High-energy-consumption industrial micro-grid hierarchical scheduling optimization method based on load classification

Abstract

The invention discloses a hierarchical scheduling optimization method for a high-energy-consumption industrial micro-grid based on load classification, and relates to the technical field of high-energy-consumption industrial micro-grid optimizing.The method comprises the steps that clustering analysis is conducted on loads of the high-energy-consumption industrial micro-grid, load groups with similar characteristics are recognized, and key characteristics are extracted; classifying the loads according to a load clustering result; a mathematical model of the high-energy-consumption industrial micro-grid is established, and coordinated optimization and operation efficiency of each subsystem are ensured through hierarchical scheduling optimization. According to the method, the high-energy-consumption industrial micro-grid load is classified based on the clustering analysis result, the accuracy and applicability of load classification are improved, the characteristics of the high-energy-consumption industrial load can be described more accurately and comprehensively by establishing an accurate mathematical model of the high-energy-consumption industrial micro-grid, and the load classification efficiency is improved through hierarchical scheduling optimization. And a hierarchical coordinated optimization model of the high-energy-consumption industrial micro-grid is established, so that coordinated optimization of each subsystem in the micro-grid is realized, and the overall performance of the system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of high-energy-consuming industrial microgrid optimization, and in particular to a high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification. Background Art

[0002] With the rapid development of high-energy-consuming industries, industrial microgrids, as a flexible energy system supporting high-energy-consuming industries, have received more and more attention. However, existing technologies still have obvious deficiencies in load analysis and classification, mathematical model construction and optimization model design, which restricts the further development of high-energy-consuming industrial microgrids.

[0003] In terms of load analysis and classification, existing technologies usually use traditional statistical analysis methods. Traditional statistical analysis methods can describe the basic characteristics of loads to a certain extent. However, for high-energy-consuming industrial loads, their characteristics are complex and have significant dynamic changes and diversity. Existing statistical analysis methods usually find it difficult to accurately capture the characteristic correlations between loads, resulting in inaccurate identification and classification results.

[0004] In terms of mathematical model construction, traditional microgrid load mathematical models are mostly based on overly simplified assumptions, such as assuming that loads are movable or of fixed categories. Although such simplified assumptions help reduce computational complexity, they cannot accurately reflect the complex characteristics of high-energy-consuming industrial loads, especially the nonlinear and uncertain characteristics of loads in a dynamically changing environment. In addition, traditional models fail to consider the interrelationships between different loads and their impact on the overall operating efficiency of the microgrid, which limits the applicability of existing models in load management and scheduling optimization.

[0005] In terms of optimization model design, optimization models in existing technologies are mostly based on a single-layer structure, and their goals are often focused on the optimal solution of a single task. However, the various subsystems within an industrial microgrid, such as power generation, energy storage, and load side, have a close collaborative relationship. The design of this single optimization model fails to effectively reflect the global coordination within the microgrid. In addition, when solving complex scenarios, existing optimization models usually lack support for hierarchical optimization, and it is difficult to take into account both global benefits and the dynamic adjustment capabilities of local subsystems. This deficiency not only limits the operating efficiency of the microgrid, but may also lead to energy waste and increased costs. Summary of the invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, the technical problems solved by the present invention are: the existing high-energy-consuming industrial microgrid load management methods have insufficient load classification accuracy, the mathematical model cannot fully reflect the load characteristics, the optimization model lacks hierarchical scheduling capabilities, and how to achieve coordinated optimization and efficient operation of various subsystems in the microgrid.

[0008] To solve the above technical problems, the present invention provides the following technical solutions: a hierarchical scheduling optimization method for high-energy-consuming industrial microgrids based on load classification, comprising clustering analysis of high-energy-consuming industrial microgrid loads, identifying load groups with similar characteristics and extracting key features; classifying loads according to load clustering results; establishing a mathematical model of high-energy-consuming industrial microgrids, and ensuring the coordinated optimization and operating efficiency of each subsystem through hierarchical scheduling optimization.

[0009] As a preferred solution of the high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification described in the present invention, the clustering analysis of the high-energy-consuming industrial microgrid load includes collecting load historical data, processing abnormal values ​​and missing values ​​in the data, extracting load characteristics, clustering the load data using the density-based DBSCAN algorithm, dividing the load into several clusters, and excluding abnormal load points, and calculating the distance between all data points based on the Euclidean distance, which is expressed as:

[0010]

[0011] Among them, p i is the power value of the i-th load data point, p j is the power value of the jth load data point, For p i The value in the nth dimension, For p j The value in the nth dimension; calculate all points within the threshold ∈ distance of the load data point p to form the ∈ neighborhood of the load data point p, expressed as:

[0012] N ∈ (p)={q∈D}|dist(p,q)≤∈

[0013] Among them, N ∈ (p) is the set of all points whose distance from the load data point p does not exceed the threshold ∈, D is the load data point data set, q is the load data point different from p, dist(p,q) is the Euclidean distance between the load data point p and the load data point q, ∈ is the distance threshold for judging the neighborhood; whether it is a core point is determined according to the number of neighborhood points, expressed as:

[0014]

[0015] Among them, minPts is the minimum number of neighboring points for judging the core point. If the neighborhood N of the load data point p is ∈ The number of points in (p) |N ∈ (p)| is greater than or equal to the minimum number of neighborhood points minPts, then the load data point p is a core point. If the neighborhood N of the load data point p∈ The number of points in (p) |N ∈ (p)| is less than the minimum number of neighborhood points minPts, then the load data point p is not a core point.

[0016] As a preferred solution of the high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification described in the present invention, wherein: the clustering analysis of the high-energy-consuming industrial microgrid load also includes evaluating the clustering effect by Davidson-Botting index and silhouette coefficient, which is expressed as:

[0017]

[0018] Where DB is the Davidson-Boting index, k is the number of clusters, σ i is the average distance of cluster i, σ j is the average distance of cluster j, d ij is the distance between clusters i and j.

[0019] As a preferred solution of the high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification described in the present invention, wherein: the load classification according to the load clustering result includes adjustable loads and unadjustable loads.

[0020] As a preferred scheme of the high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification described in the present invention, wherein: the adjustable load includes time-limited transferable load, continuous shiftable load and all adjustable loads; if the load is a time-limited transferable load, the operating time period can be adjusted within a predetermined time range without affecting the production task; if the load is a continuous shiftable load, the start-up time can be flexibly adjusted according to the requirements of the production line sequence, and the normal operation of the production line can still be ensured after adjustment.

[0021] As a preferred solution of the load classification-based high-energy-consuming industrial microgrid hierarchical scheduling optimization method described in the present invention, wherein: the mathematical model of establishing a high-energy-consuming industrial microgrid includes establishing a photovoltaic and wind turbine power model, which is expressed as:

[0022]

[0023] in, is the output power of photovoltaic and wind turbine at time t, is the maximum output power of photovoltaic and wind turbine at time t; the gas turbine power model is established and expressed as:

[0024]

[0025] in, are the electrical and thermal power output of the gas turbine at time t, is the natural gas power consumed by the gas turbine at time t, are the energy conversion efficiency of gas turbine output electricity and thermal power, are the upper and lower limits of the gas turbine output electrical and thermal power at time t respectively; a gas boiler power model is established, expressed as:

[0026]

[0027] in, is the thermal power output of the gas boiler at time t, is the natural gas power consumed by the gas boiler at time t, The energy conversion efficiency of the gas boiler output thermal power, are the upper and lower limits of the thermal power output of the gas boiler at time t respectively; the power model of the electric refrigeration / electric heating equipment is established, which is expressed as:

[0028]

[0029] in, is the output of the electric cooling / heating equipment at time t, η c / e is the energy conversion efficiency, is the power consumption of the electric cooling / heating equipment at time t, are the upper and lower limits of the output of the electric cooling / heating equipment at time t respectively; the power model of the absorption chiller is established, which is expressed as:

[0030]

[0031] in, are the heat power input and cooling power output of the absorption refrigerator at time t, respectively, AC is the conversion efficiency of the absorption chiller, are the upper and lower limits of the cooling power output of the absorption chiller at time t respectively; the design constraints of the energy storage equipment include the energy storage charging and discharging constraints, which are expressed as:

[0032]

[0033] in, is the capacity of the energy storage device at time t, is the capacity of the energy storage at time t+1, η ch is the energy storage charging efficiency, η dis is the energy storage discharge efficiency, is the charging power of the energy storage device at time t, is the discharge power of the energy storage device at time t, and Δt is the change time; if the charging and discharging power of the energy storage device is equal within the time period, the energy storage charging and discharging constraint condition is expressed as:

[0034]

[0035] Where T is the total duration of the time period; the capacity constraint and power constraint of the energy storage device at each moment are calculated, expressed as:

[0036]

[0037] in, are the upper and lower limits of the energy storage device capacity at time t, are the upper and lower limits of the charging power of the energy storage device at time t, are the upper and lower limits of the energy storage device discharge power at time t, are the 0-1 state variables of the energy storage device charging and discharging at time t, When the value is 0, it means the device stops working. When the value is 1, it means the device is in working state.

[0038] As a preferred solution of the hierarchical dispatch optimization method of high-energy-consuming industrial microgrid based on load classification described in the present invention, the hierarchical dispatch optimization includes establishing a hierarchical coordination optimization model of high-energy-consuming industrial microgrid and establishing a total revenue function of high-energy-consuming microgrid, which is expressed as:

[0039]

[0040] Among them, F IMG is the total revenue function of the high energy consumption microgrid, It is the energy purchasing function of high energy consumption microgrid from the upper power grid. is the energy selling function of the high energy consumption microgrid, For the low-carbon operation function of high-energy-consuming microgrids, It is the function for maximizing the utilization of photovoltaic and wind power in high-energy-consuming microgrids. is the maintenance rate of high energy consumption microgrid equipment; calculate the inequality constraints of high energy consumption industrial microgrid, expressed as:

[0041]

[0042] in, β e,h,c They are the upper and lower limits of the microgrid energy sales price, are the average constraints for selling and buying energy prices, are the constraints on selling and buying energy prices at time t, respectively, i Output power to the i-th device in the microgrid, P i are the upper and lower limits of equipment output, E i ,H i ,C iThey are the constraints of electricity, heat and cold energy provided by the microgrid to users; the power balance equation constraints of the high-energy-consuming industrial microgrid are calculated, which are expressed as:

[0043]

[0044] Among them, E e 、E h 、E c are the maximum power of electricity, heat and cold energy provided by high-energy-consuming industrial microgrids, P e,i , P h,i , P c,i are the electricity, heat and cold output of the i-th equipment respectively; establish the total revenue function of high-energy-consuming industrial users, conduct internal optimization of the high-energy-consuming industrial parks in the jurisdiction of the user aggregator, be responsible for connecting with load users, and integrate dispersed load resources. The total revenue function of high-energy-consuming industrial users is expressed as:

[0045]

[0046] Among them, F USER is the total revenue function of high energy-consuming industrial users, is the user utility function, To participate in the auxiliary peak load subsidy, is the energy consumption function, is the energy storage capacity leasing function, is the CO2 emission cost; the inequality constraint for high energy-consuming industrial load is calculated as:

[0047]

[0048] in, They are the upper and lower limits of the adjustable electric load for high-energy-consuming industrial loads respectively; the high-energy-consuming industrial user optimization layer is solved by the CPLEX solver, and the user's optimal electricity consumption is taken as input. The high-energy-consuming industrial microgrid dispatching layer uses a genetic algorithm to achieve the optimal dispatch of equipment within the jurisdiction.

[0049] Another object of the present invention is to provide a hierarchical scheduling optimization system for high-energy-consuming industrial microgrids based on load classification, which can establish a mathematical model of the high-energy-consuming industrial microgrid and ensure the coordinated optimization and operating efficiency of each subsystem through hierarchical scheduling optimization, thereby solving the problem that the current optimization model design contains an optimization model that lacks hierarchical scheduling capabilities.

[0050] As a preferred solution of the high-energy-consuming industrial microgrid hierarchical scheduling optimization system based on load classification described in the present invention, it includes: a load analysis module, a load classification module, and a hierarchical coordination optimization module; the load analysis module is used to perform cluster analysis on the loads of the high-energy-consuming industrial microgrid, identify load groups with similar characteristics and extract key features; the load classification module is used to classify the loads according to the load clustering results; the hierarchical coordination optimization module is used to establish a mathematical model of the high-energy-consuming industrial microgrid, and ensure the coordinated optimization and operation efficiency of each subsystem through hierarchical scheduling optimization.

[0051] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement a step of a hierarchical scheduling optimization method for high-energy-consuming industrial microgrids based on load classification.

[0052] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a hierarchical scheduling optimization method for high-energy-consuming industrial microgrids based on load classification.

[0053] Beneficial effects of the present invention: The hierarchical scheduling optimization method for high-energy-consuming industrial microgrids based on load classification provided by the present invention performs a comprehensive cluster analysis on the loads in the high-energy-consuming industrial microgrid, identifies different types of load characteristics, and classifies the loads of the high-energy-consuming industrial microgrid based on the cluster analysis results, thereby improving the accuracy and applicability of load classification, providing more accurate basic load data for the scheduling optimization of high-energy-consuming industrial microgrids, and improving the efficiency and effectiveness of energy management. By establishing an accurate mathematical model of high-energy-consuming industrial microgrids, it is possible to more accurately and comprehensively describe the characteristics of high-energy-consuming industrial loads, provide reliable model support for the scheduling optimization of high-energy-consuming industrial microgrids, and improve the overall scheduling optimization effect. Through hierarchical scheduling optimization and establishing a hierarchical coordination optimization model of high-energy-consuming industrial microgrids, coordinated optimization of each subsystem in the microgrid is achieved, and the overall performance of the system, including energy utilization efficiency, high-energy-consuming industrial load balancing capability and system stability, is improved. The present invention achieves better results in terms of accuracy, adaptability and stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0055] Figure 1 An overall flow chart of a high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification provided in the first embodiment of the present invention.

[0056] Figure 2 A module schematic diagram of a high-energy-consuming industrial microgrid hierarchical scheduling optimization system based on load classification provided in the third embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0058] Example 1, reference Figure 1 , is an embodiment of the present invention, and provides a high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification, comprising:

[0059] S1: Cluster analysis is performed on high-energy-consuming industrial microgrid loads to identify load groups with similar characteristics and extract key features.

[0060] Furthermore, cluster analysis of high-energy-consuming industrial microgrid loads includes collecting historical load data, processing outliers and missing values ​​in the data, extracting load features, clustering load data using the density-based DBSCAN algorithm, dividing the load into several clusters, and excluding abnormal load points. The distance between all data points is calculated based on the Euclidean distance, which is expressed as:

[0061]

[0062] Among them, p i is the power value of the i-th load data point, p j is the power value of the jth load data point, For p i The value in the nth dimension, For p j The value in the nth dimension; calculate all points within the threshold ∈ distance of the load data point p to form the ∈ neighborhood of the load data point p, expressed as:

[0063] N ∈ (p)={q∈D}|dist(p,q)≤∈

[0064] Among them, N ∈(p) is the set of all points whose distance from the load data point p does not exceed the threshold ∈, D is the load data point data set, q is the load data point different from p, dist(p,q) is the Euclidean distance between the load data point p and the load data point q, ∈ is the distance threshold for judging the neighborhood; whether it is a core point is determined according to the number of neighborhood points, expressed as:

[0065]

[0066] Among them, minPts is the minimum number of neighboring points for judging the core point. If the neighborhood N of the load data point p is ∈ The number of points in (p) |N ∈ (p)| is greater than or equal to the minimum number of neighborhood points minPts, then the load data point p is a core point. If the neighborhood N of the load data point p ∈ The number of points in (p) |N ∈ (p)| is less than the minimum number of neighborhood points minPts, then the load data point p is not a core point.

[0067] It should be noted that cluster analysis of high-energy-consuming industrial microgrid loads also includes the evaluation of clustering effects through the Davidson-Botting index and silhouette coefficient, which is expressed as:

[0068]

[0069] Where DB is the Davidson-Boting index, k is the number of clusters, σ i is the average distance of cluster i, σ j is the average distance of cluster j, d ij is the distance between clusters i and j.

[0070] It should also be noted that clustering analysis of various types of loads in high-energy-consuming industrial microgrids can help identify load groups with similar characteristics, thus providing a basis for subsequent load classification and optimized scheduling.

[0071] The data collection can collect historical data of various loads in high-energy-consuming industrial microgrids, including but not limited to power consumption characteristics of various high-energy-consuming loads, load change patterns, operating time, ambient temperature and other related data.

[0072] Furthermore, data preprocessing is performed to clean and standardize the collected data and handle outliers and missing values ​​to ensure the quality and consistency of the data.

[0073] Furthermore, feature extraction is performed to extract key features from the preprocessed data, such as average load, peak load, load volatility, load duration, and load time characteristics.

[0074] Furthermore, cluster analysis is performed by using density-based spatial clustering with noise and applying the DBSCAN algorithm to cluster the high-energy-consuming industrial load data. Several clusters can be obtained, each cluster represents a type of high-energy-consuming load pattern, and several noise points do not belong to any cluster, so the noise points can be regarded as abnormal loads.

[0075] S2: Classify the loads according to the load clustering results.

[0076] Furthermore, the loads are classified into controllable loads and uncontrollable loads according to the load clustering results.

[0077] It should be noted that adjustable loads include time-limited transferable loads, continuously shiftable loads and all adjustable loads; if the load is a time-limited transferable load, the operating time period can be adjusted within the predetermined time range without affecting the production task; if the load is a continuously shiftable load, the start-up time can be flexibly adjusted according to the sequence requirements of the production line, and the normal operation of the production line can still be ensured after adjustment.

[0078] It should also be noted that the loads are classified based on the clustering results to determine the characteristics and control methods of each type of load. Among them, the non-adjustable loads usually have high requirements for power supply reliability. Once their power consumption behavior is changed, it may cause the products being produced to be scrapped, damage production equipment, or threaten the personal safety of production personnel. Therefore, they cannot be easily adjusted.

[0079] Adjustable loads include time-limited transferable loads, continuous translational loads and all adjustable loads. The specific characteristics are as follows:

[0080] 1. Time-limited transferable load: refers to industrial loads whose usage time can be adjusted within a certain time range. The main feature of this type of industrial load is that their usage time has a certain flexibility and can be shifted within a predetermined time window without affecting the completion of their functions or tasks, expressed as:

[0081]

[0082] Among them, P' IMG (t) is the time-limited transferable industrial load at time t, μ is the time-limited transferable industrial load as a percentage of the total load P in a day IMG ratio, T is the maximum delay time of the transferable industrial load, P' IMG (t+i) is the load component of the limited-time transferable industrial load at time t allocated to time t+i.

[0083] The time-limited shiftable load can enable the high-energy-consuming industrial microgrid to participate in the operation and dispatch of the upper power grid as a demand response load. Due to the time-limited shiftable feature, the maximum delay processing time of such industrial load is 3 hours. The specific modeling method is as follows:

[0084] Introduce three 24×24 dimensional matrices, expressed as:

[0085]

[0086] Among them, Y is the auxiliary matrix, Z is the time-limited translation matrix, For a square array, is the power matrix.

[0087] 1) The auxiliary matrix Y limits the load translation time and is a 0-1 variable, expressed as:

[0088]

[0089] 2) Diagonalize the time-limited translation matrix load into a square matrix It is expressed as:

[0090]

[0091] Among them, Z t,t In order to adjust the time-limited translatable load before time t, the Z matrix indicates that the time-limited translatable load at time t can be translated to time t+1, t+2, and t+3, while the introduction of the auxiliary matrix Y limits the time period in which such load can be translated.

[0092] 3) Power matrix after time-limited load adjustment It is expressed as:

[0093]

[0094] Among them, the column vector represents the time period during which the time-limited shiftable load can be adjusted, and the row vector represents the load distribution after the time-limited shiftable load is adjusted at the current moment; in order to limit the shifted load from exceeding the total original load, the following constraints are set, expressed as:

[0095] y t+n,t ≤z t,t ,t=1,2,...,24;n=0,1,2,3

[0096] The tth row of the time-limited translatable matrix before adjustment is transferred to the t+1, t+2, and t+3 rows. At this time, the sum of the column vectors of the tth column is the time-limited translatable load at time t before adjustment, which is expressed as:

[0097]

[0098] For the power distribution at time t after adjustment, The sum of the row vectors in the tth row is expressed as:

[0099]

[0100] In order to avoid excessive load regulation at a certain moment, a time-limited upper limit constraint on the load that can be translated can be set, which can be expressed as:

[0101]

[0102] Continuous shiftable load: This type of industrial load needs to control the sequence of production lines and produce complete products according to certain production steps. There is no strict requirement for the start time of each production line. The start time of this type of load can be adjusted appropriately according to demand. Continuous high-energy-consuming industrial loads need to control the sequence of production lines and ensure the continuous supply of electricity within a certain period of time, but there is no strict requirement for the start time of each production line. This application takes the continuous high-energy-consuming industrial load appearing at 03:00-06:00 as an example. The specific modeling method is as follows:

[0103] 1) Introduce the continuous high-energy-consuming industrial load matrix P from 03:00 to 06:00 IMG_ys , expressed as:

[0104] P IMG_ys =[c3,c4,c5,c6]

[0105] 2) Introduce auxiliary matrix P IMG_sk , whose matrix elements are all 0-1 Boolean variables, expressed as:

[0106]

[0107] In order to ensure that the load can ensure continuous power supply within a certain period of time, and to limit the continuous high-energy-consuming industrial load to be able to be shifted to another time as a whole before and after adjustment, and not to multiple times, P should be set. IMG_sk Each column vector has only one constraint of 1, expressed as:

[0108]

[0109] 3) Introduce the post-transfer time matrix T b Determine the specific time after the continuous high energy consumption industrial load is adjusted, expressed as:

[0110] T1=[1,2,3,…,24]

[0111] T b =T a P IMG_sk

[0112] First, considering that the continuous high-energy-consuming industrial load starts at 03:00, the load after transfer should not be earlier than 03:00. b The matrix elements in the matrix are set with the following constraints:

[0113]

[0114] t1=1,2

[0115] Secondly, continuous high-energy-consuming industrial loads have a time-series characteristic, that is, they need to follow certain production steps to produce a complete product, and this sequence cannot be changed after adjustment. b The matrix elements in the matrix are set with the following constraints:

[0116]

[0117] The adjusted continuous high energy consumption industrial load matrix is ​​expressed as:

[0118]

[0119] in, is the continuous translatable load matrix that appears at time T.

[0120] 3. All adjustable loads: The start time and load size of this type of load can be transferred or reduced appropriately. It usually refers to the auxiliary load used by the enterprise for office lighting, office appliances, air conditioning, refrigeration, heating and other functions, expressed as:

[0121]

[0122] Among them, P'' IMG is the total adjustable load, P tran is the transferable load, P cut is the load that can be reduced, P"' max 、P”' min The design of maximum and minimum values ​​should be based on the principle of giving priority to production load regulation.

[0123] S3: Establish a mathematical model of high-energy-consuming industrial microgrids and ensure the coordinated optimization and operation efficiency of each subsystem through hierarchical scheduling optimization.

[0124] Furthermore, the high-energy-consuming industrial microgrid operator is the supplier of energy on the user side, which connects the energy supply system and the user together; IMG meets the energy demand of users through combined cooling, heating and power generation (CCHP) units, and purchases electricity and natural gas by directly connecting to the power grid and gas pipeline network. The combined cooling, heating and power generation unit consists of a gas turbine (GT), a gas boiler (GB), an absorption chiller (AC), an electric chiller (EC), and an electric boiler (EB). In this paper, a distributed photovoltaic generator (PV) is installed on the high-energy-consuming industrial user side, and energy storage (ES) services can be used to achieve high charging and low discharging of electricity.

[0125] Furthermore, the mathematical model of high-energy-consuming industrial microgrids includes the power models of photovoltaic and wind turbines, which can be expressed as:

[0126]

[0127] in, is the output power of photovoltaic and wind turbine at time t, is the maximum output power of photovoltaic and wind turbine at time t; the gas turbine power model is established and expressed as:

[0128]

[0129] in, are the electrical and thermal power output of the gas turbine at time t, is the natural gas power consumed by the gas turbine at time t, are the energy conversion efficiency of gas turbine output electricity and thermal power, are the upper and lower limits of the gas turbine output electrical and thermal power at time t respectively; a gas boiler power model is established, expressed as:

[0130]

[0131] in, is the thermal power output of the gas boiler at time t, is the natural gas power consumed by the gas boiler at time t, The energy conversion efficiency of the gas boiler output thermal power, are the upper and lower limits of the thermal power output of the gas boiler at time t respectively; the power model of the electric refrigeration / electric heating equipment is established, which is expressed as:

[0132]

[0133] in, is the output of the electric cooling / heating equipment at time t, η c / e is the energy conversion efficiency, is the power consumption of the electric cooling / heating equipment at time t, are the upper and lower limits of the output of the electric cooling / heating equipment at time t respectively; the power model of the absorption chiller is established, which is expressed as:

[0134]

[0135] in, are the heat power input and cooling power output of the absorption refrigerator at time t, respectively, AC is the conversion efficiency of the absorption chiller, are the upper and lower limits of the cooling power output of the absorption chiller at time t respectively; the design constraints of the energy storage equipment include the energy storage charging and discharging constraints, which are expressed as:

[0136]

[0137] in, is the capacity of the energy storage device at time t, is the capacity of the energy storage at time t+1, η ch is the energy storage charging efficiency, η dis is the energy storage discharge efficiency, is the charging power of the energy storage device at time t, is the discharge power of the energy storage device at time t, and Δt is the change time; if the charging and discharging power of the energy storage device is equal within the time period, the energy storage charging and discharging constraint condition is expressed as:

[0138]

[0139] Where T is the total duration of the time period; the capacity constraint and power constraint of the energy storage device at each moment are calculated, expressed as:

[0140]

[0141] in, are the upper and lower limits of the energy storage device capacity at time t, are the upper and lower limits of the charging power of the energy storage device at time t, are the upper and lower limits of the energy storage device discharge power at time t, are the 0-1 state variables of the energy storage device charging and discharging at time t, When the value is 0, it means the device stops working. When the value is 1, it means the device is in working state.

[0142] It should be noted that hierarchical dispatch optimization includes establishing a hierarchical coordination optimization model for high-energy-consuming industrial microgrids and establishing a total revenue function for high-energy-consuming microgrids, which is expressed as:

[0143]

[0144] Among them, F IMG is the total revenue function of the high energy consumption microgrid, It is the energy purchasing function of high energy consumption microgrid from the upper power grid. is the energy selling function of the high energy consumption microgrid, For the low-carbon operation function of high-energy-consuming microgrids, It is the function for maximizing the utilization of photovoltaic and wind power in high-energy-consuming microgrids. Indicates the wind and solar power abandonment rates, is the maintenance rate of high energy consumption microgrid equipment; calculate the inequality constraints of high energy consumption industrial microgrid, expressed as:

[0145]

[0146] in, β e,h,c They are the upper and lower limits of the microgrid energy sales price, are the average constraints for selling and buying energy prices, are the constraints on selling and buying energy prices at time t, respectively, i Output power to the i-th device in the microgrid, P i are the upper and lower limits of equipment output, E i ,H i ,C i They are the constraints of electricity, heat and cold energy provided by the microgrid to users; the power balance equation constraints of the high-energy-consuming industrial microgrid are calculated, which are expressed as:

[0147]

[0148] Among them, E e 、E h 、E c are the maximum power of electricity, heat and cold energy provided by high-energy-consuming industrial microgrids, P e,i , P h,i , P c,i are the electricity, heat and cold output of the i-th equipment respectively; establish the total revenue function of high-energy-consuming industrial users, conduct internal optimization of the high-energy-consuming industrial parks in the jurisdiction of the user aggregator, be responsible for connecting with load users, and integrate dispersed load resources. The total revenue function of high-energy-consuming industrial users is expressed as:

[0149]

[0150] Among them, F USER is the total revenue function of high energy-consuming industrial users, is the user utility function, To participate in the auxiliary peak load subsidy, is the energy consumption function, is the energy storage capacity leasing function, is the CO2 emission cost; the inequality constraint for high energy-consuming industrial load is calculated as:

[0151]

[0152] in, They are the upper and lower limits of the adjustable electric load for high-energy-consuming industrial loads respectively; the high-energy-consuming industrial user optimization layer is solved by the CPLEX solver, and the user's optimal electricity consumption is taken as input. The high-energy-consuming industrial microgrid dispatching layer uses a genetic algorithm to achieve the optimal dispatch of equipment within the jurisdiction.

[0153] It should also be noted that the establishment of the total revenue function of the high-energy-consuming microgrid includes calculating the energy purchase function of the high-energy-consuming microgrid from the superior power grid, which is expressed as:

[0154]

[0155] Among them, κ e , κ h The prices of electricity and heat purchased by high-energy-consuming microgrids from upstream energy suppliers are respectively is the power shortage of the high energy consumption microgrid at time t; the energy selling function of the high energy consumption microgrid is calculated and expressed as:

[0156]

[0157] Among them, β e , β h , β c are the electricity, heat and cooling energy prices provided by the microgrid to high-energy-consuming industrial users. are the electricity, heat and cooling power provided by the microgrid to high-energy-consuming industrial users at time t; the low-carbon operation function of the high-energy-consuming microgrid is calculated, which is expressed as:

[0158]

[0159] in, is the carbon emission coefficient, ε is the proportional coefficient, ε means that the carbon emissions generated by the electric and thermal power purchased by the microgrid from the superior network need to be shared with the superior network in a certain proportion; the maximization function of photovoltaic and wind power utilization of high-energy-consuming microgrids is calculated, which is expressed as:

[0160]

[0161] Calculate the maintenance rate of high energy consumption microgrid equipment, expressed as:

[0162]

[0163] Among them, P fa It is the rated power generated by the equipment in the microgrid.

[0164] It should also be noted that establishing the total benefit function of high-energy-consuming industrial users includes calculating the user efficiency function, which is expressed as:

[0165]

[0166] Among them, P IMG is the high energy-consuming industrial load after regulation, a, b, c are the parameters of the energy utility function; the subsidy for participating in auxiliary peak regulation is calculated as:

[0167]

[0168] Among them, μ e is the auxiliary peak regulation compensation parameter, is the total power involved in load regulation; calculate the energy consumption function, expressed as:

[0169]

[0170] Calculate the energy storage capacity leasing function, expressed as:

[0171]

[0172] Among them, μ es is the unit capacity price of energy storage for leasing; CO2 emission cost is calculated as:

[0173]

[0174] The equation constraint of high energy-consuming industrial load after calculation and regulation is expressed as:

[0175] P IMG =P′ IMG +P″ IMG +P″′ IMG

[0176] Among them, P' IMG is the time-limited transferable load, P″ IMG is the adjusted continuous translatable load matrix, P″′ IMG For all adjustable loads.

[0177] It should also be noted that to solve the hierarchical coordination optimization model of high-energy-consuming industrial microgrids, the microgrid dispatching layer needs to coordinate the internal equipment processing of high-energy-consuming industrial microgrids according to the power demand uploaded by the load aggregator, so as to pursue the maximum profit of high-energy-consuming industrial microgrids.

[0178] At the optimization layer for high-energy-consuming industrial users, load aggregators need to sign load regulation agreements with high-energy-consuming industrial users and perform demand-side response to the high-energy-consuming industrial loads they manage according to the needs of the superior power grid.

[0179] The optimization layer of high-energy-consuming industrial users is solved through CPLEX, and the optimal power consumption of users is taken as input. The high-energy-consuming industrial microgrid scheduling layer uses genetic algorithms to achieve the optimal scheduling of equipment within the jurisdiction.

[0180] The electricity and heat prices of the upper-level high-energy-consuming industrial microgrid operators are initialized and updated using genetic algorithms, and the electricity and heat consumption of the lower-level user aggregators are solved by calling the CPLEX solver. At this time, the user side only needs to accept the price signal of the multi-energy microgrid aggregator and upload the energy consumption signal of the current user, which can effectively avoid the leakage of information of various stakeholders and protect privacy and security. The specific solution process is as follows:

[0181] (1) Input basic data and initialize the parameters of each stakeholder.

[0182] (2) The upper-level multi-energy microgrid operator uses a genetic algorithm to randomly generate n sets of electricity and heat sales prices and passes these parameters to the lower level.

[0183] (3) The user aggregator receives n sets of electricity and heat prices randomly generated by the upper layer, and uses the CPLEX solver to optimize the output of each device and the user-side electricity demand response to obtain the current user operator revenue, and returns the current user power interaction to the upper layer.

[0184] (4) The multi-energy microgrid operator calculates the current revenue based on the power interaction returned by the lower layer.

[0185] (5) Generate new heat and electricity prices for the multi-energy microgrid using the selection, crossover, mutation, and boundary conditions of the genetic algorithm. Repeat steps (3) to (4) to obtain the benefits of the high-energy-consuming industrial microgrid and user aggregator.

[0186] (6)Benefit iteration.

[0187] (7) Determine convergence. If convergence occurs, terminate the program; otherwise, return to step (3).

[0188] Example 2 is an embodiment of the present invention, which provides a hierarchical scheduling optimization method for high-energy-consuming industrial microgrids based on load classification. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0189] This experiment aims to verify the innovation and advantages of clustering analysis, classification and hierarchical optimization scheduling of loads based on high-energy-consuming industrial microgrids. The experimental object is a high-energy-consuming industrial park, which contains multiple high-energy-consuming equipment, including time-limited transferable loads, continuous shiftable loads and unadjustable loads. The experiment is based on one week (7 days) of load data, covering key information such as power consumption characteristics, load fluctuation patterns, operating time and ambient temperature.

[0190] Test preparation:

[0191] 1. Data collection and preprocessing: Extract 7 days of historical data from the load monitoring system installed in the park, covering the operating characteristics of all equipment, including average power, peak power and operating time. Clean the data, remove outliers and fill in missing values, and standardize them to ensure consistency.

[0192] 2. Feature extraction: Extract key features, including average load, peak load, load volatility and continuous operation time, to provide a basis for subsequent clustering analysis.

[0193] 3. Algorithm selection: Use DBSCAN algorithm to perform load clustering analysis, set the Euclidean distance threshold and the minimum number of core point neighbors, and classify the data points into clusters.

[0194] 4. Clustering evaluation: The clustering effect is evaluated through the Davidson-Borgin index and silhouette coefficient to ensure the rationality of the grouping.

[0195] Experimental implementation:

[0196] 1. Load clustering analysis: The load data is imported into the DBSCAN algorithm, clustered according to power characteristics, and three main load clusters are identified: time-limited transferable loads, continuous transferable loads, and abnormal load points. Abnormal load points are eliminated and key load clusters are retained, laying the foundation for subsequent classification and scheduling optimization.

[0197] 2. Load classification: Based on the clustering results, the load is further subdivided. The time-limited transferable load includes processing workshop equipment, the continuous transferable load includes production line equipment, and the unadjustable load includes high-precision testing instruments. The scheduling strategy is designed according to the classification characteristics. For example, limiting the operation of the time-limited transferable load during peak time and shifting the continuous load to smooth the load curve.

[0198] 3. Hierarchical optimization scheduling: The hierarchical optimization model is used to schedule the microgrid. The scheduling layer optimizes the power of the combined heat and power equipment and distributed photovoltaics. The user-side optimization layer adjusts the operating parameters of various loads. The goal is to achieve a balance in total power and maximize energy utilization. The genetic algorithm is used to iteratively optimize the multi-objective benefit function.

[0199] Refer to Table 1 for a comparative analysis of the experimental data.

[0200] Table 1 Experimental data record table

[0201]

[0202]

[0203] It can be seen from the data in Table 1 that by implementing the clustering analysis and scheduling optimization strategy of the present invention, load management has achieved obvious results. First, the DBSCAN algorithm successfully divides the high-energy-consuming industrial microgrid load into three main load clusters (time-limited transferable load, continuous shiftable load and unadjustable load), and effectively eliminates abnormal load points. The clustering effect evaluation results show that the Davidson-Botting index reaches 1.5-1.8, and the silhouette coefficient remains above 0.8, which proves that the clustering results have good compactness and separation. Compared with traditional statistical analysis methods, it can more accurately identify load characteristics and reduce grouping errors.

[0204] Through load classification, it can be seen that the time-limited transferable load has high flexibility. After scheduling, the power is reduced from the original 800kW peak to 450kW, effectively reducing the pressure of the load peak on the microgrid. The continuous transferable load can shift the power peak from 1100kW to other time periods while maintaining the production sequence, making the overall load of the microgrid smoother, and the uncontrollable load remains stable, without causing production losses due to adjustment, which fully reflects the accuracy and rationality of the classification strategy.

[0205] The results of dispatch optimization show that the total load after dispatch is reduced from the peak value of 2350kW before dispatch to 1800kW. At the same time, the load distribution is smoothed, the stability of microgrid operation and energy utilization efficiency are improved. The genetic algorithm has a fast convergence speed in multi-objective optimization, and the optimization results are stable, which realizes the coordinated operation of cogeneration equipment and the load side. Compared with the traditional single-layer optimization model, the hierarchical optimization model proposed in this invention fully considers the dynamic characteristics of the user side and the equipment side, and realizes global coordination while ensuring the independent optimization of each subsystem.

[0206] In summary, compared with the existing technology, the present invention is innovative and advantageous in load analysis, classification and optimal scheduling. By introducing cluster analysis methods and hierarchical optimization models, the problems of inaccurate load classification, insufficient mathematical models and single optimization structure in traditional methods are effectively solved, which provides technical support for the intelligent management of high-energy-consuming industrial microgrids and improves energy utilization efficiency and operation economy.

[0207] Example 3, reference Figure 2 , is an embodiment of the present invention, and provides a high-energy-consuming industrial microgrid hierarchical scheduling optimization system based on load classification, including a load analysis module, a load classification module, and a hierarchical coordination optimization module.

[0208] The load analysis module is used to perform cluster analysis on the loads of high-energy-consuming industrial microgrids, identify load groups with similar characteristics and extract key features; the load classification module is used to classify the loads according to the load clustering results; the hierarchical coordination optimization module is used to establish a mathematical model of high-energy-consuming industrial microgrids, and ensure the coordinated optimization and operation efficiency of each subsystem through hierarchical scheduling optimization.

[0209] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0210] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0211] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0212] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and are not limited. Although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention, which should be included in the scope of the claims of the present invention.

[0213] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A hierarchical scheduling optimization method for high-energy-consuming industrial microgrids based on load classification, characterized in that: include: Cluster analysis is performed on high-energy-consuming industrial microgrid loads to identify load groups with similar characteristics and extract key features; Classify the loads according to the load clustering results; A mathematical model of high-energy-consuming industrial microgrid is established, and the coordinated optimization and operation efficiency of each subsystem are ensured through hierarchical scheduling optimization.

2. The high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification according to claim 1 is characterized in that: The cluster analysis of high-energy-consuming industrial microgrid loads includes collecting historical load data, processing outliers and missing values ​​in the data, extracting load features, clustering load data using a density-based DBSCAN algorithm, dividing the load into several clusters, and excluding abnormal load points. The distance between all data points is calculated based on the Euclidean distance, which is expressed as: Among them, p i is the power value of the i-th load data point, p j is the power value of the jth load data point, For p i The value in the nth dimension, For p j The value in the nth dimension; Calculate all points within the threshold ∈ distance of the load data point p to form the ∈ neighborhood of the load data point p, expressed as: N ∈ (p)={q∈D}|dist(p,q)≤∈ Among them, N ∈ (p) is the set of all points whose distance from the load data point p does not exceed the threshold ∈, D is the load data point data set, q is the load data point different from p, dist(p,q) is the Euclidean distance between the load data point p and the load data point q, ∈ is the distance threshold for judging the neighborhood; Judging whether it is a core point according to the number of neighborhood points is expressed as: Among them, minPts is the minimum number of neighboring points for judging the core point. If the neighborhood N of the load data point p is ∈ The number of points in (p) |N ∈ (p)| is greater than or equal to the minimum number of neighborhood points minPts, then the load data point p is a core point. If the neighborhood N of the load data point p ∈ The number of points in (p) |N ∈ (p)| is less than the minimum number of neighborhood points minPts, then the load data point p is not a core point.

3. The high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification according to claim 2 is characterized in that: The cluster analysis of high-energy-consuming industrial microgrid loads also includes evaluating the clustering effect through the Davidson-Boulding index and the silhouette coefficient, which is expressed as: Where DB is the Davidson-Boting index, k is the number of clusters, σ i is the average distance of cluster i, σ j is the average distance of cluster j, d ij is the distance between clusters i and j.

4. The method for hierarchical scheduling optimization of high-energy-consuming industrial microgrids based on load classification according to claim 3 is characterized in that: The loads are classified into adjustable loads and unadjustable loads according to the load clustering result.

5. The method for hierarchical scheduling optimization of high-energy-consuming industrial microgrids based on load classification according to claim 4 is characterized in that: The adjustable loads include time-limited transferable loads, continuous translational loads and all adjustable loads; If the load is a time-limited transferable load, the operating period can be adjusted within the predetermined time range without affecting the production task; If the load is a continuous and movable load, the start-up time can be flexibly adjusted according to the sequence requirements of the production line, and the normal operation of the production line can still be ensured after adjustment.

6. The method for hierarchical scheduling optimization of high-energy-consuming industrial microgrids based on load classification according to claim 5 is characterized in that: The mathematical model for establishing a high-energy-consuming industrial microgrid includes establishing a power model of photovoltaic and wind turbine generator sets, which is expressed as: in, is the output power of photovoltaic and wind turbine at time t, is the maximum output power of photovoltaic and wind turbine at time t; The gas turbine power model is established and expressed as: in, are the electrical and thermal power output of the gas turbine at time t, is the natural gas power consumed by the gas turbine at time t, are the energy conversion efficiency of gas turbine output electricity and thermal power, are the upper and lower limits of the electrical and thermal power output of the gas turbine at time t respectively; The gas boiler power model is established and expressed as: in, is the thermal power output of the gas boiler at time t, is the natural gas power consumed by the gas boiler at time t, The energy conversion efficiency of the gas boiler output thermal power, They are the upper and lower limits of the thermal power output of the gas boiler at time t; The power model of electric cooling / heating equipment is established and expressed as: in, is the output of the electric cooling / heating equipment at time t, η c / e is the energy conversion efficiency, is the power consumption of the electric cooling / heating equipment at time t, They are the upper and lower limits of the output of the electric cooling / heating equipment at time t; The power model of absorption chiller is established and expressed as: in, are the heat power input and cooling power output of the absorption refrigerator at time t, respectively, AC is the conversion efficiency of the absorption chiller, They are the upper and lower limits of the cooling power output of the absorption chiller at time t respectively; The design constraints of energy storage equipment include energy storage charging and discharging constraints, which are expressed as: in, is the capacity of the energy storage device at time t, is the capacity of the energy storage at time t+1, η ch is the energy storage charging efficiency, η dis is the energy storage discharge efficiency, is the charging power of the energy storage device at time t, is the discharge power of the energy storage device at time t, and Δt is the change time; If the charging and discharging power of the energy storage device is equal within a time period, the energy storage charging and discharging constraint condition is expressed as: Where T is the total duration of the time period; Calculate the capacity constraint and power constraint of the energy storage device at each moment, expressed as: in, are the upper and lower limits of the energy storage device capacity at time t, are the upper and lower limits of the charging power of the energy storage device at time t, are the upper and lower limits of the energy storage device discharge power at time t, are the 0-1 state variables of the energy storage device charging and discharging at time t, When the value is 0, it means the device stops working. When the value is 1, it means the device is in working state.

7. The method for hierarchical scheduling optimization of high-energy-consuming industrial microgrids based on load classification according to claim 6 is characterized in that: The hierarchical dispatch optimization includes establishing a hierarchical coordination optimization model for high-energy-consuming industrial microgrids and establishing a total revenue function for high-energy-consuming microgrids, which is expressed as: Among them, E IMG is the total revenue function of the high energy consumption microgrid, It is the energy purchasing function of high energy consumption microgrid from the upper power grid. is the energy selling function of the high energy consumption microgrid, For the low-carbon operation function of high-energy-consuming microgrids, It is the function for maximizing the utilization of photovoltaic and wind power in high-energy-consuming microgrids. Maintenance rate for high energy consumption microgrid equipment; Calculate the inequality constraints of high-energy-consuming industrial microgrids, expressed as: in, β e,h,c They are the upper and lower limits of the microgrid energy sales price, are the average constraints for selling and buying energy prices, are the constraints on selling and buying energy prices at time t, respectively, i Output power to the i-th device in the microgrid, P i are the upper and lower limits of equipment output, E i ,H i ,C i are the constraints of electricity, heat, and cooling energy provided by the microgrid to users, respectively; Calculate the power balance equation constraint of high-energy-consuming industrial microgrid, expressed as: Among them, E e 、E h , P c are the maximum power of electricity, heat and cold energy provided by high-energy-consuming industrial microgrids, P e,i , P h,i , P c,i are the electrical, thermal and cooling outputs of the ith device respectively; Establish the total revenue function of high-energy-consuming industrial users, conduct internal optimization of high-energy-consuming industrial parks in the jurisdiction of user aggregators, be responsible for connecting load users, and integrate dispersed load resources. The total revenue function of high-energy-consuming industrial users is expressed as: Among them, F USER is the total revenue function of high energy-consuming industrial users, is the user utility function, To participate in the auxiliary peak load subsidy, is the energy consumption function, is the energy storage capacity leasing function, Cost of CO2 emissions; Calculate the inequality constraints for high energy-consuming industrial loads, expressed as: in, They are the upper and lower limits of adjustable electric load for high energy-consuming industrial loads; The optimization layer of high-energy-consuming industrial users is solved by the CPLEX solver, and the optimal power consumption of users is taken as input. The high-energy-consuming industrial microgrid scheduling layer uses a genetic algorithm to achieve the optimal scheduling of equipment within the jurisdiction.

8. A system using the high-energy-consuming industrial microgrid hierarchical scheduling optimization method based on load classification as described in any one of claims 1 to 7, characterized in that: Including load analysis module, load classification module, hierarchical coordination optimization module; The load analysis module is used to perform cluster analysis on high-energy-consuming industrial microgrid loads, identify load groups with similar characteristics and extract key features; The load classification module is used to classify the load according to the load clustering result; The hierarchical coordination optimization module is used to establish a mathematical model of a high-energy-consuming industrial microgrid, and ensure the coordination optimization and operation efficiency of each subsystem through hierarchical scheduling optimization.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the load-classified high-energy-consuming industrial microgrid hierarchical scheduling optimization method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the load-classified high-energy-consuming industrial microgrid hierarchical scheduling optimization method described in any one of claims 1 to 7 are implemented.

Images