives and decision variables of the distribution network, wherein the edge computing method is the prior art content in this field, is not the inventive solution of this application, and is not described in detail here, and the AI ​​algorithm adopts a supervised learning algorithm. At the same time, the supervised learning algorithm is the prior art content in this field, is not the inventive solution of this application, and is not described in detail here; Step S2: According to the set multi-dimensional optimization objectives and decision variables of the distribution network, a comprehensive optimization model of the distribution network is constructed, and the fast non-dominated sorting multi-objective particle swarm algorithm with elite strategy is used to solve the comprehensive optimization model of the distribution network, and the optimal energy efficiency improvement plan is obtained according to the optimal solution set; Step S3: Deploy the obtained optimal energy efficiency improvement solution to the distribution network for infrastructure construction, build an integrated smart distribution network system, and use blockchain technology to record the operating status of the facilities; Step S4: Establish an intelligent monitoring system, use IoT sensors and AI analysis algorithms to monitor the energy efficiency of the distribution network in real time, use machine learning algorithms to build an energy efficiency prediction model, predict the trend of energy efficiency changes, and dynamically adjust the optimal energy efficiency improvement plan through an adaptive learning mechanism, combining monitoring data and prediction results. Use a neural network algorithm to regularly evaluate the energy efficiency improvement effect, compare and analyze it with the set target, and adjust the optimal energy efficiency improvement plan based on the comparison results; Furthermore, the specific steps of step S4 include: (1) IoT sensor deployment and data collection: IoT sensors are deployed at key nodes of the distribution network, such as distributed power sources, energy storage devices, and load terminals, to collect various energy efficiency-related data in real time and transmit the collected data to the data processing center or edge computing device through the sensor network; (2) Data preprocessing and storage: Perform preprocessing operations such as cleaning, denoising, and filtering on the collected raw data to improve data quality, and store the preprocessed data in the database; (3) Application of AI analysis algorithms: Use supervised learning algorithms to conduct in-depth mining of stored data, identify key points and potential problems for energy efficiency improvement, and use supervised learning algorithms to perform pattern recognition and anomaly detection to provide decision support for energy efficiency optimization; (4) Energy efficiency prediction model construction: Use machine learning algorithms, such as time series prediction algorithms, to construct energy efficiency prediction models, and use historical data to train supervised learning algorithms so that they can accurately predict the energy efficiency change trend in the future. The specific construction process of the energy efficiency prediction model constructed by the time series prediction algorithm is the prior art content in this field, and is not the inventive solution of this application, and will not be elaborated here; (5) Adaptive learning mechanism and dynamic adjustment: Design an adaptive learning mechanism to dynamically adjust the optimal energy efficiency improvement plan based on real-time monitoring data and prediction results. If it is found that the energy efficiency index deviates from the preset target, the adjustment mechanism is automatically triggered to optimize distributed power generation scheduling, energy storage equipment charging and discharging strategy, and load management; (6) Neural network algorithm evaluation and comparative analysis: Use neural network algorithms to regularly evaluate the effect of energy efficiency improvement, and compare the evaluation results with the set energy efficiency targets to identify gaps and potential improvement points; (7) Optimization scheme adjustment and feedback: The optimal energy efficiency improvement scheme is adjusted according to the comparative analysis results, and the adjusted scheme is fed back to the distribution network control system to execute the new control strategy, forming a closed-loop optimization process.

[0021] Step S5: Establish a user feedback mechanism and use big data analysis methods to conduct in-depth mining of user electricity usage behaviors, identify electricity usage patterns and energy-saving potential, and optimize load-side management based on the identification results.

[0022] Example 2 See also Figure 3-Figure 4 In this embodiment, the specific steps of step S2 include: S2.1: Set multi-dimensional optimization goals and construct comprehensive target optimization functions At the same time, constraints are established, where n Indicates the number of categories of optimization objectives, Indicates n The objective function of the optimization goal is x represents the decision variable, express The weight of j≤n ; Among them, the multi-dimensional optimization objectives include: safe and stable operation of the distribution network, minimization of the comprehensive cost of distributed power sources, minimization of active network losses, and minimization of node voltage deviation. The target optimization function includes but is not limited to: comprehensive cost of distributed power sources, active network losses of distribution networks, and node voltage deviation.

[0023] S2.2: Set the parameters of the particle swarm algorithm and randomly initialize the position and velocity of the particles. The parameters of the particle swarm algorithm include: number of particles, number of iterations, learning factor, and inertia weight; S2.3: Calculate the objective function value of each particle according to its position, and apply fast non-dominated sorting to stratify the population and calculate the value of each particle. i The crowding distance .

[0024] S2.4: Update according to the speed and position update formula of the particle swarm algorithm to obtain the updated position and speed, and perform fast non-dominated sorting on the updated population according to S2.3 to evaluate the quality of the updated particles, wherein the speed and position update formula of the particle swarm algorithm is the prior art content in this field, is not the inventive solution of the present application, and is not described in detail here; S2.5: Merge the non-dominated solutions in the current population with the non-dominated solutions in the parent population to form an extended solution set, and then select the solution based on the fast non-dominated sorting and crowding distance. Select a new parent population to continue the next round of iteration; Furthermore, the specific steps of S2.5 include: (1) Merge populations: merge the non-dominated solutions in the current child population with the non-dominated solutions in the parent population to form an extended solution set; (2) Fast non-dominated sorting: Fast non-dominated sorting is performed on the merged extended solution set, mainly dividing the solution set into multiple frontiers, where each frontier contains a set of mutually non-dominated solutions. The first layer of frontiers contains the optimal solutions, that is, no other solutions can dominate these solutions, also called non-dominated solutions; (3) Calculate crowding distance: Calculate crowding distance for each solution in the frontier , the crowding distance is the distance between a solution and its adjacent solutions in the target space, which is used to measure the diversity of solutions. The larger the crowding distance, the farther the solution is from other solutions in the target space, which helps to maintain the diversity of the population; (4) Selecting a new parent population: Based on the results of the fast non-dominated sort and the crowding distance, a new parent population is selected from the expanded solution set. Usually, all solutions in the first layer of the frontier are selected first, and then the solutions in the subsequent frontiers are selected in turn until the required population size is reached. When selecting solutions in the subsequent frontiers, solutions with a larger crowding distance are given priority to ensure the diversity of the population.

[0025] Furthermore, the iteration process includes: (1) Update particle speed and position: Update according to the speed and position update formula of the particle swarm algorithm; (2) Crossover and mutation: The crossover and mutation operations of the genetic algorithm are introduced into the particle swarm algorithm to increase the diversity of solutions; (3) Fast non-dominated sorting: fast non-dominated sorting of the updated population; (4) Elite retention strategy: merge the non-dominated solutions in the current population, i.e., the optimal solutions, with the non-dominated solutions in the parent population, and select a new parent population based on fast non-dominated sorting and crowding distance.

[0026] S2.6: Save the set of non-dominated solutions found through an external archive, and update the external archive in each iteration, remove old solutions dominated by new solutions, and add new non-dominated solutions; S2.7: Check whether the maximum number of iterations has been reached. If so, stop the iteration process and output the optimal solution set in the external archive.

[0027] The specific formula of the constraint condition in S2.1 is: ,in, Represents equality constraints, including power balance, energy balance, represents inequality constraints, including voltage limits, line capacity limits, p represents the number of equality constraints, q represents the number of inequality constraints, represents the minimum function, , .

[0028] The specific steps of S2.3 include: S2.31: Calculate each particle i The objective function value of ,in, Indicates n The particle in the optimization target i The objective function of S2.32: For each particle i , set the dominance i The number of particles and quilt i Dominated particle set , and traverse all particle pairs in the population, updating each particle and , among which, if , it is represented as a non-dominated particle and is also a non-dominated solution in the current population.

[0029] S2.33: All The particles are placed in the first layer F and the particles are updated Each particle in , repeat this process until all particles are delaminated; S2.34: Initialize each particle i The crowding distance makes , for each optimization objective, Sort by, Represents particles k The objective function value at the same layer.

[0030] S2.35: For the sorted boundary particles, set their crowding distance to ; For the sorted non-boundary particles, the formula Calculate its crowding distance, where represents the accumulated crowding distance, Represents the scaling factor, which is used to adjust the differential part to The scaling factor needs to be adjusted according to the specific needs of the problem to balance the differences between different targets or particles. and Respectively represent the optimization objectives j On the particle i The objective function value of the adjacent particles before and after is, and Respectively represent the optimization objectives in the current layer j The maximum and minimum values ​​of the objective function value, Represents a positive number, which is used to prevent the denominator from being zero and increase the numerical stability of the formula. It represents the time variation factor, which is a function of the number of iterations. This function is designed according to the dynamics of the problem, such as linear, exponential, or periodic. t represents the number of iterations, Represents particles i Its neighboring particles h The interaction strength between particles is calculated based on factors such as the distance, direction, and speed difference between particles to reflect the relative position relationship of particles in space. Indicates the impact factor of the constraint condition, which is used to introduce limiting conditions or constraints. The impact factor can be a Boolean value, with a value of 0 or 1, indicating whether a specific condition is met. b , or a continuous value, indicating the condition b The degree of satisfaction or the intensity of influence, represents the normalization factor, reflecting the particle i Its neighboring particles are on the target j The relative difference in .

[0031] Example 3

[0032] See also Figure 5 , an embodiment provided by the present invention: a virtual power plant intelligent control system based on dynamic operation monitoring, comprising: Data processing module, model solving module, system recording module, intelligent monitoring module, and optimization module; Data processing module, used to collect data of distributed power sources, energy storage devices and loads in real time and perform pre-processing to provide a basis for subsequent optimization decisions; The model solving module builds a comprehensive optimization model for the distribution network based on the set multi-dimensional optimization objectives and decision variables, and solves to obtain the optimal energy efficiency improvement solution; The system recording module deploys the optimal energy efficiency improvement solution to the distribution network, builds an integrated smart distribution network system, and records the operating status of the facilities; Intelligent monitoring module, which is used to monitor the energy efficiency of the distribution network in real time, predict the trend of energy efficiency changes, and dynamically adjust the optimal energy efficiency improvement plan based on monitoring data and prediction results; The optimization module is used to collect user opinions and suggestions, optimize the performance of the smart distribution network system, conduct in-depth mining of user electricity consumption behavior, and optimize load-side management.

[0033] The model solving module includes: a model building unit and a solving unit; The model building unit sets the optimization objectives and decision variables according to the preprocessed data and builds a comprehensive optimization model for the distribution network; The solving unit uses a fast non-dominated sorting multi-objective particle swarm algorithm with an elite strategy to solve the comprehensive optimization model of the distribution network and obtain the optimal solution set.

[0034] The system recording module includes: a system construction unit and an operation status recording unit; System construction unit, which converts the optimal energy efficiency improvement plan into actual distribution network infrastructure construction; The operation status recording unit uses blockchain technology to record the operation status of the facility to ensure the transparency and non-tamperability of the data.

[0035] The intelligent monitoring module includes: intelligent monitoring unit, energy efficiency prediction unit, and optimization adjustment unit; Intelligent monitoring units, which use IoT sensors and AI analysis algorithms to monitor the energy efficiency of the distribution network in real time; Energy efficiency prediction unit, which uses machine learning algorithms to build energy efficiency prediction models and predict energy efficiency change trends; The optimization and adjustment unit dynamically adjusts the optimal energy efficiency improvement plan through an adaptive learning mechanism, combining monitoring data and prediction results, and uses a neural network algorithm to regularly evaluate the energy efficiency improvement effect.

[0036] The optimization module includes: user feedback unit, system performance optimization unit, and management optimization unit; User feedback unit, used to establish a user feedback mechanism and collect user opinions and suggestions; System performance optimization unit, which optimizes system performance based on user feedback; The management optimization unit uses big data analysis methods to conduct in-depth mining of users' electricity usage behaviors, identify electricity usage patterns and energy-saving potential, and optimize load-side management based on the identification results.

[0037] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation modes, which are merely illustrative rather than restrictive. Under the guidance of the present invention, ordinary technicians in the field may also change, modify, replace and modify the above-mentioned embodiments without departing from the purpose and scope of protection of the present invention, and all of these are within the protection of the present invention.

Claims

1. A virtual power plant intelligent control method based on dynamic operation monitoring, characterized in that: include: Step S1: Establish an intelligent data acquisition system based on the Internet of Things, integrate the sensor network of distributed power sources, energy storage equipment, and loads, collect data in real time, and pre-process the collected data through edge computing and AI algorithms. Based on the pre-processed data, set the multi-dimensional optimization goals and decision variables of the distribution network; Step S2: According to the set multi-dimensional optimization objectives and decision variables of the distribution network, a comprehensive optimization model of the distribution network is constructed, and the fast non-dominated sorting multi-objective particle swarm algorithm with elite strategy is used to solve the comprehensive optimization model of the distribution network, and the optimal energy efficiency improvement plan is obtained according to the optimal solution set; Step S3: Deploy the obtained optimal energy efficiency improvement solution to the distribution network for infrastructure construction, build an integrated smart distribution network system, and use blockchain technology to record the operating status of the facilities; Step S4: Establish an intelligent monitoring system, use IoT sensors and AI analysis algorithms to monitor the energy efficiency of the distribution network in real time, use machine learning algorithms to build an energy efficiency prediction model, predict the trend of energy efficiency changes, and dynamically adjust the optimal energy efficiency improvement plan through an adaptive learning mechanism, combining monitoring data and prediction results. Use a neural network algorithm to regularly evaluate the energy efficiency improvement effect, compare and analyze it with the set target, and adjust the optimal energy efficiency improvement plan based on the comparison results; Step S5: Establish a user feedback mechanism and use big data analysis methods to conduct in-depth mining of user electricity usage behaviors, identify electricity usage patterns and energy-saving potential, and optimize load-side management based on the identification results.

2. The method for intelligent control of a virtual power plant based on dynamic operation monitoring according to claim 1, characterized in that: The specific steps of step S2 include: S2.1: Set multi-dimensional optimization goals and construct comprehensive target optimization functions At the same time, establish constraints, x represents the decision variable; S2.2: Set the parameters of the particle swarm algorithm and randomly initialize the position and velocity of the particles; S2.3: Calculate the objective function value of each particle according to its position, and apply fast non-dominated sorting to stratify the population and calculate the value of each particle. i The crowding distance .

3. The method for intelligent control of a virtual power plant based on dynamic operation monitoring according to claim 2, characterized in that: The specific steps of step S2 also include: S2.4: Update according to the speed and position update formula of the particle swarm algorithm to obtain the updated position and speed, and perform fast non-dominated sorting on the updated population according to S2.3 to evaluate the quality of the updated particles; S2.5: Merge the non-dominated solutions in the current population with the non-dominated solutions in the parent population to form an extended solution set, and then select the solution based on the fast non-dominated sorting and crowding distance. Select a new parent population to continue the next round of iteration; S2.6: Save the set of non-dominated solutions found through an external archive, and update the external archive in each iteration, remove old solutions dominated by new solutions, and add new non-dominated solutions; S2.7: Check whether the maximum number of iterations has been reached. If so, stop the iteration process and output the optimal solution set in the external archive.

4. The method for intelligent control of a virtual power plant based on dynamic operation monitoring according to claim 3 is characterized in that: The specific formula of the constraint condition in S2.1 is: ,in, represents the equality constraint, represents the inequality constraint, p represents the number of equality constraints, q represents the number of inequality constraints, represents the minimum function, , .

5. The method for intelligent control of virtual power plants based on dynamic operation monitoring according to claim 4, characterized in that: The specific steps of S2.3 include: S2.31: Calculate each particle i The objective function value of ,in, Indicates n The particle in the optimization target i The objective function of S2.32: For each particle i , set the control i The number of particles and quilt i Dominated particle set , and traverse all particle pairs in the population, updating each particle and , among which, if , it is represented as a non-dominated particle and is also a non-dominated solution in the current population.

6. The method for intelligent control of a virtual power plant based on dynamic operation monitoring according to claim 5, characterized in that: The specific steps of S2.3 also include: S2.33: All The particles are placed in the first layer F and the particles are updated Each particle in , repeat this process until all particles are delaminated; S2.34: Initialize each particle i The crowding distance makes , for each optimization objective, Sort by, Represents particles k The objective function value at the same layer.

7. The method for intelligent control of a virtual power plant based on dynamic operation monitoring according to claim 6, characterized in that: The specific steps of S2.3 also include: S2.35: For the sorted boundary particles, set their crowding distance to ; For the sorted non-boundary particles, the formula Calculate its crowding distance, where represents the accumulated crowding distance, represents the scaling factor, and Respectively represent the optimization objectives j On the particle i The objective function value of the adjacent particles before and after is, and Respectively represent the optimization objectives in the current layer j The maximum and minimum values ​​of the objective function value, represents a positive number, represents the time variation factor, t represents the number of iterations, Represents particles i Its neighboring particles h The interaction strength between Represents the impact factor of the constraint condition.

8. A virtual power plant intelligent control system based on dynamic operation monitoring, which is used to implement the virtual power plant intelligent control method based on dynamic operation monitoring according to any one of claims 1 to 7, characterized in that: include: Data processing module, model solving module, system recording module, intelligent monitoring module, and optimization module; The data processing module is used to collect data of distributed power sources, energy storage devices and loads in real time and perform preprocessing; The model solving module constructs a comprehensive optimization model of the distribution network according to the set multi-dimensional optimization objectives and decision variables, and solves to obtain the optimal energy efficiency improvement solution; The system recording module deploys the obtained optimal energy efficiency improvement solution to the distribution network, builds an integrated intelligent distribution network system, and records the operating status of the facilities; The intelligent monitoring module is used to monitor the energy efficiency of the distribution network in real time, predict the energy efficiency change trend, and dynamically adjust the optimal energy efficiency improvement plan based on the monitoring data and prediction results; The optimization module is used to collect user opinions and suggestions, optimize the performance of the smart distribution network system, and conduct in-depth mining of user electricity consumption behavior to optimize load-side management.

9. The virtual power plant intelligent control system based on dynamic operation monitoring according to claim 8 is characterized in that: The model solving module includes: a model building unit and a solving unit; The model building unit sets optimization objectives and decision variables according to the preprocessed data to build a comprehensive optimization model for the distribution network; The solving unit uses a fast non-dominated sorting multi-objective particle swarm algorithm with an elite strategy to solve the comprehensive optimization model of the distribution network and obtain an optimal solution set.

10. The virtual power plant intelligent control system based on dynamic operation monitoring according to claim 9, characterized in that: The intelligent monitoring module includes: an intelligent monitoring unit, an energy efficiency prediction unit, and an optimization and adjustment unit; The intelligent monitoring unit uses IoT sensors and AI analysis algorithms to monitor the energy efficiency of the distribution network in real time; The energy efficiency prediction unit uses a machine learning algorithm to build an energy efficiency prediction model to predict the energy efficiency change trend; The optimization and adjustment unit dynamically adjusts the optimal energy efficiency improvement plan through an adaptive learning mechanism in combination with monitoring data and prediction results, and uses a neural network algorithm to regularly evaluate the energy efficiency improvement effect.

Images