A simulation method for the energy efficiency of the client load

Abstract

The present invention relates to the technical field of microgrid energy management, and particularly to a simulation method for the energy efficiency of user-side loads, including: obtaining historical data of the user side, constructing a demand trend model based on a recurrent neural network to predict the electricity demand in a future time period; after obtaining the predicted value of the electricity demand in a future time period, obtaining the energy flow route, and obtaining the energy supply bottleneck points and high-energy consumption points according to the energy flow route; obtaining the data of the bottleneck points and high-energy consumption points, introducing an affiliated dimension, and predicting the power supply demand and electricity demand of the microgrid system; constructing a digital virtual model of the energy supply system to simulate and verify the electricity consumption characteristics and energy flow process of the user side under different electricity consumption scenarios, and verifying the reliability of the demand trend model and the multi-dimensional analysis model; classifying the user side into different electricity consumption type groups through a clustering analysis algorithm, and formulating corresponding energy supply plans for each electricity consumption type group to improve the energy supply efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrid energy management, and particularly to a simulation method for the energy efficiency of user-side loads. Background Art

[0002] In the context of the continuous development and transformation of the current energy field, energy efficiency management has become an important topic that has attracted much attention. However, in the actual process of energy efficiency research and application, there are many problems that need to be solved urgently. Among them, the abstraction problem of the load energy efficiency data of each power consumption node at the user side is particularly prominent. Since energy efficiency data often has the characteristics of complexity and diversity, and it involves many dimensions and factors, it is very difficult to accurately and effectively abstract the energy efficiency data. It is difficult to grasp the overall energy usage situation and efficiency level.

[0003] Therefore, we propose a method to simulate and analyze the load energy consumption of the power consumption nodes at the user side to analyze the energy efficiency of the user side. Summary of the Invention

[0004] The present invention simulates and analyzes the power consumption characteristics and energy flow process of the user side under different power consumption scenarios by constructing a digital virtual model of the energy supply system, evaluates the carbon emission energy efficiency of the user side according to the predicted values of the trend model and the multi-dimensional analysis model, and formulates corresponding energy supply plans for each power consumption type group to improve the energy supply efficiency.

[0005] The technical solution proposed by the present invention is: a simulation method for the energy efficiency of user-side loads, the method comprising:

[0006] Obtain the historical data of the user side, construct a demand trend model based on the recurrent neural network, capture the dynamic changes and periodic characteristics of the power consumption data of the user side, and predict the power consumption demand within a future time period;

[0007] After obtaining the predicted value of the power consumption demand within a future time period, analyze the energy supply demand among multiple nodes of the user side, obtain the energy flow route, and obtain the energy supply bottleneck points and high energy consumption points according to the energy flow route;

[0008] Obtain the data of the bottleneck points and high energy consumption points, introduce additional dimensions, construct a multi-dimensional analysis model, and predict the power supply demand and power consumption demand of the microgrid system through the multi-dimensional analysis model;

[0009] Construct a digital virtual model of the energy supply system, simulate and analyze the power consumption characteristics and energy flow process of the user side under different power consumption scenarios, and verify the reliability of the demand trend model and the multi-dimensional analysis model;

[0010] Establish an efficiency evaluation model, and evaluate the carbon emission energy efficiency of the user side according to the predicted values of the trend model and the multi-dimensional analysis model;

[0011] According to the evaluation results of the previous step, the user terminals are divided into different power consumption type groups through a clustering analysis algorithm, and corresponding energy supply plans are formulated for each power consumption type group to improve the energy supply efficiency.

[0012] Preferably, the method for obtaining the historical data of the user terminal, constructing a demand trend model based on a recurrent neural network, capturing the dynamic changes and periodic characteristics of the power consumption data of the user terminal, and predicting the power consumption demand within a future time period includes:

[0013] Obtain the historical power consumption data of the user terminal within a preset time period, where the historical power consumption data includes power supply amount, power consumption amount, and power consumption timestamp;

[0014] Preprocess the obtained historical power consumption data to form a historical power consumption data set, and extract multiple power consumption characteristics from the historical power consumption data set to form a power consumption characteristic sample set;

[0015] Use the TensorFlow or PyTorch deep learning framework to construct a demand trend model;

[0016] Divide the power consumption characteristic sample set into a training set and a test set, train the demand trend model through the training set, and verify the demand trend model through the test set; then, deploy the trained demand trend model to the system;

[0017] Collect the power consumption data of multiple power consumption nodes of the user terminal within a preset time length at a preset frequency, and extract multiple power consumption data characteristics to form a power consumption data time series; the power consumption data characteristics include maximum power consumption, minimum power consumption, maximum power supply amount, minimum power supply amount, maximum power supply time, minimum power supply time, maximum power consumption time, and minimum power consumption time;

[0018] After normalizing the power consumption data time series, input it into the demand trend model, and output the predicted values of the maximum power supply amount, maximum power supply time, maximum power consumption, and maximum power consumption time of a power consumption node within a future time period.

[0019] Preferably, after obtaining the predicted values of the power consumption demand within a future time period, analyze the energy supply demands among multiple nodes of the user terminal, obtain the energy flow route, and obtain the energy supply bottleneck points and high energy consumption points according to the energy flow route, including:

[0020] Obtain the predicted values of the power consumption demands of each node within a future time period from the power grid management system;

[0021] According to the obtained predicted values, establish a linear programming model, with the goal of minimizing the energy cost and / or maximizing the energy efficiency, and analyze the energy supply demands of each node;

[0022] Using graph theory algorithms, determine the optimal energy flow route from the energy supply source to each power consumption node according to the capacity limit and transmission loss of the line;

[0023] By analyzing the energy flow route, identify the bottleneck points in the system by calculating the flow and pressure at each node, that is, those bottleneck points and high energy consumption points that limit the overall system performance.

[0024] Preferably, according to the obtained prediction values, establish a linear programming model, with the goal of minimizing energy cost and / or maximizing energy efficiency, and analyze the energy supply demand of each node, including:

[0025] Use to represent the energy demand of the power consumption node ;

[0026] If the goal is to minimize the energy cost, the objective function is: ; where represents the unit energy cost of the th power consumption node, represents the total number of power consumption nodes at the user end;

[0027] If the goal is to maximize the energy efficiency, the objective function is: ; where represents the unit energy efficiency of the th power consumption node;

[0028] Add constraint one: , where represents the maximum supply capacity of the power consumption node ;

[0029] Add constraint two: , where represents the total energy supply at the user end;

[0030] Solve the objective function by the simplex method or the interior point method;

[0031] According to the solution results, determine the optimal energy demand at each stage and calculate the corresponding cost or efficiency.

[0032] Preferably, the use of graph theory algorithms to determine the optimal energy flow route from the energy supply source to each power consumption node according to the capacity limit and transmission loss of the line includes:

[0033] Let there be a directed graph G=(V, E); where V is the set of vertices, and each vertex represents an energy supply source and a power consumption node; E is the set of edges, and each edge e=(u, v) has a capacity limit c(e) and a transmission loss w(e);

[0034] Create a distance array D to store the shortest distances from the source node to other nodes. Initially, the distances of all nodes are set to infinity, and the distance of the source node is set to 0;

[0035] Create a priority queue P to store the nodes to be processed and their current distances;

[0036] Add the source node to the priority queue with a distance of 0;

[0037] When the priority queue is not empty, extract the node u with the minimum distance;

[0038] For each adjacent node v of node u, calculate the distance new_D to reach v through u as new_D = D[u] + w(u,v); where w(u, v) represents the weight of the edge from node u to node v;

[0039] If new_D is less than D[v], then update D[v] and add v to the priority queue;

[0040] When the priority queue is empty, end the calculation; at this time, the D array stores the shortest distances from the source node to each node;

[0041] Thus, obtain the optimal energy flow routes from the energy supply source to each power consumption node, minimizing the total transmission loss.

[0042] Preferably, by analyzing the energy flow routes, identify the bottleneck points in the system, i.e., those points that limit the overall system performance and high - energy - consumption points, by calculating the flow and pressure of each node, including:

[0043] Obtain the energy flow lines;

[0044] Calculate the maximum flow from the source node to the sink node through the Ford - Fulkerson algorithm or the Edmonds - Karp algorithm; this process is achieved by continuously finding augmenting paths and updating the residual graph;

[0045] After calculating the maximum flow, check the utilization rate of each edge, i.e., the ratio of the actual flow to the capacity; the edges with a utilization rate of 100% ± 2% are the bottleneck points;

[0046] Calculate the difference between the total input and output flows of each power consumption node, and the nodes with a difference greater than the preset threshold are the high - energy - consumption points.

[0047] Preferably, for obtaining the data of bottleneck points and high - energy - consumption points, introduce additional dimensions to construct a multi - dimensional analysis model, and predict the power supply demand and power consumption demand of the micro - grid system through the multi - dimensional analysis model, including:

[0048] Obtain the data of high - energy - consumption points and bottleneck points to form a two - dimensional energy - consumption data set;

[0049] Obtain the ambient temperature, weather events, and maintenance events at the power - consuming nodes, and after quantifying the ambient temperature, weather events, and maintenance events, form an affiliated dimension data set;

[0050] After normalizing the two - dimensional energy - consumption data set and the affiliated dimension data set, combine them into a multi - dimensional data set;

[0051] Construct a multi - dimensional analysis model through the random forest algorithm, using the multi - dimensional data set as the input variable to predict the power supply demand of high - energy - consumption points and bottleneck points in the next time period. The power supply demand includes the maximum power supply, the maximum power supply demand time, the minimum power supply, and the minimum power supply demand time.

[0052] Preferably, construct a digital virtual model of the energy supply system, simulate and simulate the power - consumption characteristics and energy - flow process of the user side under different power - consumption scenarios, and verify the reliability of the demand trend model and the multi - dimensional analysis model, including:

[0053] Obtain the power - consumption characteristic data of the user side. The power - consumption characteristic data includes equipment type, power consumption, and usage time.

[0054] Obtain the power - consumption demand in the next time period predicted by the demand trend model and the multi - dimensional analysis model;

[0055] According to the power - grid structure, power - generation capacity, and power - consumption demand, simulate the energy - flow process from the power - generation station to the user side; specifically: use the discrete - event simulation or system - dynamics simulation method to simulate the energy - flow of the entire energy supply system.

[0056] Preferably, establish an energy - efficiency evaluation model, and evaluate the carbon - emission energy - efficiency of the user side according to the predicted values of the trend model and the multi - dimensional analysis model, including:

[0057] Obtain the predicted power consumption of the power - consuming nodes in the user side. The power - consuming nodes include bottleneck points, high - energy - consumption points, non - bottleneck points, and non - high - energy - consumption points;

[0058] Import the carbon - emission coefficient, calculate and obtain the carbon emissions of each power - consuming node, \(C = W\times p\) i , where \(p\) i represents the carbon - emission coefficient of the \(i\) - th power - consuming node;

[0059] Compare the carbon - emission value of the actual power consumption with the carbon - emission value of the predicted power consumption, and judge the carbon - emission energy - efficiency according to the difference; that is, the smaller the difference, the higher the carbon - emission energy - efficiency; on the contrary, the lower the carbon - emission energy - efficiency;

[0060] According to the evaluation results of the previous step, the user terminals are divided into different power consumption type groups through a clustering analysis algorithm, and corresponding energy supply plans are formulated for each power consumption type group to improve the energy supply efficiency, including:

[0061] Obtain the power consumption data time series, the power consumption data at bottleneck points and high energy consumption points, and form a classification data set;

[0062] Analyze the classification data set through a clustering algorithm, classify the user terminals, and divide the user terminals into multiple power consumption groups; the power consumption groups include high carbon emission efficiency power consumption groups, low carbon emission efficiency power consumption groups, and abnormal carbon emission efficiency power consumption groups;

[0063] Guide corresponding energy supply plans for different power consumption groups, and the energy supply plans include adjusting the equipment usage time and improving the equipment usage efficiency.

[0064] The present invention also provides a simulation model, and the simulation model includes one or more of a demand trend model, a linear programming model, a multi-dimensional analysis model, a digital virtual model of an energy supply system, and an efficiency evaluation model.

[0065] Advantages of the present invention:

[0066] In the present invention, the power consumption demand of the user terminals in a future time period is predicted through a demand trend model, the bottleneck points and high energy consumption points are obtained by analyzing the energy flow route, and the model is simulated in the digital virtual model to verify the effectiveness of the prediction. At the same time, taking the carbon emission energy efficiency as the evaluation basis, the energy efficiency of the user terminals is evaluated, the user terminals are divided into multiple type groups according to the evaluation, and different energy supply plans are provided for different type groups to improve the load energy efficiency of the user terminals. Description of the Drawings

[0067] Figure 1 It is a flowchart of a simulation method for the load energy efficiency of a user terminal according to the present invention. Detailed Embodiments

[0068] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious deformations. The basic principles defined in the following description can be applied to other implementation schemes, deformation schemes, improvement schemes, equivalent schemes, and other technical schemes that do not depart from the spirit and scope of the present invention.

[0069] It can be understood that the term "one" should be understood as "at least one" or "one or more". That is, in one embodiment, the number of an element can be one, and in other embodiments, the number of the element can be multiple. The term "one" cannot be understood as a limitation on the number.

[0070] Reference Figure 1 , the technical solution provided by the present invention is: a simulation method for the energy efficiency of the user - side load, including the following steps:

[0071] Step 1: Collect and organize the historical electricity consumption data of the user - side, including key parameters such as electricity quantity and voltage. Pre - process the data, such as cleaning, normalization, etc., for subsequent analysis.

[0072] Step 2: Based on the recurrent neural network (RNN), especially the long short - term memory network (LSTM), construct a demand trend model. Train the model to capture the dynamic changes and periodic characteristics of the electricity consumption data, such as daily, weekly, or seasonal changes. Use the model to predict the electricity demand within a future time period.

[0073] Step 3: According to the predicted electricity demand, analyze the energy supply demand among multiple nodes of the user - side. Determine the energy flow route, identify the energy supply bottleneck points and high - energy - consumption points.

[0074] Step 4: Obtain the data of the bottleneck points and high - energy - consumption points, and introduce auxiliary dimensions (such as time, temperature, humidity, etc.) to enrich the data set. Construct a multi - dimensional analysis model, considering the influence of multiple factors on the power supply demand. Predict the power supply demand and electricity consumption demand of the micro - grid system through the model.

[0075] Step 5: Establish a digital virtual model of the energy supply system, including components such as power sources, power grids, and electrical equipment. Simulate the electricity consumption characteristics and energy flow process of the user - side under different electricity consumption scenarios. Verify the reliability of the demand trend model and the multi - dimensional analysis model.

[0076] Step 6: According to the predicted values of the trend model and the multi - dimensional analysis model, evaluate the carbon emission energy efficiency of the user - side. Consider different electricity consumption scenarios and energy efficiency standards, and formulate appropriate evaluation indicators for each scenario.

[0077] Step 7: According to the effectiveness evaluation results, divide the user - side into different electricity consumption type groups through the clustering analysis algorithm. Formulate corresponding energy supply plans for each electricity consumption type group to improve the energy utilization efficiency and energy supply efficiency.

[0078] In this embodiment, step 2 specifically includes the following steps: Obtain the historical electricity consumption data of the user - side within a preset time period, and the historical electricity consumption data includes power supply quantity, electricity consumption quantity, and electricity consumption timestamp;

[0079] Pre - process the obtained historical electricity consumption data to form a historical electricity consumption data set, and extract multiple electricity consumption characteristics from the historical electricity consumption data set to form an electricity consumption characteristic sample set;

[0080] Use the TensorFlow or PyTorch deep learning framework to construct a demand trend model;

[0081] Divide the power consumption feature sample set into a training set and a test set, train the demand trend model through the training set, and verify the demand trend model through the test set; then, deploy the trained demand trend model to the system.

[0082] Collect the power consumption data of multiple power consumption nodes at the user end within a preset time length at a preset frequency, and extract multiple power consumption data features to form a power consumption data time series; the power consumption data features include the maximum power consumption, the minimum power consumption, the maximum power supply, the minimum power supply, the maximum power supply time, the minimum power supply time, the maximum power consumption time, and the minimum power consumption time.

[0083] After normalizing the power consumption data time series, input it into the demand trend model, and output the predicted values of the maximum power supply, the maximum power supply time, the maximum power consumption, and the maximum power consumption time within a future time period for one power consumption node.

[0084] In this embodiment, step 3 specifically includes the following steps:

[0085] Connect to the database of the power grid management system and extract the predicted power consumption demand data of each node within the future time period.

[0086] According to the predicted demand value, construct a linear programming model, and set the objective function to minimize the energy cost or maximize the energy efficiency. Specifically:

[0087] Use to represent the energy demand of the power consumption node . If the goal is to minimize the energy cost, the objective function is: ; where represents the unit energy cost of the th power consumption node, and represents the total number of power consumption nodes at the user end; if the goal is to maximize the energy efficiency, the objective function is: ; where represents the unit energy efficiency of the th power consumption node;

[0088] Add constraint condition one: , where represents the maximum supply capacity of the power consumption node ; add constraint condition two: , where represents the total energy supply at the user end;

[0089] Solve the objective function by the simplex method and the interior point method; according to the solution results, determine the optimal energy demand for each stage, and calculate and obtain the corresponding cost or efficiency. Consider the supply-demand balance of each node to ensure that the decision variables (such as the supply volume of each node) meet the demand constraint conditions.

[0090] Using graph theory algorithms (such as Dijkstra's algorithm, Bellman-Ford algorithm, etc.), calculate the shortest path or the lowest-cost path from the energy supply source to each power-consuming node according to the capacity limit and transmission loss of the line, as the optimal energy flow route. Specifically: Let the directed graph G=(V, E); where, V is the set of vertices, and each vertex represents an energy supply source and a power-consuming node; E is the set of edges, and each edge e=(u, v) has a capacity limit c(e) and a transmission loss w(e);

[0091] Create a distance array D to store the shortest distances from the source node to other nodes. Initially, the distances of all nodes are infinite, and the distance of the source node is 0;

[0092] Create a priority queue P to store the nodes to be processed and their current distances;

[0093] Add the source node to the priority queue with a distance of 0;

[0094] When the priority queue is not empty, take out the node u with the smallest distance;

[0095] For each adjacent node v of node u, calculate the distance new_D = D[u] + w(u,v) to reach v through u; where, w(u, v) represents the weight of the edge from node u to node v;

[0096] If new_D is less than D[v], then update D[v] and add v to the priority queue;

[0097] When the priority queue is empty, end the calculation; at this time, the D array stores the shortest distances from the source node to each node;

[0098] Thus, obtain the optimal energy flow route from the energy supply source to each power-consuming node, minimizing the total transmission loss.

[0099] According to the obtained optimal energy flow route, calculate the flow rate (i.e., the energy passing through this node) and pressure (i.e., the ratio of the energy processed by the node to its maximum processing capacity) of each node. Specifically: Obtain the energy flow line;

[0100] Calculate the maximum flow from the source node to the sink node through the Ford-Fulkerson algorithm or the Edmonds-Karp algorithm; this process is achieved by continuously finding augmenting paths and updating the residual graph;

[0101] After calculating the maximum flow, check the utilization rate of each edge, i.e., the ratio of the actual flow to the capacity; the edge with a utilization rate of 100% ± 2% is the bottleneck point.

[0102] Calculate the difference between the total input and output flows of each power consumption node, and the node with a difference greater than the preset threshold is the high - energy - consumption point.

[0103] Identify the nodes where the flow is close to or reaches the line capacity limit, and the nodes where the pressure is greater than the preset threshold. These are the bottleneck points and high - energy - consumption points in the system.

[0104] In this embodiment, step 4 includes the following steps:

[0105] Obtain and merge multiple data sources, including the energy consumption data of high - energy - consumption points and bottleneck points, the environmental temperature, the quantified data of weather events and maintenance events. The merged dataset contains the energy consumption, temperature, weather events, and maintenance event information of each point.

[0106] Normalize the data to eliminate the dimensional differences between different features and ensure the stability and accuracy of model training. Use MinMaxScaler to scale the data to the interval [0, 1].

[0107] Select feature variables (temperature, weather events, maintenance events) from the pre - processed data as the input of the model. Define the target variable as the total energy consumption of each point, i.e., the sum of the energy consumption of high - energy - consumption points and bottleneck points.

[0108] Use the random forest regression model (RandomForestRegressor) to predict the power supply demand. Set the parameters of the random forest, such as the number of trees (n_estimators) and the random seed (random_state), to improve the generalization ability and reproducibility of the model. Train the model so that it can predict the target variable based on the feature variables.

[0109] Use the trained model to predict the power supply demand for a future time period. Analyze the prediction results to find the maximum and minimum power supply demands and their corresponding time points.

[0110] In this embodiment, step 5 includes the following steps:

[0111] Collect data on equipment type, power consumption, and usage time from the power system database or user surveys. Use the previously constructed demand trend model and multi - dimensional analysis model to predict the power consumption of different user terminals in a future time period.

[0112] Build a discrete event model of the power grid, including components such as power stations, transformers, and transmission lines. Simulate the changes in the operating states of each component, such as the adjustment of the power generation power of the power station and the load change of the transformer. Consider the real-time demand at the user end and dynamically allocate electric energy. Output the simulation results, including information such as voltage, current, and power at each node.

[0113] Establish a dynamic model of the power grid energy supply system to describe the interaction and energy conversion relationship between components. Set model parameters, such as power generation capacity, line capacity, electricity demand, etc. Simulate the dynamic behavior of the system under different operating conditions and analyze the stability and efficiency of energy flow. Optimize the power grid configuration and dispatching strategy according to the simulation results.

[0114] To simulate the dynamic behavior of the system under different operating conditions, analyze the stability and efficiency of energy flow, and optimize the power grid configuration and dispatching strategy according to the simulation results, we need to build a simulation model. This model can be realized based on the basic theory and mathematical modeling methods of the power system. The simulation model here includes a demand trend model, a linear programming model, a multi-dimensional analysis model, a digital virtual model of the energy supply system, and an effectiveness evaluation model.

[0115] Specifically:

[0116] First, define system parameters: including parameters of generators, loads, transmission lines, etc.

[0117] Establish a mathematical model of the system based on the basic equations of the power system (such as the power balance equation). Simulate different operating conditions, such as different load levels, generator outputs, etc. Use numerical methods (such as the Newton-Raphson method) to solve the steady-state or dynamic equations of the system. Calculate the stability indicators (such as damping ratio, frequency deviation) and efficiency indicators (such as losses, costs) of the system. Adjust the power grid configuration and dispatching strategy according to the analysis results, such as reallocating loads, adjusting generator outputs, etc. Repeat the above steps until the optimal power grid configuration and dispatching strategy are found.

[0118] In the complex field of power system operation and management, in order to more accurately grasp the energy usage and carbon emission-related information of each electricity consumption node within the user end, it is necessary to obtain the predicted electricity consumption of the electricity consumption nodes within the user end. The electricity consumption nodes here cover various types, specifically including bottleneck points, high-energy consumption points, as well as non-bottleneck points and non-high-energy consumption points.

[0119] Bottleneck points usually refer to the links in the process of power transmission or distribution where power supply shortages are likely to occur due to factors such as equipment capacity and line carrying capacity. The operating conditions of these points have a crucial impact on the stable and efficient operation of the entire power system. High-energy consumption points refer to those areas or equipment where energy consumption during production exceeds a preset threshold. For example, in some specific production processes of large industrial enterprises, their electricity consumption often accounts for more than 80% of the total electricity consumption of the enterprise. In contrast, non-bottleneck points and non-high-energy consumption points are relatively stable in terms of power supply and energy consumption, but they are also an indispensable part of the power system.

[0120] After obtaining the predicted electricity consumption of the power consumption nodes, it is necessary to import the carbon emission coefficient to calculate the carbon emissions of each power consumption node. This calculation process follows a specific formula, that is, C = W * pi. Where W represents the predicted electricity consumption of this power consumption node, and pi represents the carbon emission coefficient of the i-th power consumption node. The carbon emission coefficient is a key parameter that reflects the difference in carbon emissions per unit of electricity consumption among different power consumption nodes. Different power consumption nodes will have different carbon emission coefficients due to factors such as their equipment types, production processes, and energy structures.

[0121] After calculating the carbon emissions, the next step is to compare the carbon emission value of actual electricity consumption with the carbon emission value of predicted electricity consumption. Through this comparison, the carbon emission energy efficiency can be judged based on the difference between the two. Specifically, when the difference is less than the preset comparison threshold, it means that the carbon emissions of actual electricity consumption are relatively close to the predicted value, indicating that during the power use process, the energy utilization efficiency is high and carbon emissions are well controlled, that is, the carbon emission energy efficiency is high. For example, in some enterprises that have adopted advanced energy-saving technologies and optimized production processes, the difference between the carbon emission value of actual electricity consumption and the predicted value is often less than the preset comparison threshold, indicating that the enterprise has achieved good results in energy conservation and emission reduction. On the contrary, when the difference is greater than the preset threshold G, it means that the carbon emissions of actual electricity consumption exceed the predicted range, reflecting that there may be problems such as energy waste and equipment aging during the power use process, resulting in a decrease in carbon emission efficiency. For example, some old high-energy-consuming equipment may cause the actual carbon emissions to be much higher than the predicted value due to reasons such as backward technology and improper maintenance.

[0122] In this embodiment, step 7 includes the following steps: obtaining the time series of power consumption data, the power consumption data of bottleneck points and high-energy consumption points, and forming a classification data set;

[0123] Analyzing the classification data set through a clustering algorithm to classify the user side and divide the user side into multiple power consumption groups; the power consumption groups include high-carbon emission efficiency power consumption groups, low-carbon emission efficiency power consumption groups, and abnormal carbon emission efficiency power consumption groups;

[0124] Guide corresponding energy supply plans for different electricity - consuming groups, where the energy supply plans include adjusting the equipment usage time and improving the equipment usage efficiency.

[0125] In the embodiments disclosed by the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. The embodiments disclosed by the present invention include a computer program product, which includes a computer program carried on a computer - readable medium. The computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above - defined functions in the methods of the present application are executed. It should be noted that the computer - readable medium in the present application can be a computer - readable signal medium or a computer - readable storage medium or any combination of the two. A computer - readable storage medium can, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer - readable storage medium can include, but are not limited to: an electrical connection having one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read - only memory (ROM), an erasable programmable read - only memory (EPROM or flash memory), an optical fiber, a portable compact disk read - only memory (CD - ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer - readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction - execution system, apparatus, or device. In the present application, a computer - readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which the computer - readable program code is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer - readable signal medium can also be any computer - readable medium other than a computer - readable storage medium, and this computer - readable medium can send, propagate, or transmit a program for use by or in combination with an instruction - execution system, apparatus, or device. The program code contained on the computer - readable medium can be transmitted by any suitable medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination of the above.

[0126] The flowcharts in the accompanying drawings illustrate the architecture, functionality, and operation that the methods according to various embodiments of the present invention may achieve. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0127] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the said principles, any changes or modifications can be made to the embodiments of the present invention.

Claims

1. A simulation method for the energy efficiency of a client load, characterized in that, The method includes: Obtaining the historical data of the client side, constructing a demand trend model based on a recurrent neural network to capture the dynamic changes and periodic characteristics of the electricity consumption data of the client side, and predicting the electricity demand within a future time period; After obtaining the predicted value of the electricity demand within a future time period, analyzing the energy supply demands among multiple nodes of the client side, obtaining the energy flow route, and obtaining the energy supply bottleneck points and high energy consumption points according to the energy flow route; Among them, according to the predicted demand value, a linear programming model is constructed, and the objective function is set to minimize the energy cost or maximize the energy efficiency; using graph theory algorithms, according to the capacity limit and transmission loss of the line, calculate the shortest path or the lowest cost path from the energy supply source to each electricity consumption node as the optimal energy flow route; according to the obtained optimal energy flow route, calculate the flow rate of each node, which is the energy passing through the node, and the pressure, which is the ratio of the energy processed by the node to its maximum processing capacity; specifically: obtain the energy flow line; calculate the maximum flow from the source node to the sink node through the Ford-Fulkerson algorithm or the Edmonds-Karp algorithm; this process is achieved by continuously finding the augmenting path and updating the residual graph; After calculating the maximum flow, check the utilization rate of each edge, that is, the ratio of the actual flow to the capacity; the edge with a utilization rate of 100% ± 2% is the bottleneck point; Calculate the difference between the total input and output flows of each electricity consumption node, and the node with a difference greater than the preset threshold is the high energy consumption point; Obtain the data of the bottleneck points and high energy consumption points, introduce additional dimensions, construct a multi-dimensional analysis model, and predict the power supply demand and electricity consumption demand of the microgrid system through the multi-dimensional analysis model; Among them, obtain and merge multiple data sources, including the energy consumption data of high energy consumption points and bottleneck points, the quantization data of environmental temperature, weather events, and maintenance events; the merged data set contains the energy consumption, environmental temperature, weather events, and maintenance event information of each point; Perform normalization processing on the data to eliminate the dimensional differences between different features and ensure the stability and accuracy of model training; select feature variables from the preprocessed data, such as environmental temperature, weather events, and maintenance events as the input of the model; define the target variable as the total energy consumption of each point, that is, the sum of the energy consumption of high energy consumption points and bottleneck points; use the Random Forest Regression model (RandomForestRegressor) to predict the power supply demand; use the trained model to predict the power supply demand for a future time; analyze the prediction results to find the maximum and minimum power supply demands and their corresponding time points; Construct a digital virtual model of the energy supply system to simulate and verify the electricity consumption characteristics and energy flow process of the client side under different electricity consumption scenarios, and verify the reliability of the demand trend model and the multi-dimensional analysis model; Establish an efficiency evaluation model to evaluate the carbon emission energy efficiency of the client side according to the predicted values of the trend model and the multi-dimensional analysis model; According to the evaluation results of the previous step, divide the client side into different electricity consumption type groups through the clustering analysis algorithm, and formulate corresponding energy supply plans for each electricity consumption type group to improve the energy supply efficiency.

2. The simulation method for the energy efficiency of the client load according to claim 1, wherein Obtaining the historical data of the user side, constructing a demand trend model based on a recurrent neural network to capture the dynamic changes and periodic characteristics of the electricity consumption data of the user side, and predicting the electricity demand within a future time period, including: Obtaining the historical electricity consumption data of the user side within a preset time period, where the historical electricity consumption data includes the power supply amount, the electricity consumption amount, and the electricity consumption timestamp; Preprocessing the obtained historical electricity consumption data to form a historical electricity consumption data set, and extracting multiple electricity consumption characteristics from the historical electricity consumption data set to form an electricity consumption characteristic sample set; Using the TensorFlow or PyTorch deep learning framework to construct a demand trend model; Dividing the electricity consumption characteristic sample set into a training set and a test set, training the demand trend model through the training set, and validating the demand trend model through the test set; then, deploying the trained demand trend model to the system; Collecting the electricity consumption data of multiple electricity consumption nodes of the user side within a preset time length at a preset frequency, and extracting multiple electricity consumption data characteristics to form an electricity consumption data time series; the electricity consumption data characteristics include the maximum electricity consumption, the minimum electricity consumption, the maximum power supply amount, the minimum power supply amount, the maximum power supply time, the minimum power supply time, the maximum electricity consumption time, and the minimum electricity consumption time; After normalizing the electricity consumption data time series, inputting it into the demand trend model, and outputting the predicted values of the maximum power supply amount, the maximum power supply time, the maximum electricity consumption, and the maximum electricity consumption time of an electricity consumption node within a future time period.

3. The simulation method for user - side load energy efficiency according to claim 2, wherein, After obtaining the predicted values of the electricity demand within a future time period, analyzing the energy supply demands among multiple nodes of the user side, obtaining the energy flow route, and obtaining the energy supply bottleneck points and high energy consumption points according to the energy flow route, including: Obtaining the predicted values of the electricity demand of each node within a future time period from the power grid management system; According to the obtained predicted values, establishing a linear programming model, with the goal of minimizing the energy cost and / or maximizing the energy efficiency, and analyzing the energy supply demands of each node; Using graph theory algorithms to determine the optimal energy flow route from the energy supply source to each electricity consumption node according to the capacity limit and transmission loss of the line; By analyzing the energy flow route, identifying the bottleneck points in the system, that is, those bottleneck points and high energy consumption points that limit the overall system performance, by calculating the flow and pressure of each node.

4. The simulation method for the user-side load energy efficiency according to claim 3, wherein According to the obtained predicted values, establishing a linear programming model, with the goal of minimizing the energy cost and / or maximizing the energy efficiency, and analyzing the energy supply demands of each node, including: Use to represent the energy demand of the electricity-consuming node ; If the goal is to minimize the energy cost, the objective function is: ; where represents the unit energy cost of the th electricity consumption node, and represents the total number of electricity consumption nodes at the user side; If the goal is to maximize energy efficiency, the objective function is: ; where represents the unit energy efficiency of the th electricity-consuming node; Add constraint condition 1: , where represents the maximum supply capacity of the electricity-consuming node ; Add constraint condition two: , where represents the total energy supply of the user side; Solving the objective function by the simplex method or the interior point method; According to the solution results, determining the optimal energy demand for each stage, and calculating and obtaining the corresponding cost or efficiency.

5. The simulation method for user - end load energy efficiency according to claim 4, wherein, Using graph theory algorithms to determine the optimal energy flow route from the energy supply source to each electricity consumption node according to the capacity limit and transmission loss of the line, including: Let there be a directed graph G=(V, E); where, V is the set of vertices, and each vertex represents an energy supply source and an electricity consumption node; E is the set of edges, and each edge e=(u, v) has a capacity limit c(e) and a transmission loss w(e); Create a distance array D to store the shortest distances from the source node to other nodes. Initially, the distances of all nodes are set to infinity, and the distance of the source node is set to 0; Create a priority queue P to store the nodes to be processed and their current distances; Add the source node to the priority queue with a distance of 0; When the priority queue is not empty, extract the node u with the minimum distance; For each adjacent node v of node u, calculate the distance new_D to reach v through u as new_D = D[u] + w(u, v); where w(u, v) represents the weight of the edge from node u to node v; If new_D is less than D[v], then update D[v] and add v to the priority queue; When the priority queue is empty, end the calculation; at this time, the D array stores the shortest distances from the source node to each node; Thus, obtain the optimal energy flow routes from the energy supply source to each power-consuming node, minimizing the total transmission loss.

6. The simulation method for the user terminal load energy efficiency according to claim 5, wherein By analyzing the energy flow routes, identify the bottleneck points in the system, i.e., those points that limit the overall system performance and high-energy-consuming points, by calculating the flow and pressure at each node, including: Obtain the energy flow lines; Calculate the maximum flow from the source node to the sink node using the Ford-Fulkerson algorithm or the Edmonds-Karp algorithm; this process is achieved by continuously finding augmenting paths and updating the residual graph; After calculating the maximum flow, check the utilization rate of each edge, i.e., the ratio of the actual flow to the capacity; the edges with a utilization rate of 100% ± 2% are the bottleneck points; Calculate the difference between the total input and output flows of each power-consuming node, and the nodes with a difference greater than the preset threshold are the high-energy-consuming points.

7. A simulation method for the energy efficiency of the user terminal load according to claim 6, characterized in that Obtain the data of the bottleneck points and high-energy-consuming points, introduce auxiliary dimensions, and construct a multi-dimensional analysis model to predict the power supply demand and power consumption demand of the microgrid system through the multi-dimensional analysis model, including: Obtain the data of the high-energy-consuming points and the bottleneck points to form a two-dimensional energy consumption data set; Obtain the ambient temperature, weather events, and maintenance events at the power-consuming nodes, and after quantifying the ambient temperature, weather events, and maintenance events, form an auxiliary dimension data set; After normalizing the two-dimensional energy consumption data set and the auxiliary dimension data set, combine them into a multi-dimensional data set; Construct a multi-dimensional analysis model using the random forest algorithm, with the multi-dimensional data set as the input variable, to predict the power supply demand of the high-energy-consuming points and bottleneck points in the next time period, where the power supply demand includes the maximum power supply, the maximum power supply demand time, the minimum power supply, and the minimum power supply demand time.

8. The simulation method for the user-side load energy efficiency according to claim 7, wherein Construct a digital virtual model of the energy supply system to simulate and verify the reliability of the demand trend model and the multi-dimensional analysis model for the power consumption characteristics and energy flow process at the user side under different power consumption scenarios, including: Obtain the power consumption characteristic data of the user side, where the power consumption characteristic data includes device type, power consumption, and usage time; Obtain the power consumption demand predicted by the demand trend model and the multi-dimensional analysis model for the next time period; According to the power grid structure, power generation capacity, and electricity demand, simulate the flow process of electric energy from the power generation station to the user side; specifically: use the discrete event simulation or system dynamics simulation method to simulate the energy flow of the entire energy supply system.

9. A simulation method for the energy efficiency of the client load according to claim 8, characterized in that The establishment of the energy efficiency evaluation model is to evaluate the carbon emission energy efficiency of the user side according to the predicted values of the trend model and the multi-dimensional analysis model, including: Obtain the predicted electricity consumption of the electricity consumption nodes in the user side, and the electricity consumption nodes include bottleneck points, high energy consumption points, non-bottleneck points, and non-high energy consumption points; Import the carbon emission coefficient and calculate the carbon emissions of each power consumption node, C = W * p i , where W represents the predicted power consumption of the power consumption node, and p i represents the carbon emission coefficient of the i-th power consumption node; Compare the carbon emission value of the actual electricity consumption with the carbon emission value of the predicted electricity consumption, and judge the carbon emission energy efficiency according to the difference; that is, the smaller the difference, the higher the carbon emission energy efficiency; on the contrary, the lower the carbon emission efficiency; According to the evaluation results of the previous step, use the clustering analysis algorithm to divide the user side into different electricity consumption type groups, and formulate corresponding energy supply plans for each electricity consumption type group to improve the energy supply efficiency, including: Obtain the time series of electricity consumption data, the electricity consumption data of bottleneck points and high energy consumption points, and form a classification data set; Analyze the classification data set through the clustering algorithm, classify the user side, and divide the user side into multiple electricity consumption groups; the electricity consumption groups include high carbon emission efficiency electricity consumption groups, low carbon emission efficiency electricity consumption groups, and abnormal carbon emission efficiency electricity consumption groups; Guide corresponding energy supply plans for different electricity consumption groups, and the energy supply plans include adjusting the equipment use time and improving the equipment use efficiency.

Images